issue_owner_repo listlengths 2 2 | issue_body stringlengths 0 261k ⌀ | issue_title stringlengths 1 925 | issue_comments_url stringlengths 56 81 | issue_comments_count int64 0 2.5k | issue_created_at stringlengths 20 20 | issue_updated_at stringlengths 20 20 | issue_html_url stringlengths 37 62 | issue_github_id int64 387k 2.46B | issue_number int64 1 127k |
|---|---|---|---|---|---|---|---|---|---|
[
"MonetDB",
"MonetDB"
] | Date: 2015-04-21 16:39:05 +0200
From: John <<caroptions>>
To: SQL devs <<bugs-sql>>
Version: 11.21.19 (Jul2015-SP4)
CC: martin.van.dinther
Last updated: 2019-07-10 14:07:00 +0200
## Comment 20826
Date: 2015-04-21 16:39:05 +0200
From: John <<caroptions>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_3) AppleWebKit/600.5.17 (KHTML, like Gecko) Version/8.0.5 Safari/600.5.17
Build Identifier:
> I am experiencing very strange behavior:
>
> CREATE TRIGGER marketdata.calc_sdate BEFORE INSERT ON marketdata.quotes
> FOR EACH ROW
> BEGIN ATOMIC
> -- select * from marketdata.quotes limit 1;
> END;
>
>
> ERROR: syntax error, unexpected '.', expecting WHILE in: "create
> trigger marketdata.calc_sdate before insert on
> marketdata."
>
> When I am creating trigger on table without schema it works fine.
> CREATE TRIGGER marketdata.calc_sdate BEFORE INSERT ON t1 ...
>
> Assuming t1 created without schema.
>
> All combination of " around schema and table, schema only, table only
> didn't help.
>
> Am I doing something wrong or this is the bug?
Indeed a bug, we should allow for a fully qualified name (ie including
schema name).
Niels
Reproducible: Always
### Steps to Reproduce:
> I am experiencing very strange behavior:
>
> CREATE TRIGGER marketdata.calc_sdate BEFORE INSERT ON marketdata.quotes
> FOR EACH ROW
> BEGIN ATOMIC
> -- select * from marketdata.quotes limit 1;
> END;
>
>
> ERROR: syntax error, unexpected '.', expecting WHILE in: "create
> trigger marketdata.calc_sdate before insert on
> marketdata."
>
> When I am creating trigger on table without schema it works fine.
> CREATE TRIGGER marketdata.calc_sdate BEFORE INSERT ON t1 ...
>
> Assuming t1 created without schema.
>
> All combination of " around schema and table, schema only, table only
> didn't help.
>
> Am I doing something wrong or this is the bug?
Indeed a bug, we should allow for a fully qualified name (ie including
schema name).
Niels
## Comment 21596
Date: 2015-11-25 22:30:36 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [995872b4375d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=995872b4375d) made by Martin Kersten <mk@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=995872b4375d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=995872b4375d)
Changeset description:
Add test for bug #3710
## Comment 25199
Date: 2017-04-06 13:21:31 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [faac32625f98](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=faac32625f98) made by Martin van Dinther <martin.van.dinther@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=faac32625f98](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=faac32625f98)
Changeset description:
Added missing desired output files for test: schema-trigger.Bug-3710
## Comment 27120
Date: 2019-07-10 13:58:47 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [769e84ad3c38](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=769e84ad3c38) made by Martin van Dinther <martin.van.dinther@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=769e84ad3c38](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=769e84ad3c38)
Changeset description:
Extending trigger test with body action and some inserts to trigger the body action.
Bug #3710 is resolved.
| Strange syntax error when create trigger | https://api.github.com/repos/MonetDB/MonetDB/issues/3710/comments | 0 | 2020-11-30T13:01:20Z | 2024-06-27T12:07:06Z | https://github.com/MonetDB/MonetDB/issues/3710 | 753,459,671 | 3,710 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-04-19 15:33:23 +0200
From: sorear
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
CC: @njnes
Last updated: 2015-05-07 12:37:25 +0200
## Comment 20817
Date: 2015-04-19 15:33:23 +0200
From: sorear
This is a reliably reproducable error condition:
sql>drop table s10; create table s10 (x int); start transaction; insert into s10 values (8); alter table s10 set read only; select * from s10; rollback;
operation successful (3.3s)
operation successful (3.3s)
auto commit mode: on
1 affected row (3.3s)
operation successful (3.3s)
GDK reported error.
BATproject: does not match always
auto commit mode: on
sql>
It seems that in this case (readonly=1, nonempty ibat) sql.tid returns one row while sql.bind(x) returns zero, and algebra.leftfetchjoin throws the error above. Works fine if the table creation is moved into the transaction.
## Comment 20824
Date: 2015-04-20 21:14:12 +0200
From: @njnes
disallow changing to readonly on tables without standing inserts.
## Comment 20825
Date: 2015-04-20 21:21:06 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [df0d1a4be0ed](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=df0d1a4be0ed) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=df0d1a4be0ed](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=df0d1a4be0ed)
Changeset description:
added test for bug #3709
| "BATproject: does not match always" on abusive use of ALTER TABLE SET READ ONLY | https://api.github.com/repos/MonetDB/MonetDB/issues/3709/comments | 0 | 2020-11-30T13:01:17Z | 2024-06-27T12:07:05Z | https://github.com/MonetDB/MonetDB/issues/3709 | 753,459,636 | 3,709 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-04-19 15:10:20 +0200
From: sorear
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
CC: @njnes
Last updated: 2015-08-28 13:42:11 +0200
## Comment 20816
Date: 2015-04-19 15:10:20 +0200
From: sorear
Just noticed this bit of odd behavior:
sql>create schema s1;
operation successful (26.216ms)
sql>create schema s2;
operation successful (2.487ms)
sql>set current_schema='s1';
auto commit mode: on
sql>create table s3 (x int);
operation successful (5.756ms)
sql>create view s4 as select * from s3;
operation successful (2.070ms)
sql>select * from s1.s4;
+---+
| x |
+===+
+---+
0 tuples (3.550ms)
sql>set current_schema='s2';
auto commit mode: on
sql>select * from s1.s4;
SELECT: no such table 's3'
Apparently the view expansion is not taking into account the view's own schema and using the connection current schema; this can also be shown by creating another version of the table and using PLAN:
sql>create schema s1;
operation successful (2.167ms)
sql>create schema s2;
operation successful (1.642ms)
sql>create table s1.s3 (x int);
operation successful (2.331ms)
sql>create table s2.s3 (x int);
operation successful (2.316ms)
sql>set current_schema='s1';
auto commit mode: on
sql>create view s4 as select * from s3;
operation successful (2.749ms)
sql>plan select * from s1.s4;
+--------------------------------+
| rel |
+================================+
| project ( |
| | table(s1.s3) [ s3.x ] COUNT |
| ) [ s3.x as s4.x ] |
+--------------------------------+
3 tuples (0.398ms)
sql>set current_schema = 's2';
auto commit mode: on
sql>plan select * from s1.s4;
+--------------------------------+
| rel |
+================================+
| project ( |
| | table(s2.s3) [ s3.x ] COUNT |
| ) [ s3.x as s4.x ] |
+--------------------------------+
3 tuples (0.264ms)
I'd expect the plan to not depend on the current_schema. Note that this is *not* an issue for us in any way; just odd behavior noticed while testing something else.
Incidentally while trying to find duplicates of this I got a bizarre error from bugzilla (here reproduced with curl -v -o- https://www.monetdb.org/bugzilla/buglist.cgi?content=view%20schema)
> GET /bugzilla/buglist.cgi?content=view%20schema HTTP/1.1
> User-Agent: curl/7.19.7 (x86_64-redhat-linux-gnu) libcurl/7.19.7 NSS/3.13.1.0 zlib/1.2.3 libidn/1.18 libssh2/1.2.2
> Host: www.monetdb.org
> Accept: */*
>
< HTTP/1.1 500 Internal Server Error
< Date: Sun, 19 Apr 2015 13:08:23 GMT
< Server: Apache/2.4.10 (Fedora) OpenSSL/1.0.1e-fips mod_fcgid/2.3.9 PHP/5.5.23 mod_perl/2.0.9-dev Perl/v5.18.4
< Strict-Transport-Security: max-age=15768000
< Connection: close
< Transfer-Encoding: chunked
< Content-Type: text/html; charset=UTF-8
<
<h1>Software error:</h1>
<pre>Invalid local time for date in time zone: Europe/Amsterdam
</pre>
<p>
For help, please send mail to the webmaster (<a href="mailto:root@localhost">root@localhost</a>), giving this error message
and the time and date of the error.
</p>
## Comment 20873
Date: 2015-05-20 21:02:26 +0200
From: @njnes
Fixed. We now check views schema
## Comment 20874
Date: 2015-05-20 21:03:46 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [87af5a5903a8](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=87af5a5903a8) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=87af5a5903a8](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=87af5a5903a8)
Changeset description:
fixed bug in handling view's and schema's (fixed bug #3708)
## Comment 21205
Date: 2015-08-28 13:42:11 +0200
From: @sjoerdmullender
Jul2015 has been released.
| wrong scoping for cross-schema view references | https://api.github.com/repos/MonetDB/MonetDB/issues/3708/comments | 0 | 2020-11-30T13:01:15Z | 2024-06-27T12:07:04Z | https://github.com/MonetDB/MonetDB/issues/3708 | 753,459,606 | 3,708 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-04-19 14:23:31 +0200
From: sorear
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.19.7 (Oct2014-SP1)
CC: @mlkersten, @drstmane
Last updated: 2015-06-22 22:32:25 +0200
## Comment 20815
Date: 2015-04-19 14:23:31 +0200
From: sorear
When I run "select * from var();" in a freshly started database, I get an immediate assertion failure:
2015-04-19 05:05:51 ERR gt[6067]: mserver5: ../../../../MonetDB-11.19.7/sql/backends/monet5/sql.c:1378: sql_variables: Assertion `v->vtype == 5' failed.
v->vtype is 4 (TYPE_bat) in this case; I guess this is because the sql_variables MAL op is declared to return a bat in the sql.mal, but the argument checking wants a pointer to int.
+------------------------------------------------------------------+
| mal |
+==================================================================+
| function user.s2_1{autoCommit=true}():void; |
| X_2:bat[:oid,:str] := sql.sql_variables(); |
| X_3 := sql.resultSet(1,1,X_2); |
| sql.rsColumn(X_3,".","name","varchar",1024,0,X_2); |
| X_11 := io.stdout(); |
| sql.exportResult(X_11,X_3); |
| end s2_1; |
| querylog.define("explain select * from var();","default_pipe") |
+------------------------------------------------------------------+
MonetDB 5 server v11.19.7 "Oct2014-SP1" (64-bit, 64-bit oids)
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2014 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 1.8GiB available memory, 2 available cpu cores
Libraries:
libpcre: 7.8 2008-09-05 (compiled with 7.8)
openssl: OpenSSL 1.0.1e 11 Feb 2013 (compiled with OpenSSL 1.0.1e-fips 11 Feb 2013)
libxml2: 2.7.6 (compiled with 2.7.6)
Compiled by: REDACTED (x86_64-unknown-linux-gnu)
Compilation: gcc -g
Linking : /usr/bin/ld -m elf_x86_64
## Comment 20877
Date: 2015-05-20 22:41:09 +0200
From: @mlkersten
The issue can not be reproduced on the default branch.
I consider it closed until further notice.
## Comment 20878
Date: 2015-05-20 23:05:34 +0200
From: @drstmane
For the records:
The problem is reproducible as reported with Oct2014 and Oct2014-SP1,
but appears to be fixed as of Oct2014-SP2.
Please upgrade to the latest Oct2014-SP3 (or the upcoming Oct2014-SP4 once it will be released).
| var() possibly not working in debug builds | https://api.github.com/repos/MonetDB/MonetDB/issues/3707/comments | 0 | 2020-11-30T13:01:12Z | 2024-06-27T12:07:02Z | https://github.com/MonetDB/MonetDB/issues/3707 | 753,459,570 | 3,707 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-04-16 16:08:24 +0200
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
CC: @njnes
Last updated: 2015-05-07 12:37:20 +0200
## Comment 20803
Date: 2015-04-16 16:08:24 +0200
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Fedora; Linux x86_64; rv:36.0) Gecko/20100101 Firefox/36.0
Build Identifier:
SQL Query:
SELECT (SELECT count(*) FROM t1 AS x WHERE x.b<t1.b) FROM t1 WHERE (a>b-2 AND a<b+2) OR c>d ORDER BY 1;
causes assertion failure
>mserver5: ../../MonetDB-11.19.9/gdk/gdk_bat.c:2881: BATassertHeadProps: Assertion `!b->H->sorted || cmp <= 0' failed.
Also happens on MonetDB 5 server v11.20.0:
>mserver5: ../../dev/gdk/gdk_bat.c:2847: BATassertHeadProps: Assertion `!b->H->sorted || cmp <= 0' failed.
Reproducible: Always
### Steps to Reproduce:
1. Start mserver5 (MonetDB 5 server v11.19.9 "Oct2014-SP2")
2. Start mclient
3. execute SQL commands:
CREATE TABLE t1(a INTEGER, b INTEGER, c INTEGER, d INTEGER, e INTEGER);
INSERT INTO t1(e,c,b,d,a) VALUES(103,102,100,101,104);
INSERT INTO t1(a,c,d,e,b) VALUES(107,106,108,109,105);
INSERT INTO t1(e,d,b,a,c) VALUES(110,114,112,111,113);
INSERT INTO t1(d,c,e,a,b) VALUES(116,119,117,115,118);
SELECT (SELECT count(*) FROM t1 AS x WHERE x.b<t1.b) FROM t1 WHERE (a>b-2 AND a<b+2) OR c>d ORDER BY 1;
### Actual Results:
builtin opt gdk_dbpath = /export/scratch1/dinther/Downloads/INSTALL/var/monetdb5/dbfarm/demo
builtin opt gdk_debug = 0
builtin opt gdk_vmtrim = no
builtin opt monet_prompt = >
builtin opt monet_daemon = no
builtin opt mapi_port = 50000
builtin opt mapi_open = false
builtin opt mapi_autosense = false
builtin opt sql_optimizer = default_pipe
builtin opt sql_debug = 0
cmdline opt gdk_debug = 10
MonetDB 5 server v11.19.9 "Oct2014-SP2"
Serving database 'demo', using 8 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs dynamically linked
Found 15.590 GiB available main-memory.
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2015 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Listening for connection requests on mapi:monetdb://127.0.0.1:50000/
MonetDB/GIS module loaded
MonetDB/SQL module loaded
>mserver5: ../../MonetDB-11.19.9/gdk/gdk_bat.c:2881: BATassertHeadProps: Assertion `!b->H->sorted || cmp <= 0' failed.
bash-4.2$ mclient
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.19.9 (Oct2014-SP2), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>CREATE TABLE t1(a INTEGER, b INTEGER, c INTEGER, d INTEGER, e INTEGER);
operation successful (6.762ms)
sql>INSERT INTO t1(e,c,b,d,a) VALUES(103,102,100,101,104);
1 affected row (0.581ms)
sql>INSERT INTO t1(a,c,d,e,b) VALUES(107,106,108,109,105);
1 affected row (0.460ms)
sql>INSERT INTO t1(e,d,b,a,c) VALUES(110,114,112,111,113);
1 affected row (0.499ms)
sql>INSERT INTO t1(d,c,e,a,b) VALUES(116,119,117,115,118);
1 affected row (0.493ms)
sql>SELECT (SELECT count(*) FROM t1 AS x WHERE x.b<t1.b) FROM t1 WHERE (a>b-2 AND a<b+2) OR c>d ORDER BY 1;
sql>
bash-4.2$
### Expected Results:
no assertion failure
## Comment 20821
Date: 2015-04-19 22:09:30 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [b3a1a04bea24](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=b3a1a04bea24) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=b3a1a04bea24](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=b3a1a04bea24)
Changeset description:
added test for bug #3706
## Comment 20822
Date: 2015-04-19 22:10:02 +0200
From: @njnes
fixed after recent changes (probably same fix as 3705)
## Comment 20837
Date: 2015-04-23 16:34:39 +0200
From: Martin van Dinther <<martin.van.dinther>>
Unfortunately it is NOT fixed yet. It still happens.
I re-tested with MonetDB 5 server v11.19.11 "Oct2014-SP3" built of 23 april
builtin opt gdk_dbpath = /export/scratch1/dinther/Downloads/INSTALL/var/monetdb5/dbfarm/demo
builtin opt gdk_debug = 0
builtin opt gdk_vmtrim = no
builtin opt monet_prompt = >
builtin opt monet_daemon = no
builtin opt mapi_port = 50000
builtin opt mapi_open = false
builtin opt mapi_autosense = false
builtin opt sql_optimizer = default_pipe
builtin opt sql_debug = 0
cmdline opt gdk_debug = 10
MonetDB 5 server v11.19.11 "Oct2014-SP3"
Serving database 'demo', using 8 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs dynamically linked
Found 15.590 GiB available main-memory.
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2015 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Listening for connection requests on mapi:monetdb://127.0.0.1:50000/
MonetDB/GIS module loaded
MonetDB/SQL module loaded
> SQL catalog created, loading sql scripts once
loading sql script: 09_like.sql
loading sql script: 10_math.sql
loading sql script: 11_times.sql
loading sql script: 12_url.sql
loading sql script: 13_date.sql
loading sql script: 14_inet.sql
loading sql script: 15_querylog.sql
loading sql script: 16_tracelog.sql
loading sql script: 19_cluster.sql
loading sql script: 20_vacuum.sql
loading sql script: 21_dependency_functions.sql
loading sql script: 22_clients.sql
loading sql script: 23_skyserver.sql
loading sql script: 24_zorder.sql
loading sql script: 25_debug.sql
loading sql script: 26_sysmon.sql
loading sql script: 39_analytics.sql
loading sql script: 40_geom.sql
loading sql script: 40_json.sql
loading sql script: 41_jsonstore.sql
loading sql script: 45_uuid.sql
loading sql script: 46_gsl.sql
loading sql script: 75_storagemodel.sql
loading sql script: 80_statistics.sql
loading sql script: 80_udf.sql
loading sql script: 85_bam.sql
loading sql script: 90_generator.sql
loading sql script: 99_system.sql
mserver5: ../../MonetDB-11.19.11/gdk/gdk_bat.c:2839: BATassertHeadProps: Assertion `!b->H->sorted || cmp <= 0' failed.
bash-4.2$
Idem for: MonetDB 5 server v11.20.0:
mserver5: ../../dev/gdk/gdk_bat.c:2863: BATassertHeadProps: Assertion `!b->H->sorted || cmp <= 0' failed.
## Comment 20841
Date: 2015-04-24 11:26:29 +0200
From: @njnes
Runs fine according to the current stable branch nightly test output
http://monetdb.cwi.nl/testweb/web/showtestoutput.php?serial=55351:bc4923c9aad2&target=Int-Fedora-x86_64-assert&module=sql&test=sql%2Ftest%2FBugTracker-2015&which=out
## Comment 20846
Date: 2015-04-28 14:07:18 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [59b3760cadc0](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=59b3760cadc0) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=59b3760cadc0](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=59b3760cadc0)
Changeset description:
Don't copy properties if there are nils (probably due to candidate list).
This fixes bug #3706.
## Comment 20847
Date: 2015-04-28 14:07:58 +0200
From: @sjoerdmullender
It turned out that there really still was a bug. This should now be fixed.
| Assertion failure in gdk_bat.c: BATassertHeadProps: Assertion `!b->H->sorted || cmp <= 0' failed. | https://api.github.com/repos/MonetDB/MonetDB/issues/3706/comments | 0 | 2020-11-30T13:01:10Z | 2024-06-27T12:07:01Z | https://github.com/MonetDB/MonetDB/issues/3706 | 753,459,527 | 3,706 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-04-16 14:26:39 +0200
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
CC: @njnes
Last updated: 2015-06-22 22:32:24 +0200
## Comment 20802
Date: 2015-04-16 14:26:39 +0200
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Fedora; Linux x86_64; rv:36.0) Gecko/20100101 Firefox/36.0
Build Identifier:
SQL Query:
SELECT CASE WHEN c>(SELECT avg(c) FROM t1) THEN a*2 ELSE b*10 END FROM t1;
causes assertion failure
Reproducible: Always
### Steps to Reproduce:
1. Start mserver5 (MonetDB 5 server v11.19.9 "Oct2014-SP2")
2. Start mclient
3. execute SQL commands:
CREATE TABLE t1(a INTEGER, b INTEGER, c INTEGER, d INTEGER, e INTEGER);
INSERT INTO t1(e,c,b,d,a) VALUES(103,102,100,101,104);
SELECT CASE WHEN c>(SELECT avg(c) FROM t1) THEN a*2 ELSE b*10 END FROM t1;
### Actual Results:
builtin opt gdk_dbpath = /export/scratch1/dinther/Downloads/INSTALL/var/monetdb5/dbfarm/demo
builtin opt gdk_debug = 0
builtin opt gdk_vmtrim = no
builtin opt monet_prompt = >
builtin opt monet_daemon = no
builtin opt mapi_port = 50000
builtin opt mapi_open = false
builtin opt mapi_autosense = false
builtin opt sql_optimizer = default_pipe
builtin opt sql_debug = 0
cmdline opt gdk_debug = 10
MonetDB 5 server v11.19.9 "Oct2014-SP2"
Serving database 'demo', using 8 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs dynamically linked
Found 15.590 GiB available main-memory.
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2015 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Listening for connection requests on mapi:monetdb://127.0.0.1:50000/
MonetDB/GIS module loaded
MonetDB/SQL module loaded
>could not find t1.a
t1.c
L1.L1
mserver5: ../../../../MonetDB-11.19.9/sql/backends/monet5/rel_bin.c:2274: rel2bin_project: Assertion `0' failed.
bash-4.2$
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.19.9 (Oct2014-SP2), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>CREATE TABLE t1(a INTEGER, b INTEGER, c INTEGER, d INTEGER, e INTEGER);
operation successful (7.575ms)
sql>INSERT INTO t1(e,c,b,d,a) VALUES(103,102,100,101,104);
1 affected row (1.117ms)
sql>SELECT avg(c) FROM t1;
+--------------------------+
| L1 |
+==========================+
| 102 |
+--------------------------+
1 tuple (1.247ms)
sql>SELECT CASE WHEN c>(SELECT avg(c) FROM t1) THEN a*2 ELSE b*10 END FROM t1;
sql>
bash-4.2$
### Expected Results:
no assertion failure
Also happens on MonetDB 5 server v11.20.0:
mserver5: ../../../../dev/sql/backends/monet5/rel_bin.c:2234: rel2bin_project: Assertion `0' failed.
## Comment 20814
Date: 2015-04-19 13:27:22 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [33ac46bb0409](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=33ac46bb0409) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=33ac46bb0409](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=33ac46bb0409)
Changeset description:
fixed bug #3705, ie handle subquery projections better
## Comment 20818
Date: 2015-04-19 21:29:05 +0200
From: @njnes
Fixed (properly pass all projection expressions) and added test
## Comment 20896
Date: 2015-05-28 16:44:53 +0200
From: Martin van Dinther <<martin.van.dinther>>
Reopened because when I extend the SELECT query with "abs(b-c)" as in:
SELECT abs(b-c), CASE WHEN c>(SELECT avg(c) FROM t1) THEN a*2 ELSE b*10 END FROM t1;
or
SELECT CASE WHEN c>(SELECT avg(c) FROM t1) THEN a*2 ELSE b*10 END, abs(b-c) FROM t1;
I again get the assertion failure:
mserver5: ../../../../MonetDB-11.19.11/sql/backends/monet5/rel_bin.c:2279: rel2bin_project: Assertion `0' failed.
This happens on MonetDB 5 server v11.19.11 "Oct2014-SP3" and on MonetDB 5 server v11.20.0. (built on 28 may 2015).
On MonetDB 5 server v11.20.0 I get:
mserver5: ../../../../dev/sql/backends/monet5/rel_bin.c:2352: rel2bin_project: Assertion `0' failed.
I have extended BugTracker-2015/Tests/case_aggr.Bug-3705.sql and stable.out with the new SELECT statements and desired output.
## Comment 20897
Date: 2015-05-28 16:47:45 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [e946d6eed01a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e946d6eed01a) made by Martin van Dinther <martin.van.dinther@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=e946d6eed01a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=e946d6eed01a)
Changeset description:
Variants of SELECT query still produce same assertion failure as reported in bug #3705
Extended test case with those SELECTs and reopened bug #3705
| Assertion failure in rel_bin.c:2274: rel2bin_project: Assertion `0' failed. | https://api.github.com/repos/MonetDB/MonetDB/issues/3705/comments | 0 | 2020-11-30T13:01:07Z | 2024-06-27T12:07:00Z | https://github.com/MonetDB/MonetDB/issues/3705 | 753,459,494 | 3,705 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-04-16 14:01:08 +0200
From: Richard Hughes <<richard.monetdb>>
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
CC: @njnes
Last updated: 2015-05-07 12:37:26 +0200
## Comment 20801
Date: 2015-04-16 14:01:08 +0200
From: Richard Hughes <<richard.monetdb>>
Build is Oct2014 [9c80de069c44](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9c80de069c44) (head of that branch as of creation of this bug report)
create table foo (z int);
select aaa from (select 0,0,0,0,0,foo.z from foo)as t(aaa,b,c,d,e,f);
Expected:
null
Actual:
SELECT: identifier 'aaa' unknown
Some interesting observations:
- Any fewer than 6 columns in the subquery and it'll work
- Any more than 6 and it will still not work
- Remove the qualified identifier 'foo.z' (i.e. just make it 'z') and it'll work
- None of the subquery columns work (e.g. "select b from (select 0,0,0,0,0,foo.z from foo)as t(aaa,b,c,d,e,f);" doesn't work either)
## Comment 20811
Date: 2015-04-18 22:36:51 +0200
From: @njnes
added missing (name) hash after adding the aliases.
## Comment 20812
Date: 2015-04-18 22:39:43 +0200
From: @njnes
added test BugTracker-2015/alias.Bug-3704.sql
## Comment 20813
Date: 2015-04-18 22:59:33 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [8f6b8d6f6366](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=8f6b8d6f6366) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=8f6b8d6f6366](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=8f6b8d6f6366)
Changeset description:
fixed bug #3704 (ie cleanup hash table after renaming expressions)
| Unknown identifier from subquery | https://api.github.com/repos/MonetDB/MonetDB/issues/3704/comments | 0 | 2020-11-30T13:01:04Z | 2024-06-27T12:06:59Z | https://github.com/MonetDB/MonetDB/issues/3704 | 753,459,465 | 3,704 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-04-16 11:28:17 +0200
From: @yzchang
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
CC: @njnes
Last updated: 2015-05-07 12:38:02 +0200
## Comment 20799
Date: 2015-04-16 11:28:17 +0200
From: @yzchang
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.10; rv:37.0) Gecko/20100101 Firefox/37.0
Build Identifier:
INSERT INTO a MERGE TABLE crashes mserver5. GDB info:
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7fffecd89700 (LWP 23059)]
0x00007fffef9fd622 in timestamp_delta (d=0x0, ts=15)
at /ufs/zhang/monet/Oct2014/sql/storage/bat/bat_storage.c:32
32 while (d->next && d->wtime > ts)
Reproducible: Always
### Steps to Reproduce:
A minimal query set to trigger the crash:
create table t1 (i int);
create table t2 (i int);
create merge table m (i int);
alter table m add table t1;
alter table m add table t2;
insert into m values (1);
### Actual Results:
crash
### Expected Results:
An error message is expected.
## Comment 20805
Date: 2015-04-16 17:00:52 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [810b430bcf72](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=810b430bcf72) made by Jennie Zhang <y.zhang@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=810b430bcf72](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=810b430bcf72)
Changeset description:
Added test for bug #3703
## Comment 20808
Date: 2015-04-17 13:25:56 +0200
From: @njnes
added checks to disallow inserting into merge table
| INSERT INTO a MERGE TABLE crashes mserver5 | https://api.github.com/repos/MonetDB/MonetDB/issues/3703/comments | 0 | 2020-11-30T13:01:02Z | 2024-06-27T12:06:58Z | https://github.com/MonetDB/MonetDB/issues/3703 | 753,459,428 | 3,703 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-04-16 08:14:19 +0200
From: @swingbit
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
CC: @njnes
Last updated: 2015-08-28 13:41:50 +0200
## Comment 20797
Date: 2015-04-16 08:14:19 +0200
From: @swingbit
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2272.76 Safari/537.36
Build Identifier:
I created a filter function:
CREATE filter function strmaxdist(s1 string, s2 string, k int)
external name str.maxdist;
to be used as:
SELECT *
FROM a,b
WHERE a.s strmaxdist(2) b.s;
It works fine if I create and use the filter function in the "sys" schema.
If I create and use it when I am on a different schema, then I get:
Subquery result missing
Reproducible: Always
$ mserver5 --version
MonetDB 5 server v11.19.10 (64-bit, 64-bit oids)
This is an unreleased version
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2015 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 15.6GiB available memory, 8 available cpu cores
Libraries:
libpcre: 8.35 2014-04-04 (compiled with 8.35)
openssl: OpenSSL 1.0.1k 8 Jan 2015 (compiled with OpenSSL 1.0.1k-fips 8 Jan 2015)
libxml2: 2.9.1 (compiled with 2.9.1)
Compiled by: roberto@photon.spinque.com (x86_64-unknown-linux-gnu)
Compilation: gcc -g -Werror -Wall -Wextra -W -Werror-implicit-function-declaration -Wpointer-arith -Wdeclaration-after-statement -Wundef -Wformat=2 -Wno-format-nonliteral -Winit-self -Winvalid-pch -Wmissing-declarations -Wmissing-format-attribute -Wmissing-prototypes -Wold-style-definition -Wpacked -Wunknown-pragmas -Wvariadic-macros -fstack-protector-all -Wstack-protector -Wpacked-bitfield-compat -Wsync-nand -Wjump-misses-init -Wmissing-include-dirs -Wlogical-op -Wunreachable-code
Linking : /usr/bin/ld -m elf_x86_64
## Comment 20798
Date: 2015-04-16 11:10:50 +0200
From: @swingbit
UPDATE:
If I issue the same CREATE FILTER FUNCTION command in both the "sys" and my custom schema, then I can successfully use the filter function in the custom schema.
## Comment 20820
Date: 2015-04-19 21:48:21 +0200
From: @njnes
This is fixed (only) in default. There we changed filter functions calling interface as well (ie change to [a1,a2] filter.func [b1,b...].
## Comment 21189
Date: 2015-08-28 13:41:50 +0200
From: @sjoerdmullender
Jul2015 has been released.
| Filter function not found if created in a user schema | https://api.github.com/repos/MonetDB/MonetDB/issues/3702/comments | 0 | 2020-11-30T13:00:59Z | 2024-06-27T12:06:57Z | https://github.com/MonetDB/MonetDB/issues/3702 | 753,459,403 | 3,702 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-04-15 06:21:51 +0200
From: Raghu <<rdubey>>
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
CC: mtucker, @njnes
Last updated: 2015-05-15 12:55:21 +0200
## Comment 20796
Date: 2015-04-15 06:21:51 +0200
From: Raghu <<rdubey>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:37.0) Gecko/20100101 Firefox/37.0
Build Identifier:
I am using JDBC driver 2.14 as of now. I have created a primary key constraint on a table and expect it to fail and tell me on which line it failed. That is the behaviour I would expect from JDBC driver so that I know the culprit row.
It appears that the MonetDB driver does not behave in a similar fashion as other JDBC drivers, in that it is throwing an error on the "COMMIT" statement, instead of throwing an error when the row of data is actually inserted. This causes an issue as the error is no longer tied to a specific incoming data row.
Here is what it says. I am trying to get it to work with Pentaho. Just ignore the pentaho stuff and search for the 'ERROR'.
---
2015/04/09 15:06:23 - LoadFactSales_test - Transformation is killing the other steps!
2015/04/09 15:06:23 - LoadFactSales_test - Looking at step: MongoDB Input
2015/04/09 15:06:23 - myportal - Commit on database connection [myportal]
2015/04/09 15:06:23 - Table output.0 - ERROR (version 5.3.0.0-213, build 1 from 2015-02-02_12-17-08 by buildguy) : Unexpected error committing the database connection.
2015/04/09 15:06:23 - Table output.0 - ERROR (version 5.3.0.0-213, build 1 from 2015-02-02_12-17-08 by buildguy) : org.pentaho.di.core.exception.KettleDatabaseException:
2015/04/09 15:06:23 - Table output.0 - Error comitting connection
2015/04/09 15:06:23 - Table output.0 - COMMIT: failed
2015/04/09 15:06:23 - Table output.0 -
2015/04/09 15:06:23 - Table output.0 - at org.pentaho.di.core.database.Database.commit(Database.java:755)
2015/04/09 15:06:23 - Table output.0 - at org.pentaho.di.core.database.Database.commit(Database.java:727)
2015/04/09 15:06:23 - Table output.0 - at org.pentaho.di.core.database.Database.emptyAndCommit(Database.java:1331)
2015/04/09 15:06:23 - Table output.0 - at org.pentaho.di.trans.steps.tableoutput.TableOutput.dispose(TableOutput.java:571)
2015/04/09 15:06:23 - Table output.0 - at org.pentaho.di.trans.step.RunThread.run(RunThread.java:96)
2015/04/09 15:06:23 - Table output.0 - at java.lang.Thread.run(Unknown Source)
2015/04/09 15:06:23 - Table output.0 - Caused by: java.sql.SQLException: COMMIT: failed
2015/04/09 15:06:23 - Table output.0 - at nl.cwi.monetdb.jdbc.MonetConnection$ResponseList.executeQuery(MonetConnection.java:2535)
2015/04/09 15:06:23 - Table output.0 - at nl.cwi.monetdb.jdbc.MonetConnection$ResponseList.processQuery(MonetConnection.java:2284)
2015/04/09 15:06:23 - Table output.0 - at nl.cwi.monetdb.jdbc.MonetConnection.commit(MonetConnection.java:374)
2015/04/09 15:06:23 - Table output.0 - at org.pentaho.di.core.database.Database.commit(Database.java:746)
2015/04/09 15:06:23 - Table output.0 - ... 5 more
2015/04/09 15:06:23 - Table output.0 - ERROR (version 5.3.0.0-213, build 1 from 2015-02-02_12-17-08 by buildguy) : org.pentaho.di.core.exception.KettleDatabaseException:
2015/04/09 15:06:23 - Table output.0 - Error comitting connection
2015/04/09 15:06:23 - Table output.0 - COMMIT: failed
2015/04/09 15:06:23 - Table output.0 -
2015/04/09 15:06:23 - Table output.0 - at org.pentaho.di.core.database.Database.commit(Database.java:755)
2015/04/09 15:06:23 - Table output.0 - at org.pentaho.di.core.database.Database.commit(Database.java:727)
2015/04/09 15:06:23 - Table output.0 - at org.pentaho.di.core.database.Database.emptyAndCommit(Database.java:1331)
2015/04/09 15:06:23 - Table output.0 - at org.pentaho.di.trans.steps.tableoutput.TableOutput.dispose(TableOutput.java:571)
2015/04/09 15:06:23 - Table output.0 - at org.pentaho.di.trans.step.RunThread.run(RunThread.java:96)
2015/04/09 15:06:23 - Table output.0 - at java.lang.Thread.run(Unknown Source)
2015/04/09 15:06:23 - Table output.0 - Caused by: java.sql.SQLException: COMMIT: failed
2015/04/09 15:06:23 - Table output.0 - at nl.cwi.monetdb.jdbc.MonetConnection$ResponseList.executeQuery(MonetConnection.java:2535)
2015/04/09 15:06:23 - Table output.0 - at nl.cwi.monetdb.jdbc.MonetConnection$ResponseList.processQuery(MonetConnection.java:2284)
2015/04/09 15:06:23 - Table output.0 - at nl.cwi.monetdb.jdbc.MonetConnection.commit(MonetConnection.java:374)
2015/04/09 15:06:23 - Table output.0 - at org.pentaho.di.core.database.Database.commit(Database.java:746)
2015/04/09 15:06:23 - Table output.0 - ... 5 more
2015/04/09 15:06:23 - Table output.0 - Signaling 'output done' to 0 output rowsets.
----
Thanks
Raghu
Reproducible: Always
### Expected Results:
It should fail and confirm which row it failed on and why?
## Comment 20870
Date: 2015-05-15 12:55:21 +0200
From: @njnes
raghu monetdb doesn't do deferred key checking but (as other dbms-es) uses immediate checking (on a per statement basis). So any key failure should have been reported earlier. From your output I cannot see what when wrong (other
then a transaction failure).
| Ambiguous message - JDBC driver issue | https://api.github.com/repos/MonetDB/MonetDB/issues/3701/comments | 0 | 2020-11-30T13:00:56Z | 2024-06-27T12:06:56Z | https://github.com/MonetDB/MonetDB/issues/3701 | 753,459,369 | 3,701 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-04-12 17:03:06 +0200
From: John <<caroptions>>
To: SQL devs <<bugs-sql>>
Version: 11.21.19 (Jul2015-SP4)
CC: martin.van.dinther, @njnes
Last updated: 2018-08-08 10:31:49 +0200
## Comment 20786
Date: 2015-04-12 17:03:06 +0200
From: John <<caroptions>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_3) AppleWebKit/600.5.17 (KHTML, like Gecko) Version/8.0.5 Safari/600.5.17
Build Identifier:
drop function mmm(timestamptz,timestamptz);
create function mmm(ts1 timestamptz, ts2 timestamptz) returns string
begin
DECLARE t1 timetz;
DECLARE t2 timetz;
set t1=cast(ts1 as timetz);
set t2=cast(ts2 as timetz);
if t1>t2 THEN
return 'T1 > T2: ' || t1 || '<=>' || t2 ;
elseif t1=t2 THEN
return 'T1 = T2: ' || t1 || '<=>' || t2 ;
else
return 'T1 < T2: ' || t1 || '<=>' || t2 ;
end if;
end;
select mmm('2001-01-01 19:00:00 -04:00','2001-01-01 23:00:00 -04:00');
RESULT:
T1 > T2: 19:00:00.000000-04:00<=>23:00:00.000000-04:00
Reproducible: Always
### Steps to Reproduce:
drop function mmm(timestamptz,timestamptz);
create function mmm(ts1 timestamptz, ts2 timestamptz) returns string
begin
DECLARE t1 timetz;
DECLARE t2 timetz;
set t1=cast(ts1 as timetz);
set t2=cast(ts2 as timetz);
if t1>t2 THEN
return 'T1 > T2: ' || t1 || '<=>' || t2 ;
elseif t1=t2 THEN
return 'T1 = T2: ' || t1 || '<=>' || t2 ;
else
return 'T1 < T2: ' || t1 || '<=>' || t2 ;
end if;
end;
select mmm('2001-01-01 19:00:00 -04:00','2001-01-01 23:00:00 -04:00');
RESULT:
T1 > T2: 19:00:00.000000-04:00<=>23:00:00.000000-04:00
### Actual Results:
T1 > T2: 19:00:00.000000-04:00<=>23:00:00.000000-04:00
### Expected Results:
T1 < T2: 19:00:00.000000-04:00<=>23:00:00.000000-04:00
it seems time conversion converts TIME to UTC and then compare. This way
23:00 become -> 3:00 AM and 19:00 -> 23:00 and accordingly 23:00 is bigger than 03:00.
This bug does not allow correctly compare times in range from 24:00 - TZ to 24:00
## Comment 20787
Date: 2015-04-12 17:26:56 +0200
From: John <<caroptions>>
Statement:
SET TIME ZONE INTERVAL '-04:00' HOUR TO MINUTE;
Didn't help either.
Not sure what would be the correct functionality, but it seems the time should be converted not to UTC, but to session time zone, or it should be specified somehow which time zone to use.
And at least if time zone for both parameters is the same use time zone of parameters.
Please pay attention that session time zone setting does not affect any function as well as statement.
select cast(timestamptz '2001-01-01 19:00:00 -04:00' as timetz)< cast( timestamptz '2001-01-01 23:00:00 -04:00' as timetz);
-> false.
## Comment 20788
Date: 2015-04-12 17:31:32 +0200
From: John <<caroptions>>
As an idea (and it seems PostgreSQL uses this approach) when you convert to UTC, you should not "reset" count at 24 but allow hours more that 24:00.
This way 23:00 -04:00 will become 26:00 UTC then the comparison will be correct regardless timezone.
## Comment 20804
Date: 2015-04-16 16:54:38 +0200
From: Martin van Dinther <<martin.van.dinther>>
Hi John,
Your function cast/convert a timestamptz value into a timetz value. During this step the date part information is cut, so you loose (day) information.
To get your function mmm() to work properly you can replace the lines:
if t1>t2 THEN
return 'T1 > T2: ' || t1 || '<=>' || t2 ;
elseif t1=t2 THEN
into
if ts1>ts2 THEN
return 'T1 > T2: ' || t1 || '<=>' || t2 ;
elseif ts1=ts2 THEN
This way you compare the timestamptz values (including the date part).
This may be a workaround for your application.
It is not a fix of the reported issue of course.
## Comment 21065
Date: 2015-07-29 16:16:25 +0200
From: @sjoerdmullender
MonetDB stores timestamps in UTC. Time zones are only used during ingestion and display of timestamp values. When mclient (or indeed, the ODBC and JDBC drivers) connects, it first sends a query like the one you gave in comment 1, so it doesn't surprise me in the least that that didn't make a difference.
Changing this would be a pretty major undertaking.
## Comment 26591
Date: 2018-08-08 10:31:49 +0200
From: @njnes
the storage for timestamp's with and without time zone's is (in size) equal. No
room for storing the timezone. The timestamps are converted to gmt asap, ie (when converted from string).
There are no plans to change this, ie this problem is likely to stay.
| time comparison is wrong | https://api.github.com/repos/MonetDB/MonetDB/issues/3700/comments | 0 | 2020-11-30T13:00:54Z | 2024-06-28T13:15:39Z | https://github.com/MonetDB/MonetDB/issues/3700 | 753,459,343 | 3,700 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-04-11 20:10:05 +0200
From: John <<caroptions>>
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
CC: @njnes
Last updated: 2015-05-07 12:37:45 +0200
## Comment 20782
Date: 2015-04-11 20:10:05 +0200
From: John <<caroptions>>
CREATE USER "trader" WITH PASSWORD '1234' NAME 'Market User' SCHEMA "sys";
CREATE SCHEMA "marketdata" AUTHORIZATION "trader";
ALTER USER "trader" SET SCHEMA "marketdata";
CREATE SEQUENCE marketdata.seq_quotes_1 AS INTEGER ;
CREATE TABLE marketdata.quotes (
"id" INTEGER NOT NULL DEFAULT next value for marketdata.seq_quotes_1,
"qtime" TIMESTAMP WITH TIME ZONE NOT NULL,
"sdate" TIMESTAMP WITH TIME ZONE,
"sym" VARCHAR(10) NOT NULL,
"cur" VARCHAR(10) NOT NULL,
"open" DOUBLE NOT NULL,
"high" DOUBLE NOT NULL,
"low" DOUBLE NOT NULL,
"close" DOUBLE NOT NULL,
"volume" DOUBLE,
CONSTRAINT quotes_id_pkey PRIMARY KEY ("id")
);
CREATE FUNCTION marketdata.ohlc_d2(psym text, since date, close_time time, till date)
RETURNS TABLE(sdate date, open double, high double, low double, close double, volume double)
BEGIN
DECLARE tsince DATE;
IF since is NULL THEN
SET tsince = DATE '1990-01-01';
ELSE
SET tsince = since;
END IF;
IF close_time is NULL THEN
RETURN TABLE ( SELECT * FROM (
WITH open(sdate, open, rnk) as (
select sdate, open, row_number() over (partition by sdate order by qtime asc) as rnk from marketdata.quotes
where sym=psym and sdate is not NULL and cast(sdate as date) > tsince
),
close(sdate, close, rnk) as (
select sdate, close, row_number() over (partition by sdate order by qtime desc) as rnk from marketdata.quotes
where sym=psym and sdate is not NULL and cast(sdate as date) > tsince
),
hilw(sdate,high,low, volume) as (
select sdate, max(high) as high, min(low) as low, sum(volume) from marketdata.quotes
where sym=psym and sdate is not NULL and cast(sdate as date) > tsince group by sdate
)
select open.sdate, open.open, hilw.high, hilw.low, close.close , hilw.volume from open, close, hilw where open.rnk=1 and close.rnk=1 and open.sdate=close.sdate and open.sdate=hilw.sdate ORDER BY sdate DESC
) AS x
);
ELSE
RETURN TABLE ( SELECT * FROM (
WITH open(sdate, open, rnk) as (
select sdate, open, row_number() over (partition by sdate order by qtime asc) as rnk from marketdata.quotes
where sym=psym and sdate is not NULL and cast(sdate as date) > tsince
),
close(sdate, close, rnk) as (
select sdate, close, row_number() over (partition by sdate order by qtime desc) as rnk from marketdata.quotes
where sym=psym and sdate is not NULL and cast(sdate as date) > tsince
),
hilw(sdate,high,low, volume) as (
select sdate, max(high) as high, min(low) as low, sum(volume) from marketdata.quotes
where sym=psym and sdate is not NULL and cast(sdate as date) > tsince group by sdate
)
select open.sdate, open.open, hilw.high, hilw.low, close.close , hilw.volume from open, close, hilw where open.rnk=1 and close.rnk=1 and open.sdate=close.sdate and open.sdate=hilw.sdate ORDER BY sdate DESC
) AS x
);
END IF;
END;
## Comment 20783
Date: 2015-04-12 00:13:46 +0200
From: John <<caroptions>>
Even simpler:
DROP FUNCTION marketdata.test(text);
CREATE FUNCTION marketdata.test(a text) RETURNS table(SOMEFIELD date)
BEGIN
IF a is NULL THEN
return table(
select cast(sdate as date) from marketdata.quotes limit 10
);
ELSE
return table(
select cast(sdate as date) from marketdata.quotes limit 10
);
END IF;
END;
## Comment 20784
Date: 2015-04-12 00:36:58 +0200
From: John <<caroptions>>
One more case:
CREATE FUNCTION marketdata.test(a text) RETURNS table(SOMEFIELD double)
BEGIN
declare table t(f double);
select 0.0 into t;
return t;
END;
I wonder if somebody used mounted before?
## Comment 20785
Date: 2015-04-12 00:38:56 +0200
From: John <<caroptions>>
CREATE FUNCTION marketdata.test(a text) RETURNS table(SOMEFIELD double)
BEGIN
declare table t(f double);
set t = table (select 0.0);
return t;
END;
## Comment 20807
Date: 2015-04-17 13:02:57 +0200
From: @njnes
just tested your last bit of example.
set t = table(...) isn't part of the sql standard.
Change this into insert into t (select ..).
That runs and doesn't crash in my tests.
## Comment 20810
Date: 2015-04-17 17:35:44 +0200
From: John <<caroptions>>
DROP FUNCTION marketdata.test(text);
CREATE FUNCTION marketdata.test(a text) RETURNS table(SOMEFIELD date)
BEGIN
IF a is NULL THEN
return table(
select cast(sdate as date) from marketdata.quotes limit 10
);
ELSE
return table(
select cast(sdate as date) from marketdata.quotes limit 10
);
END IF;
END;
Still crashes the DB.
Side question. Why statement in function: RETURN TABLE ( WITH .... SELECT ...) generates syntax error?
## Comment 20823
Date: 2015-04-19 22:50:11 +0200
From: @njnes
found (and fixed) a problem with the type checking of the if then statement.
The with is currently not supported in the return, will be added in
the next feature release.
| segfault again! (during last week I found 3 segfault bugs already) | https://api.github.com/repos/MonetDB/MonetDB/issues/3699/comments | 0 | 2020-11-30T13:00:51Z | 2024-06-27T12:06:54Z | https://github.com/MonetDB/MonetDB/issues/3699 | 753,459,312 | 3,699 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-04-10 23:43:28 +0200
From: John <<caroptions>>
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
CC: @njnes
Last updated: 2015-04-17 17:31:18 +0200
## Comment 20781
Date: 2015-04-10 23:43:28 +0200
From: John <<caroptions>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_3) AppleWebKit/600.5.17 (KHTML, like Gecko) Version/8.0.5 Safari/600.5.17
Build Identifier:
-- as any regular user
CREATE FUNCTION user.db_users () RETURNS TABLE( name varchar(2048)) EXTERNAL NAME sql.db_users;
select * from user.db_users();
Reproducible: Always
### Steps to Reproduce:
-- as any regular user ru script:
CREATE FUNCTION user.db_users () RETURNS TABLE( name varchar(2048)) EXTERNAL NAME sql.db_users;
select * from user.db_users();
### Actual Results:
mserver5[19791]: segfault at 0 ip 00007f2f0d09e64c sp 00007f2f0bd76960 error 4 in lib_sql.so
### Expected Results:
no segfault
## Comment 20806
Date: 2015-04-17 12:49:45 +0200
From: @njnes
I'm unable to reproduce your crash on the oct2014 branch, nor on the default branch.
## Comment 20809
Date: 2015-04-17 17:31:18 +0200
From: John <<caroptions>>
I've just checked on my latest compiled from sources monetdb instance. It works as intended and didn't crash. Strange.. I agree with close... Will observe further..
| segfault at 0 error 4 in lib_sql.so | https://api.github.com/repos/MonetDB/MonetDB/issues/3698/comments | 0 | 2020-11-30T13:00:49Z | 2024-06-27T12:06:53Z | https://github.com/MonetDB/MonetDB/issues/3698 | 753,459,283 | 3,698 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-04-08 05:14:47 +0200
From: John <<caroptions>>
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
CC: martin.van.dinther, @njnes
Last updated: 2015-05-07 12:37:40 +0200
## Comment 20762
Date: 2015-04-08 05:14:47 +0200
From: John <<caroptions>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_2) AppleWebKit/600.4.10 (KHTML, like Gecko) Version/8.0.4 Safari/600.4.10
Build Identifier:
DB crash every time I execute statement:
WITH cte(qtime, open, close, sdate, id, rnk) as (
select qtime, sdate, id, row_number() over (partition by sdate order by qtime asc) as rnk
from marketdata.quotes where sym='SPY'
) select * from cte where rnk=1;
Reproducible: Always
### Steps to Reproduce:
1. Create New DB, Create Table, Load data ~3M records
2. Connect over jdbc to monetdb server
3. Execute statement specified above.
### Actual Results:
368925.504436] mserver5[26946]: segfault at 0 ip 00007f3d0e1ab808 sp 00007f3cefbfcad0 error 4 in lib_sql.so[7f3d0e180000+16c000]
### Expected Results:
No crash
System: CentOS Linux release 7.1.1503 (Core) 32 GB,
DB -> SSD Disk, mount options: rw,noexec,nodev,noatime,nodiratime,nobarrier
Bug is repeatable with package installed from yum repository as well as compiled manually.
## Comment 20764
Date: 2015-04-08 19:19:03 +0200
From: @njnes
I'm missing the ddl statement to create the table.
## Comment 20765
Date: 2015-04-08 22:02:54 +0200
From: John <<caroptions>>
Sorry for missing this information,
Here you go:
CREATE TABLE "marketdata"."quotes" (
"id" INTEGER NOT NULL DEFAULT next value for "marketdata"."seq_6391",
"qtime" TIMESTAMP WITH TIME ZONE NOT NULL,
"sdate" TIMESTAMP WITH TIME ZONE,
"sym" VARCHAR(10) NOT NULL,
"cur" VARCHAR(10) NOT NULL,
"open" DOUBLE NOT NULL,
"high" DOUBLE NOT NULL,
"low" DOUBLE NOT NULL,
"close" DOUBLE NOT NULL,
"volume" DOUBLE,
CONSTRAINT "quotes_id_pkey" PRIMARY KEY ("id")
);
if any additional info required, just let me know.
## Comment 20766
Date: 2015-04-09 14:01:22 +0200
From: Martin van Dinther <<martin.van.dinther>>
For a working DDL, you will also need to create a sequence first, e.g.:
CREATE SEQUENCE "seq_6391" AS INTEGER ;
Next the:
CREATE TABLE "quotes" (
"id" INTEGER NOT NULL DEFAULT next value for "seq_6391",
"qtime" TIMESTAMP WITH TIME ZONE NOT NULL,
"sdate" TIMESTAMP WITH TIME ZONE,
"sym" VARCHAR(10) NOT NULL,
"cur" VARCHAR(10) NOT NULL,
"open" DOUBLE NOT NULL,
"high" DOUBLE NOT NULL,
"low" DOUBLE NOT NULL,
"close" DOUBLE NOT NULL,
"volume" DOUBLE,
CONSTRAINT "quotes_id_pkey" PRIMARY KEY ("id")
);
It appears that the query:
WITH cte(qtime, open, close, sdate, id, rnk) as (
select qtime, sdate, id, row_number() over (partition by sdate order by qtime asc) as rnk
from marketdata.quotes where sym='SPY'
) select * from cte where rnk=1;
is not well defined. The select has less columns (4) than the WITH definition (6).
The corrected version would be:
WITH cte(qtime, open, close, sdate, id, rnk) as (
select qtime, open, close, sdate, id, row_number() over (partition by sdate order by qtime asc) as rnk
from marketdata.quotes where sym='SPY'
) select * from cte where rnk=1;
Alternatively you can do:
CREATE VIEW cte (qtime, open, close, sdate, id, rnk) as
select qtime, open, close, sdate, id, row_number() over (partition by sdate order by qtime asc) as rnk
from "quotes" where sym='SPY';
and next use the simple query:
select * from cte where rnk=1;
I tested above on 11.19.9 SP2 and on 11.20.0 (dev version) and the reported problem did not occurr, so it is NOT reproducable without data.
John, could you please provide data (preferably a subset) which allows us to reproduce the problem.
Also try yourself the version with the VIEW instead of the WITH and let us know if that makes a difference.
## Comment 20768
Date: 2015-04-09 16:29:20 +0200
From: John <<caroptions>>
Created attachment 331
Data Subset for bug reproduction
> Attached file: [SPY.csvaa.zip](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3697_SPY.csvaa.zip_331) (application/zip, 946778 bytes)
> Description: Data Subset for bug reproduction
## Comment 20769
Date: 2015-04-09 17:03:29 +0200
From: John <<caroptions>>
Hi Martin,
I have attached the DataSet, I've checked it and the bug is reproducible on this specific file. So for bug debuging purposes it should work.
Thank you for pointing out on the problem with sql! I've tried your edition and it works fine and DB did not crash.
It seems the bug is related exactly to the not valid SQL.
Even though the criticality of this bug is much lower now for my specific case I am available to continue debug and provide any additional information.
Thanks,
John
## Comment 20770
Date: 2015-04-10 09:45:46 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [855e0d266900](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=855e0d266900) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=855e0d266900](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=855e0d266900)
Changeset description:
Check number of columns in WITH clause with those in subquery.
This fixes bug #3697.
## Comment 20771
Date: 2015-04-10 09:47:27 +0200
From: @sjoerdmullender
The bug was that there was no proper check that the number of columns specified in the WITH clause was equal to the number of columns produced in the query. The fix doesn't make the query run but it does make that the server doesn't crash.
## Comment 20779
Date: 2015-04-10 16:23:22 +0200
From: John <<caroptions>>
Hi,
I've checked change set and see:
if (!create) {
+ if (column_spec) {
+ dnode *n = column_spec->h;
+ node *m = sq->exps->h;
+
+ for (; n && m; n = n->next, m = m->next)
+ ;
+ if (n || m) {
+ sql_error(sql, 01, "21S02!WITH CLAUSE: number of columns does not match");
+ rel_destroy(sq);
+ return NULL;
+ }
+ }
Is it intention was:
+ for (; n && m; n = n->next, m = m->next)
+ //Need to be removed --> ;
+ if (n || m) { ...
Just checking ...
Thanks,
John
## Comment 20780
Date: 2015-04-10 17:03:43 +0200
From: @sjoerdmullender
No, the intention is what's in the code.
The for loop just loops until the end of either list, the if then checks whether that was the end of both lists.
| mserver5[26946]: segfault at 0 ip 00007f3d0e1ab808 sp 00007f3cefbfcad0 error 4 in lib_sql.so[7f3d0e180000+16c000] | https://api.github.com/repos/MonetDB/MonetDB/issues/3697/comments | 0 | 2020-11-30T13:00:46Z | 2024-06-27T12:06:52Z | https://github.com/MonetDB/MonetDB/issues/3697 | 753,459,259 | 3,697 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-04-02 17:37:48 +0200
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
Last updated: 2015-05-07 12:38:10 +0200
## Comment 20758
Date: 2015-04-02 17:37:48 +0200
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Fedora; Linux x86_64; rv:36.0) Gecko/20100101 Firefox/36.0
Build Identifier:
A double accepts conversion of special values such as 'Inf', 'Infinity', 'inf', 'infinity' where a float (SQL real) does not accept these (return an error). However the negative infinity ('-Inf', '-Infinity', '-inf', '-infinity') is accepted for both data types. A strange inconsistency.
Also on MS Windows for a float (SQL real) the special values 'NaN', '-NaN', 'Inf', 'Infinity', 'inf', 'infinity', '-Inf', '-Infinity', '-inf', '-infinity' are NOT accepted at all resulting in different MonetDB behavior across platforms.
See for instance http://monetdb.cwi.nl/testweb/web/showtestoutput.php?serial=55220:750c51b86faf&target=Mic-Windows7-x86_64-installer&module=sql&test=sql%2Ftest%2Fpg_regress%2Ffloat4&which=out
Also it appears that conversions resulting in overflow (such as '1e+39' or '-1e+39') for a float (SQL real) is a accepted on MS Windows where it is not accepted on other platforms. This difference should also not occur.
Also it appears that string representation of Infinity is different on different platforms. On Linux it is 'inf' or '-inf', On GNU-Solaris-sparcv9-dbfarm and GNU-Solaris-sparc-dbfarm and GNU-OpenIndiana-x86_64 it is 'Inf' or '-Inf', so with a capital I. This output difference should also be harmonised across all supported platforms.
Reproducible: Always
### Steps to Reproduce:
1. Start mserver5 (MonetDB v11.19.9 (Oct2014-SP2))
2. Start mclient (or a JDBC client)
3. Execute SQL commands:
CREATE TABLE DOUBLE_TBL(x double);
INSERT INTO DOUBLE_TBL(x) VALUES ('NaN');
SELECT x, cast(x as varchar(30)) as x_str FROM DOUBLE_TBL;
-- lists 1 row with null as double and as string value (in JDBC)
INSERT INTO DOUBLE_TBL(x) VALUES ('-NaN');
SELECT x, cast(x as varchar(30)) as x_str FROM DOUBLE_TBL;
-- lists 2 rows with null as double and as string value (in JDBC)
INSERT INTO DOUBLE_TBL(x) VALUES ('Inf');
INSERT INTO DOUBLE_TBL(x) VALUES ('Infinity');
INSERT INTO DOUBLE_TBL(x) VALUES ('inf');
INSERT INTO DOUBLE_TBL(x) VALUES ('infinity');
SELECT x, cast(x as varchar(30)) as x_str FROM DOUBLE_TBL;
-- lists 6 rows. last 4 return 'inf' as string value, o as double value (in JDBC)
INSERT INTO DOUBLE_TBL(x) VALUES ('-Inf');
INSERT INTO DOUBLE_TBL(x) VALUES ('-Infinity');
INSERT INTO DOUBLE_TBL(x) VALUES ('-inf');
INSERT INTO DOUBLE_TBL(x) VALUES ('-infinity');
SELECT x, cast(x as varchar(30)) as x_str FROM DOUBLE_TBL;
-- lists 10 rows. last 4 return '-inf' as string value, o as double value (in JDBC)
CREATE TABLE REAL_TBL(x real);
INSERT INTO REAL_TBL(x) VALUES ('NaN');
SELECT x, cast(x as varchar(30)) as x_str FROM REAL_TBL;
-- lists 1 row with null as double and as string value (in JDBC)
INSERT INTO REAL_TBL(x) VALUES ('-NaN');
SELECT x, cast(x as varchar(30)) as x_str FROM REAL_TBL;
-- lists 2 rows with null as double and as string value (in JDBC)
INSERT INTO REAL_TBL(x) VALUES ('Inf');
-- Error: conversion of string 'Inf' to type flt failed. SQLState: 22018 ErrorCode: 0
INSERT INTO REAL_TBL(x) VALUES ('Infinity');
-- Error: conversion of string 'Infinity' to type flt failed. SQLState: 22018 ErrorCode: 0
INSERT INTO REAL_TBL(x) VALUES ('inf');
-- Error: conversion of string 'inf' to type flt failed. SQLState: 22018 ErrorCode: 0
INSERT INTO REAL_TBL(x) VALUES ('infinity');
-- Error: conversion of string 'infinity' to type flt failed. SQLState: 22018 ErrorCode: 0
SELECT x, cast(x as varchar(30)) as x_str FROM REAL_TBL;
-- lists 2 instead of 6 rows.
INSERT INTO REAL_TBL(x) VALUES ('-Inf');
INSERT INTO REAL_TBL(x) VALUES ('-Infinity');
INSERT INTO REAL_TBL(x) VALUES ('-inf');
INSERT INTO REAL_TBL(x) VALUES ('-infinity');
SELECT x, cast(x as varchar(30)) as x_str FROM REAL_TBL;
-- lists 6 rows. last 4 return '-inf' as string value, o as double value (in JDBC)
-- test overflow for real
INSERT INTO REAL_TBL(x) VALUES ('1e+39');
-- this should result in: Error: conversion of string '1e+39' to type flt failed. SQLState: 22018 ErrorCode: 0
INSERT INTO REAL_TBL(x) VALUES ('-1e+39');
-- this should result in: Error: conversion of string '-1e+39' to type flt failed. SQLState: 22018 ErrorCode: 0
-- cleanup
DROP TABLE DOUBLE_TBL;
DROP TABLE REAL_TBL;
### Actual Results:
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.19.9 (Oct2014-SP2), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>CREATE TABLE DOUBLE_TBL(x double);
operation successful (1.499ms)
sql>INSERT INTO DOUBLE_TBL(x) VALUES ('NaN');
1 affected row (0.714ms)
sql>SELECT x, cast(x as varchar(30)) as x_str FROM DOUBLE_TBL;
+--------------------------+-------+
| x | x_str |
+==========================+=======+
| null | null |
+--------------------------+-------+
1 tuple (1.150ms)
sql>INSERT INTO DOUBLE_TBL(x) VALUES ('-NaN');
1 affected row (0.519ms)
sql>SELECT x, cast(x as varchar(30)) as x_str FROM DOUBLE_TBL;
+--------------------------+-------+
| x | x_str |
+==========================+=======+
| null | null |
| null | null |
+--------------------------+-------+
2 tuples (0.365ms)
sql>INSERT INTO DOUBLE_TBL(x) VALUES ('Inf');
1 affected row (0.308ms)
sql>INSERT INTO DOUBLE_TBL(x) VALUES ('Infinity');
1 affected row (0.519ms)
sql>INSERT INTO DOUBLE_TBL(x) VALUES ('inf');
1 affected row (0.401ms)
sql>INSERT INTO DOUBLE_TBL(x) VALUES ('infinity');
1 affected row (0.376ms)
sql>SELECT x, cast(x as varchar(30)) as x_str FROM DOUBLE_TBL;
+--------------------------+-------+
| x | x_str |
+==========================+=======+
| null | null |
| null | null |
| inf | inf |
| inf | inf |
| inf | inf |
| inf | inf |
+--------------------------+-------+
6 tuples (0.487ms)
sql>INSERT INTO DOUBLE_TBL(x) VALUES ('-Inf');
1 affected row (0.292ms)
sql>INSERT INTO DOUBLE_TBL(x) VALUES ('-Infinity');
1 affected row (0.395ms)
sql>INSERT INTO DOUBLE_TBL(x) VALUES ('-inf');
1 affected row (0.279ms)
sql>INSERT INTO DOUBLE_TBL(x) VALUES ('-infinity');
1 affected row (0.257ms)
sql>SELECT x, cast(x as varchar(30)) as x_str FROM DOUBLE_TBL;
+--------------------------+-------+
| x | x_str |
+==========================+=======+
| null | null |
| null | null |
| inf | inf |
| inf | inf |
| inf | inf |
| inf | inf |
| -inf | -inf |
| -inf | -inf |
| -inf | -inf |
| -inf | -inf |
+--------------------------+-------+
10 tuples (0.302ms)
sql>
sql>CREATE TABLE REAL_TBL(x real);
operation successful (1.570ms)
sql>INSERT INTO REAL_TBL(x) VALUES ('NaN');
1 affected row (0.533ms)
sql>SELECT x, cast(x as varchar(30)) as x_str FROM REAL_TBL;
+-----------------+-------+
| x | x_str |
+=================+=======+
| null | null |
+-----------------+-------+
1 tuple (1.671ms)
sql>INSERT INTO REAL_TBL(x) VALUES ('-NaN');
1 affected row (0.449ms)
sql>SELECT x, cast(x as varchar(30)) as x_str FROM REAL_TBL;
+-----------------+-------+
| x | x_str |
+=================+=======+
| null | null |
| null | null |
+-----------------+-------+
2 tuples (0.333ms)
sql>INSERT INTO REAL_TBL(x) VALUES ('Inf');
conversion of string 'Inf' to type flt failed.
sql>INSERT INTO REAL_TBL(x) VALUES ('Infinity');
conversion of string 'Infinity' to type flt failed.
sql>INSERT INTO REAL_TBL(x) VALUES ('inf');
conversion of string 'inf' to type flt failed.
sql>INSERT INTO REAL_TBL(x) VALUES ('infinity');
conversion of string 'infinity' to type flt failed.
sql>SELECT x, cast(x as varchar(30)) as x_str FROM REAL_TBL;
+-----------------+-------+
| x | x_str |
+=================+=======+
| null | null |
| null | null |
+-----------------+-------+
2 tuples (0.547ms)
sql>INSERT INTO REAL_TBL(x) VALUES ('-Inf');
1 affected row (0.380ms)
sql>INSERT INTO REAL_TBL(x) VALUES ('-Infinity');
1 affected row (0.370ms)
sql>INSERT INTO REAL_TBL(x) VALUES ('-inf');
1 affected row (0.384ms)
sql>INSERT INTO REAL_TBL(x) VALUES ('-infinity');
1 affected row (0.288ms)
sql>SELECT x, cast(x as varchar(30)) as x_str FROM REAL_TBL;
+-----------------+-------+
| x | x_str |
+=================+=======+
| null | null |
| null | null |
| -inf | -inf |
| -inf | -inf |
| -inf | -inf |
| -inf | -inf |
+-----------------+-------+
6 tuples (0.302ms)
sql>INSERT INTO REAL_TBL(x) VALUES ('1e+39');
conversion of string '1e+39' to type flt failed.
sql>INSERT INTO REAL_TBL(x) VALUES ('-1e+39');
conversion of string '-1e+39' to type flt failed.
sql>
sql>DROP TABLE DOUBLE_TBL;
operation successful (1.169ms)
sql>DROP TABLE REAL_TBL;
operation successful (0.899ms)
sql>
### Expected Results:
Scanning and conversion of special values ('Nan', 'Inf', '-Inf') and overflow values to a double or a float (SQL real) should be treated in a uniform way and the same across platforms.
Also the string output should be consistent across platform for these special double and float values.
See also tests in sql/test/pg_regress/float4.sql and sql/test/pg_regress/float8.sql
In gdk/gdk_atoms.c the implementation of fltFromStr() should be more similar to dblFromStr().
## Comment 20776
Date: 2015-04-10 14:23:49 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [61ee04c1073b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=61ee04c1073b) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=61ee04c1073b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=61ee04c1073b)
Changeset description:
Don't accept strings like NaN or Infinite when converting to floating point.
The SQL standard says (ISO_9075_02_Foundation_2011_E, 6.13, General
Rules 9, b) that when converting a string to an approximate number,
the string must conform to the rules for a "signed numeric literal"
which is an optional sign, digits, period, more digits, exponent (with
some optional bits). But definitely not a string such as "NaN". This
fixes bug #3696.
## Comment 20791
Date: 2015-04-14 10:42:11 +0200
From: @sjoerdmullender
Fixed with changesets [61ee04c1073b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=61ee04c1073b) and [9c80de069c44](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9c80de069c44)
## Comment 20800
Date: 2015-04-16 12:47:00 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [9dc8e2433dc3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9dc8e2433dc3) made by Martin van Dinther <martin.van.dinther@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=9dc8e2433dc3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=9dc8e2433dc3)
Changeset description:
With fix for bug #3696 the special values 'nan', 'Inf', 'Infinity', 'inf', 'infinity' are no longer accepted for floating point data types.
Updated the float4 and float8 outputs.
| Inconsistent behavior between dbl (SQL double) and flt (SQL real) data types and across platforms | https://api.github.com/repos/MonetDB/MonetDB/issues/3696/comments | 0 | 2020-11-30T13:00:43Z | 2024-06-27T12:06:51Z | https://github.com/MonetDB/MonetDB/issues/3696 | 753,459,226 | 3,696 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-04-02 17:26:06 +0200
From: Ralf Karschnick <<r.karschnick>>
To: GDK devs <<bugs-common>>
Version: 11.21.19 (Jul2015-SP4)
CC: akkaran046
Last updated: 2018-12-31 07:13:06 +0100
## Comment 20757
Date: 2015-04-02 17:26:06 +0200
From: Ralf Karschnick <<r.karschnick>>
User-Agent: Mozilla/5.0 (Windows NT 6.3; WOW64; rv:36.0) Gecko/20100101 Firefox/36.0
Build Identifier:
mserver5 crashes very often while loading files using "copy into ...". I am trying to load small files (5000 - 20000 lines) after some "BATs do not match" errors which occured after the attempt to load a lot (some Million lines) into quite a big table (about 150 Million lines). I am the only one using this machine while the crashe occur.
Reproducible: Always
OS: SUSE Linux Enterprise Server 11 (x86_64)
free:
total used free shared buffers cached
Mem: 12186476 7410124 4776352 0 8 6894256
-/+ buffers/cache: 515860 11670616
Swap: 44036268 112896 43923372
grep sql /var/log/messages yield tonnes of lines like this one:
Apr 2 16:49:37 adgbenchdb2 kernel: [7881367.696292] mserver5[25940]: segfault at 18 ip 00007f64951240ac sp 00007f6493622d60 error 4 in lib_sql.so[7f31575a4000+147000]
merovingian.log: (this happen very often, always after - lets say 10 - successful loads
2015-04-02 17:13:00 MSG merovingian[31545]: target connection is on local UNIX domain socket, passing on filedescriptor instead of proxying
2015-04-02 17:13:00 MSG merovingian[31545]: proxying client (local) for database 'ptr' to mapi:monetdb:///monetdb_vault/ptr/.mapi.sock?database=ptr
2015-04-02 17:13:00 MSG merovingian[31545]: target connection is on local UNIX domain socket, passing on filedescriptor instead of proxying
2015-04-02 17:13:00 MSG merovingian[31545]: proxying client (local) for database 'ptr' to mapi:monetdb:///monetdb_vault/ptr/.mapi.sock?database=ptr
2015-04-02 17:13:00 MSG merovingian[31545]: target connection is on local UNIX domain socket, passing on filedescriptor instead of proxying
2015-04-02 17:13:00 ERR ptr[32112]: GDKmmap(70320128) fails, try to free up space [memory in use=39612888,virtual memory in use=18284179928]
2015-04-02 17:13:00 ERR ptr[32112]: GDKmmap(70320128) result [mem=31320216,vm=26339480]
2015-04-02 17:13:00 ERR ptr[32112]: GDKmmap: recovery ok. Continuing..
2015-04-02 17:13:00 MSG merovingian[31545]: database 'ptr' (32112) was killed by signal SIGSEGV
2015-04-02 17:13:05 MSG merovingian[31545]: starting database 'ptr', up min/avg/max: 9s/4h/22h, crash average: 1.00 0.80 0.93 (217-11=206)
2015-04-02 17:13:06 MSG ptr[1717]: arguments: /usr/local/bin/mserver5 --dbpath=/monetdb_vault/ptr --set merovingian_uri=mapi:monetdb://adgbenchdb2:50000/ptr --set mapi_open=false --set mapi_port=0 --set mapi_usock=/monetdb_vault/ptr/.mapi.sock --set monet_vault_key=/monetdb_vault/ptr/.vaultkey --set gdk_nr_threads=16 --set max_clients=64 --set sql_optimizer=default_pipe --set monet_daemon=yes
2015-04-02 17:13:06 MSG ptr[1717]: MonetDB 5 server v11.19.9 "Oct2014-SP2"
2015-04-02 17:13:06 MSG ptr[1717]: Serving database 'ptr', using 16 threads
2015-04-02 17:13:06 MSG ptr[1717]: Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs dynamically linked
2015-04-02 17:13:06 MSG ptr[1717]: Found 11.622 GiB available main-memory.
2015-04-02 17:13:06 MSG ptr[1717]: Copyright (c) 1993-July 2008 CWI.
2015-04-02 17:13:06 MSG ptr[1717]: Copyright (c) August 2008-2015 MonetDB B.V., all rights reserved
2015-04-02 17:13:06 MSG ptr[1717]: Visit http://www.monetdb.org/ for further information
2015-04-02 17:13:06 MSG ptr[1717]: Listening for UNIX domain connection requests on mapi:monetdb:///monetdb_vault/ptr/.mapi.sock
2015-04-02 17:13:06 MSG ptr[1717]: MonetDB/SQL module loaded
2015-04-02 17:13:06 MSG merovingian[31545]: proxying client (local) for database 'ptr' to mapi:monetdb:///monetdb_vault/ptr/.mapi.sock?database=ptr
2015-04-02 17:13:06 MSG merovingian[31545]: target connection is on local UNIX domain socket, passing on filedescriptor instead of proxying
2015-04-02 17:13:06 MSG merovingian[31545]: proxying client (local) for database 'ptr' to mapi:monetdb:///monetdb_vault/ptr/.mapi.sock?database=ptr
2015-04-02 17:13:06 MSG merovingian[31545]: target connection is on local UNIX domain socket, passing on filedescriptor instead of proxying
2015-04-02 17:13:06 MSG merovingian[31545]: proxying client (local) for database 'ptr' to mapi:monetdb:///monetdb_vault/ptr/.mapi.sock?database=ptr
the scrip I use:
lala () {
echo $1
mclient -d ptr -s "delete from abvgaga; copy into abvgaga from STDIN " - < $1
mclient -d ptr -s "insert into abvdata select * from abvgaga; delete from abvgaga;"
}
for i in new/[0-9]*
do
lala $i
sleep 5
done
## Comment 20759
Date: 2015-04-02 17:39:06 +0200
From: Ralf Karschnick <<r.karschnick>>
just to clarify: I am not talking about the "BATs do not match" error. That happended yesterday. As a workaround I was trying to load small portions of the data. Yesterday, I had no Problems loading the Files by piping them into "mclien -d ptr -s "copy into ..." from a formatting program.
I forgot df' output:
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/cciss/c0d2p1 62879792 37538228 25341564 60% /
devtmpfs 6093236 460 6092776 1% /dev
tmpfs 6093236 16348 6076888 1% /dev/shm
/dev/cciss/c0d0p1 124427 89396 28607 76% /boot
/dev/dm-13 419328000 27677140 391650860 7% /monetdb_vault
I'll have some days off next week, so maybe I won't be able to answer
## Comment 26767
Date: 2018-12-31 07:13:06 +0100
From: martin garix <<akkaran046>>
Download mozilla firefox latest version,this is batter search browser then more,i am sure you felling very well https://www.monetdb.org/bugzilla/show_bug.cgi?id=3150 i have giving the 4 star this game,now looking here,and using the one of the simple method.
| crashes after "GDKmmap(...) fails" | https://api.github.com/repos/MonetDB/MonetDB/issues/3695/comments | 0 | 2020-11-30T13:00:40Z | 2024-06-28T13:15:38Z | https://github.com/MonetDB/MonetDB/issues/3695 | 753,459,187 | 3,695 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-04-02 14:29:24 +0200
From: Martin van Dinther <<martin.van.dinther>>
To: GDK devs <<bugs-common>>
Version: -- development
Last updated: 2015-04-09 16:17:22 +0200
## Comment 20756
Date: 2015-04-02 14:29:24 +0200
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Fedora; Linux x86_64; rv:36.0) Gecko/20100101 Firefox/36.0
Build Identifier:
mserver5: ../../dev/gdk/gdk_group.c:388: BATgroup_internal: Assertion `g == ((void *)0) || ((g)->T->type == 0 && (g)->T->seq != oid_nil ? 6 : (g)->T->type) == 6' failed.
Reproducible: Always
### Steps to Reproduce:
1. Start mserver5 (MonetDB v11.20.0, built from default branch on 2 april 2015)
2. Start mclient
3. Execute following 2 SQL commands:
create table atacc1 ( test int, test2 int);
alter table atacc1 add constraint atacc_test1 unique (test, test2);
### Actual Results:
builtin opt gdk_dbpath = /export/scratch1/dinther/INSTALL/var/monetdb5/dbfarm/demo
builtin opt gdk_debug = 0
builtin opt gdk_vmtrim = no
builtin opt monet_prompt = >
builtin opt monet_daemon = no
builtin opt mapi_port = 50000
builtin opt mapi_open = false
builtin opt mapi_autosense = false
builtin opt sql_optimizer = default_pipe
builtin opt sql_debug = 0
cmdline opt embedded_r = true
cmdline opt mapi_port = 41000
cmdline opt gdk_debug = 10
MonetDB 5 server v11.20.0
This is an unreleased version
Serving database 'demo', using 8 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs and 128bit integers dynamically linked
Found 15.590 GiB available main-memory.
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2015 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Listening for connection requests on mapi:monetdb://127.0.0.1:41000/
MonetDB/GIS module loaded
MonetDB/SQL module loaded
MonetDB/R module loaded
>mserver5: ../../dev/gdk/gdk_group.c:388: BATgroup_internal: Assertion `g == ((void *)0) || ((g)->T->type == 0 && (g)->T->seq != oid_nil ? 6 : (g)->T->type) == 6' failed.
mserver5.sh: line 5: 7639 Aborted /export/scratch1/dinther/INSTALL/bin/mserver5 -d10 --set embedded_r=true --set mapi_port=41000
bash-4.2$
bash-4.2$ mclient -p 41000
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.20.0 (unreleased), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>create table atacc1 ( test int, test2 int);
operation successful (8.921ms)
sql>select * from atacc1;
+------+-------+
| test | test2 |
+======+=======+
+------+-------+
0 tuples (1.444ms)
sql>alter table atacc1 add constraint atacc_test1 unique (test, test2);
sql>select * from atacc1;
bash-4.2$
### Expected Results:
No assertion failure.
This only happens on the default branch and is new (did not occur 1 week ago)
See also testweb:
http://monetdb.cwi.nl/testweb/web/showtestoutput.php?serial=55220:750c51b86faf&target=Cla-Darwin-x86_64-assert-propcheck&module=sql&test=sql%2Ftest%2Fpg_regress%2Falter_table&which=err
## Comment 20767
Date: 2015-04-09 16:17:02 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [5c2d208b7b1d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=5c2d208b7b1d) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=5c2d208b7b1d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=5c2d208b7b1d)
Changeset description:
g can be void/nil when it's empty.
This fixes bug #3694.
| Assertion failure in gdk_group.c after adding a unique constraint to a table | https://api.github.com/repos/MonetDB/MonetDB/issues/3694/comments | 0 | 2020-11-30T13:00:36Z | 2024-06-27T12:06:49Z | https://github.com/MonetDB/MonetDB/issues/3694 | 753,459,124 | 3,694 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-03-30 17:53:39 +0200
From: @swingbit
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
CC: @njnes
Last updated: 2015-05-07 12:37:24 +0200
## Comment 20749
Date: 2015-03-30 17:53:39 +0200
From: @swingbit
Created attachment 330
SQL script to reproduce the bug
I get this error when executing the query in attachment:
TypeException:user.s14_1[131]:'algebra.join' undefined in: (_210:any, r1_210:any) := algebra.join(_10:bat[:oid,:int], _209:bat[:oid,:str])
The MAL plan in the explain for the same query shows this instruction:
(X_210,r1_210) := algebra.join(X_10,X_209);
but variable X_10 was NEVER DEFINED.
I thought some of the optimizers could have performed a wrong rewrite, but the same error happens when using 'minimal_pipe'.
> Attached file: [test.sql](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3693_test.sql_330) (application/sql, 5093 bytes)
> Description: SQL script to reproduce the bug
## Comment 20750
Date: 2015-03-30 18:06:38 +0200
From: @swingbit
Additional info: I'm running Oct2014 updated to changeset 55186:2f3c782345a2, which is the latest at this point in time.
## Comment 20777
Date: 2015-04-10 15:21:01 +0200
From: @njnes
fixed a problem in project rewrites
## Comment 20778
Date: 2015-04-10 15:26:49 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [ed162e68df7f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ed162e68df7f) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ed162e68df7f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ed162e68df7f)
Changeset description:
fixed bug #3693 (fixed problem with project rewrites)
## Comment 20789
Date: 2015-04-13 15:34:20 +0200
From: @swingbit
Created attachment 332
Second test, it still fails after the first fix
> Attached file: [BUG-3693-_2.sql](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3693_BUG-3693-_2.sql_332) (application/sql, 19535 bytes)
> Description: Second test, it still fails after the first fix
## Comment 20790
Date: 2015-04-13 15:35:19 +0200
From: @swingbit
The first fix does make my first test succeed, but it still fails on my second test (see second attachment)
## Comment 20819
Date: 2015-04-19 21:29:36 +0200
From: @njnes
now even more protection is added, added also the second part of the test.
| algebra.join undefined (caused by non-existing variables in the plan) | https://api.github.com/repos/MonetDB/MonetDB/issues/3693/comments | 0 | 2020-11-30T13:00:33Z | 2024-06-27T12:06:48Z | https://github.com/MonetDB/MonetDB/issues/3693 | 753,459,101 | 3,693 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-03-27 16:19:24 +0100
From: winning
To: GDK devs <<bugs-common>>
Version: 11.19.9 (Oct2014-SP2)
CC: @mlkersten, winning
Last updated: 2015-11-11 10:34:03 +0100
## Comment 20748
Date: 2015-03-27 16:19:24 +0100
From: winning
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2272.101 Safari/537.36
Build Identifier:
I am trying to use COPY INTO to import 100,000 rows into a table with 173,866,526 rows already and it is crashing with an error:
Failed to extend the BAT, perhaps disk full
Failed to import table
I am sure my disk isn't full:
df -h
/dev/mapper/fedora--server-home 1.5T 342M 1.4T 1% /home
I am running this on a Fedora 21 virtual machine in VirtualBox with 64 GB memory allocated from a RHEL 5.9 host with 98 GB RAM total. The virtual disk size is 1.5TB. Note that I was able to load all of the data into a monetdb instance on the host (same monetdb Oct2014-SP2), but cannot now that I am trying to virtualize it.
Reproducible: Always
### Steps to Reproduce:
1.Import 17 large files into table
2.Crashes on 18th
### Actual Results:
Failed to extend the BAT, perhaps disk full
Failed to import table
### Expected Results:
Should have imported the table (plus more for a total of 58)
mserver5 --version
MonetDB 5 server v11.19.9 "Oct2014-SP2" (64-bit, 64-bit oids)
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2015 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 62.9GiB available memory, 1 available cpu core
Libraries:
libpcre: 8.35 2014-04-04 (compiled with 8.35)
openssl: OpenSSL 1.0.1k 8 Jan 2015 (compiled with )
libxml2: 2.9.1 (compiled with 2.9.1)
Compiled by: mockbuild@ (x86_64-redhat-linux-gnu)
Compilation: gcc -O3 -fomit-frame-pointer -pipe -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches -m64 -mtune=generic -D_FORTIFY_SOURCE=2
Linking : /usr/bin/ld -m elf_x86_64 -Wl,-z,relro
merovingian.log output:
2015-03-27 11:13:30 ERR agdb[7172]: = gdk_posix.c:428: MT_mremap(./bat/27/2704.tail,7fa5af344000,92405760,116523008): GDKextendf() failed
2015-03-27 11:13:30 ERR agdb[7172]: = gdk_posix.c:428: MT_mremap(./bat/27/2703.tail,7fa6848d0000,46202880,58261504): GDKextendf() failed
2015-03-27 11:13:30 ERR agdb[7172]: = gdk_posix.c:428: MT_mremap(./bat/27/2702.tail,7fa66d770000,46202880,58261504): GDKextendf() failed
2015-03-27 11:13:30 ERR agdb[7172]: = gdk_posix.c:428: MT_mremap(./bat/27/2701.tail,7fa4b4214000,92405760,116523008): GDKextendf() failed
2015-03-27 11:13:30 ERR agdb[7172]: = gdk_posix.c:428: MT_mremap(./bat/27/2700.tail,7fa67c8b0000,46202880,58261504): GDKextendf() failed
2015-03-27 11:13:30 ERR agdb[7172]: = gdk_posix.c:428: MT_mremap(./bat/26/2677.tail,7fa4ae9f4000,92405760,116523008): GDKextendf() failed
2015-03-27 11:13:30 ERR agdb[7172]: = gdk_posix.c:428: MT_mremap(./bat/26/2676.tail,7fa4a91d4000,92405760,116523008): GDKextendf() failed
2015-03-27 11:13:30 ERR agdb[7172]: = gdk_posix.c:428: MT_mremap(./bat/26/2675.tail,7fa5a6944000,92405760,116523008): GDKextendf() failed
2015-03-27 11:13:30 ERR agdb[7172]: = gdk_posix.c:428: MT_mremap(./bat/26/2674.tail,7fa666e40000,46202880,58261504): GDKextendf() failed
2015-03-27 11:13:30 ERR agdb[7172]: = gdk_posix.c:428: MT_mremap(./bat/26/2673.tail,7fa6d4aed000,11599872,14614528): GDKextendf() failed
2015-03-27 11:13:30 ERR agdb[7172]: = gdk_posix.c:428: MT_mremap(./bat/26/2672.tail,7fa6112a4000,92405760,116523008): GDKextendf() failed
2015-03-27 11:13:30 ERR agdb[7172]: = gdk_posix.c:428: MT_mremap(./bat/26/2671.tail,7fa6cf3cd000,11599872,14614528): GDKextendf() failed
2015-03-27 11:13:30 ERR agdb[7172]: = gdk_posix.c:428: MT_mremap(./bat/26/2670.tail,7fa60b904000,92405760,116523008): GDKextendf() failed
2015-03-27 11:13:30 ERR agdb[7172]: = gdk_posix.c:428: MT_mremap(./bat/26/2667.tail,7fa678cf0000,46202880,58261504): GDKextendf() failed
2015-03-27 11:13:30 ERR agdb[7172]: = gdk_posix.c:428: MT_mremap(./bat/26/2666.tail,7fa6d3e3d000,11599872,14614528): GDKextendf() failed
2015-03-27 11:13:30 ERR agdb[7172]: = gdk_posix.c:428: MT_mremap(./bat/26/2665.tail,7fa6a4550000,11599872,14614528): GDKextendf() failed
2015-03-27 11:13:30 ERR agdb[7172]: = gdk_posix.c:428: MT_mremap(./bat/27/2726.tail,7fa6953f0000,46202880,58261504): GDKextendf() failed
2015-03-27 11:13:30 ERR agdb[7172]: = gdk_posix.c:428: MT_mremap(./bat/27/2727.tail,7fa6596f0000,46202880,58261504): GDKextendf() failed
2015-03-27 11:13:30 ERR ag
## Comment 20753
Date: 2015-04-01 14:55:57 +0200
From: winning
Switched to a centos guest OS and everything works fine, as it did on the RHEL non-virtual machine. Must be something in fedora.
## Comment 21509
Date: 2015-11-11 10:34:03 +0100
From: @mlkersten
Seems resolved by the complainant.
Furthermore, COPY FROM has been changed significantly.
| COPY INTO crashes with "Failed to extend the BAT, perhaps disk full" | https://api.github.com/repos/MonetDB/MonetDB/issues/3692/comments | 0 | 2020-11-30T13:00:30Z | 2024-06-28T13:15:37Z | https://github.com/MonetDB/MonetDB/issues/3692 | 753,459,070 | 3,692 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-03-26 18:42:22 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
Last updated: 2015-05-07 12:37:32 +0200
## Comment 20747
Date: 2015-03-26 18:42:22 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Fedora; Linux x86_64; rv:36.0) Gecko/20100101 Firefox/36.0
Build Identifier:
conversion of whitespaces string to double or float is accepted without an error during insert
Reproducible: Always
### Steps to Reproduce:
1. Start mserver5 (MonetDB v11.19.9 (Oct2014-SP2))
2. Start mclient
3. execute SQL commands:
CREATE TABLE DOUBLE_TBL(f1 double);
INSERT INTO DOUBLE_TBL(f1) VALUES (' ');
INSERT INTO DOUBLE_TBL(f1) VALUES ('\t');
INSERT INTO DOUBLE_TBL(f1) VALUES ('\n');
INSERT INTO DOUBLE_TBL(f1) VALUES (' \t \t\t \n ');
SELECT f1 FROM DOUBLE_TBL;
-- lists 4 rows with all null
INSERT INTO DOUBLE_TBL(f1) VALUES ('');
-- this fails
CREATE TABLE FLOAT_TBL(f1 float);
INSERT INTO FLOAT_TBL(f1) VALUES (' ');
INSERT INTO FLOAT_TBL(f1) VALUES ('\t');
INSERT INTO FLOAT_TBL(f1) VALUES ('\n');
INSERT INTO FLOAT_TBL(f1) VALUES (' \t \t\t \n ');
SELECT f1 FROM FLOAT_TBL;
-- lists 4 rows with all null
INSERT INTO FLOAT_TBL(f1) VALUES ('');
-- this fails
### Actual Results:
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.19.9 (Oct2014-SP2), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>CREATE TABLE DOUBLE_TBL(f1 double);
operation successful (7.731ms)
sql>INSERT INTO DOUBLE_TBL(f1) VALUES (' ');
1 affected row (0.542ms)
sql>INSERT INTO DOUBLE_TBL(f1) VALUES ('\t');
1 affected row (0.308ms)
sql>INSERT INTO DOUBLE_TBL(f1) VALUES ('\n');
1 affected row (0.324ms)
sql>INSERT INTO DOUBLE_TBL(f1) VALUES (' \t \t\t \n ');
1 affected row (0.660ms)
sql>SELECT f1 FROM DOUBLE_TBL;
+--------------------------+
| f1 |
+==========================+
| null |
| null |
| null |
| null |
+--------------------------+
4 tuples (1.040ms)
sql>INSERT INTO DOUBLE_TBL(f1) VALUES ('');
conversion of string '' to type dbl failed.
sql>
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.19.9 (Oct2014-SP2), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>CREATE TABLE FLOAT_TBL(f1 float);
operation successful (1.506ms)
sql>INSERT INTO FLOAT_TBL(f1) VALUES (' ');
1 affected row (0.777ms)
sql>INSERT INTO FLOAT_TBL(f1) VALUES ('\t');
1 affected row (0.531ms)
sql>INSERT INTO FLOAT_TBL(f1) VALUES ('\n');
1 affected row (0.509ms)
sql>INSERT INTO FLOAT_TBL(f1) VALUES (' \t \t\t \n ');
1 affected row (0.745ms)
sql>SELECT f1 FROM FLOAT_TBL;
+--------------------------+
| f1 |
+==========================+
| null |
| null |
| null |
| null |
+--------------------------+
4 tuples (1.731ms)
sql>INSERT INTO FLOAT_TBL(f1) VALUES ('');
conversion of string '' to type dbl failed.
sql>
### Expected Results:
strings with only whitespace characters do not represent a number and therefore should result in an error, like when the string has length 0.
In default branch the behavior for whitespace string to float has recently changed and it now returns an error.
## Comment 20751
Date: 2015-03-31 11:31:51 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [476a699b2b1e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=476a699b2b1e) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=476a699b2b1e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=476a699b2b1e)
Changeset description:
Fix for bug #3691 by using basically same code for dbl as for flt.
## Comment 20754
Date: 2015-04-02 13:31:30 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [c64098b19544](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=c64098b19544) made by Martin van Dinther <martin.van.dinther@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=c64098b19544](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=c64098b19544)
Changeset description:
Updated and approved new outputs after fix for bug #3691
## Comment 20755
Date: 2015-04-02 13:31:38 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [543a90055d48](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=543a90055d48) made by Martin van Dinther <martin.van.dinther@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=543a90055d48](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=543a90055d48)
Changeset description:
Updated float8.sql outputs after fix 3691
## Comment 20763
Date: 2015-04-08 14:14:35 +0200
From: @sjoerdmullender
It's NEXTRELEASE until the release.
| conversion of whitespaces string to double or float is accepted without an error during insert | https://api.github.com/repos/MonetDB/MonetDB/issues/3691/comments | 0 | 2020-11-30T13:00:28Z | 2024-06-27T12:06:46Z | https://github.com/MonetDB/MonetDB/issues/3691 | 753,459,049 | 3,691 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-03-26 12:57:47 +0100
From: @swingbit
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
CC: @njnes
Last updated: 2015-05-07 12:37:46 +0200
## Comment 20741
Date: 2015-03-26 12:57:47 +0100
From: @swingbit
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2272.76 Safari/537.36
Build Identifier:
This assertions happens in the reorder_joins optimizer, see gdb trace below.
The query it fails with is a rather large one, with many subqueries.
I attach a reproducible test that fails.
0 0x0000003633c348d7 in raise () from /lib64/libc.so.6
1 0x0000003633c3653a in abort () from /lib64/libc.so.6
2 0x0000003633c2d47d in __assert_fail_base () from /lib64/libc.so.6
3 0x0000003633c2d532 in __assert_fail () from /lib64/libc.so.6
4 0x00007f05666602ed in find_fk (sql=0x7f05580d18d0, rels=0x7f0558363920, exps=0x7f055835c5b0) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:722
5 0x00007f05666605a9 in order_joins (sql=0x7f05580d18d0, rels=0x7f0558363920, exps=0x7f055835c5b0) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:762
6 0x00007f0566661147 in reorder_join (sql=0x7f05580d18d0, rel=0x7f055835c730) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:991
7 0x00007f056666136d in rel_join_order (changes=0x7f05654c7604, sql=0x7f05580d18d0, rel=0x7f055835c730) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:1043
8 0x00007f056667b2dd in rewrite (sql=0x7f05580d18d0, rel=0x7f055835c730, rewriter=0x7f05666612cc <rel_join_order>, has_changes=0x7f05654c7948) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:7014
9 0x00007f056667b22f in rewrite (sql=0x7f05580d18d0, rel=0x7f05582a5a50, rewriter=0x7f05666612cc <rel_join_order>, has_changes=0x7f05654c7948) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:7001
10 0x00007f056667b22f in rewrite (sql=0x7f05580d18d0, rel=0x7f05582a8e70, rewriter=0x7f05666612cc <rel_join_order>, has_changes=0x7f05654c7948) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:7001
11 0x00007f056667b203 in rewrite (sql=0x7f05580d18d0, rel=0x7f055834e060, rewriter=0x7f05666612cc <rel_join_order>, has_changes=0x7f05654c7948) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:6994
12 0x00007f056667b22f in rewrite (sql=0x7f05580d18d0, rel=0x7f05582ac400, rewriter=0x7f05666612cc <rel_join_order>, has_changes=0x7f05654c7948) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:7001
13 0x00007f056667b22f in rewrite (sql=0x7f05580d18d0, rel=0x7f05582b28d0, rewriter=0x7f05666612cc <rel_join_order>, has_changes=0x7f05654c7948) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:7001
14 0x00007f056667b203 in rewrite (sql=0x7f05580d18d0, rel=0x7f055835cf00, rewriter=0x7f05666612cc <rel_join_order>, has_changes=0x7f05654c7948) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:6994
15 0x00007f056667b1dc in rewrite (sql=0x7f05580d18d0, rel=0x7f055835e100, rewriter=0x7f05666612cc <rel_join_order>, has_changes=0x7f05654c7948) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:6993
16 0x00007f056667b22f in rewrite (sql=0x7f05580d18d0, rel=0x7f05582b8a30, rewriter=0x7f05666612cc <rel_join_order>, has_changes=0x7f05654c7948) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:7001
17 0x00007f056667b1dc in rewrite (sql=0x7f05580d18d0, rel=0x7f0558353d70, rewriter=0x7f05666612cc <rel_join_order>, has_changes=0x7f05654c7948) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:6993
18 0x00007f056667b22f in rewrite (sql=0x7f05580d18d0, rel=0x7f05582f8950, rewriter=0x7f05666612cc <rel_join_order>, has_changes=0x7f05654c7948) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:7001
19 0x00007f056667b22f in rewrite (sql=0x7f05580d18d0, rel=0x7f05582f8a80, rewriter=0x7f05666612cc <rel_join_order>, has_changes=0x7f05654c7948) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:7001
20 0x00007f056667b22f in rewrite (sql=0x7f05580d18d0, rel=0x7f05582f8c00, rewriter=0x7f05666612cc <rel_join_order>, has_changes=0x7f05654c7948) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:7001
21 0x00007f056667baa5 in _rel_optimizer (sql=0x7f05580d18d0, rel=0x7f05582f8c00, level=1) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:7185
22 0x00007f056667bd12 in _rel_optimizer (sql=0x7f05580d18d0, rel=0x7f05582f8c00, level=1) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:7228
23 0x00007f056667bd65 in rel_optimizer (sql=0x7f05580d18d0, rel=0x7f05582f8c00) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/server/rel_optimizer.c:7237
24 0x00007f05665739cb in sql_symbol2relation (c=0x7f05580d18d0, sym=0x7f0558234840) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/backends/monet5/sql.c:129
25 0x00007f056659deee in SQLparser (c=0x6dc480) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/sql/backends/monet5/sql_scenario.c:2488
26 0x00007f056e7e55cd in runPhase (c=0x6dc480, phase=1) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/monetdb5/mal/mal_scenario.c:526
27 0x00007f056e7e5768 in runScenarioBody (c=0x6dc480) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/monetdb5/mal/mal_scenario.c:561
28 0x00007f056e7e590b in runScenario (c=0x6dc480) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/monetdb5/mal/mal_scenario.c:590
29 0x00007f056e7e7242 in MSserveClient (dummy=0x6dc480) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/monetdb5/mal/mal_session.c:450
30 0x00007f056e7e6c3e in MSscheduleClient (command=0x7f05580eca10 "", challenge=0x7f05654c7df0 "6RGoNPfpd", fin=0x7f0558100790, fout=0x7f0558105820) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/monetdb5/mal/mal_session.c:330
31 0x00007f056e8d5596 in doChallenge (data=0x7f05600008d0) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/monetdb5/modules/mal/mal_mapi.c:192
32 0x00007f056e27a976 in thread_starter (arg=0x7f0560000a50) at /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/gdk/gdk_system.c:535
33 0x000000363400752a in start_thread () from /lib64/libpthread.so.0
34 0x0000003633d0022d in clone () from /lib64/libc.so.6
Reproducible: Always
### Steps to Reproduce:
1. run the attached SQL script
2.
3.
### Actual Results:
find_fk: Assertion `t && i' failed.
$ mserver5 --version
MonetDB 5 server v11.19.10 (64-bit, 64-bit oids)
This is an unreleased version
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2015 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 15.6GiB available memory, 8 available cpu cores
Libraries:
libpcre: 8.35 2014-04-04 (compiled with 8.35)
openssl: OpenSSL 1.0.1k 8 Jan 2015 (compiled with OpenSSL 1.0.1k-fips 8 Jan 2015)
libxml2: 2.9.1 (compiled with 2.9.1)
Compiled by: roberto@photon.spinque.com (x86_64-unknown-linux-gnu)
Compilation: gcc -g -Werror -Wall -Wextra -W -Werror-implicit-function-declaration -Wpointer-arith -Wdeclaration-after-statement -Wundef -Wformat=2 -Wno-format-nonliteral -Winit-self -Winvalid-pch -Wmissing-declarations -Wmissing-format-attribute -Wmissing-prototypes -Wold-style-definition -Wpacked -Wunknown-pragmas -Wvariadic-macros -fstack-protector-all -Wstack-protector -Wpacked-bitfield-compat -Wsync-nand -Wjump-misses-init -Wmissing-include-dirs -Wlogical-op -Wunreachable-code
Linking : /usr/bin/ld -m elf_x86_64
## Comment 20742
Date: 2015-03-26 12:59:12 +0100
From: @swingbit
Created attachment 326
SQL script to reproduce the bug
> Attached file: [BUG-3690.sql](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3690_BUG-3690.sql_326) (application/sql, 9055 bytes)
> Description: SQL script to reproduce the bug
## Comment 20772
Date: 2015-04-10 11:04:47 +0200
From: @njnes
Fixed added to stable. Added protection agains missing OID columns.
## Comment 20773
Date: 2015-04-10 11:42:07 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [2e15db9d79b4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=2e15db9d79b4) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=2e15db9d79b4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=2e15db9d79b4)
Changeset description:
fixes for Bug #3690, ie protect against missing OID columns
| find_fk: Assertion `t && i' failed. | https://api.github.com/repos/MonetDB/MonetDB/issues/3690/comments | 0 | 2020-11-30T13:00:25Z | 2024-06-27T12:06:45Z | https://github.com/MonetDB/MonetDB/issues/3690 | 753,459,007 | 3,690 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-03-19 16:45:39 +0100
From: Richard Hughes <<richard.monetdb>>
To: Merovingian devs <<bugs-merovingian>>
Version: 11.19.9 (Oct2014-SP2)
Last updated: 2015-05-07 12:37:56 +0200
## Comment 20725
Date: 2015-03-19 16:45:39 +0100
From: Richard Hughes <<richard.monetdb>>
Build is Oct2014 [e58372859532](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e58372859532)
To reproduce:
1) Run "nc 127.0.0.1 50000" in one terminal. Ignore the negotiation/authentication blurb that's printed.
2) Run "mclient -h 127.0.0.1 -d mydb" in another terminal. Nothing will appear to happen.
3) Ctrl-C the netcat in the first terminal. The second terminal will unstick itself.
The second terminal will never unstick until nc is Ctrl-Ced.
While this could be considered an exploitable unauthenticated DoS, anybody running a database on an untrusted network is an idiot. I'm creating this bug because I've managed to trip over this on several occasions simply by normal network glitches and/or clients going nuts, and it's inconvenient to have to restart the whole database simply to unstick it.
## Comment 20729
Date: 2015-03-20 10:21:14 +0100
From: @sjoerdmullender
It looks like this is actually two bugs. One in mserver5 and one in monetdbd.
My first attempt to reproduce worked (in the sense that the bug showed up), but the problem was in monetdbd. It was waiting for a response from nc to continue the authentication process. mserver5 was blissfully unaware of any attempt at connecting.
My second attempt was to connect directly to mserver5, and now it was waiting for an authentication response.
## Comment 20730
Date: 2015-03-20 11:13:29 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [312dc4e64454](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=312dc4e64454) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=312dc4e64454](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=312dc4e64454)
Changeset description:
In monetdbd, handle client authentication in separate thread.
This is a partial fix to bug #3689.
## Comment 20731
Date: 2015-03-20 13:22:47 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [8acd025b3ee1](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=8acd025b3ee1) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=8acd025b3ee1](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=8acd025b3ee1)
Changeset description:
Run authentication exchange in separate thread.
This should fix bug #3689.
## Comment 20752
Date: 2015-03-31 11:35:01 +0200
From: @sjoerdmullender
Since I could reproduce the problem before my fixes and I can't after, I'm assuming this is indeed fixed.
| No more connections accepted if a single client misbehaves | https://api.github.com/repos/MonetDB/MonetDB/issues/3689/comments | 0 | 2020-11-30T13:00:23Z | 2024-06-27T12:06:44Z | https://github.com/MonetDB/MonetDB/issues/3689 | 753,458,984 | 3,689 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-03-19 13:58:17 +0100
From: Richard Hughes <<richard.monetdb>>
To: GDK devs <<bugs-common>>
Version: 11.19.9 (Oct2014-SP2)
Last updated: 2015-05-07 12:37:45 +0200
## Comment 20722
Date: 2015-03-19 13:58:17 +0100
From: Richard Hughes <<richard.monetdb>>
Build is Oct2014 [e58372859532](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e58372859532)
I got this crash at shutdown:
(gdb) bt
0 0x00007f4dd2eff44a in pthread_join (threadid=4294967296,
thread_return=thread_return@entry=0x0) at pthread_join.c:47
1 0x00007f4dd4084c36 in MT_join_thread (t=<optimized out>)
at gdk_system.c:656
2 0x00007f4dd3ff9014 in GDKexit (status=status@entry=0) at gdk_utils.c:1230
3 0x00007f4dd45122dc in mal_exit () at mal.c:215
4 <signal handler called>
5 0x00007f4dd2c2bf23 in select () at ../sysdeps/unix/syscall-template.S:81
6 0x00007f4dd4085c09 in MT_sleep_ms (ms=<optimized out>) at gdk_posix.c:1096
7 0x00000000004026da in main (argc=24, av=0x0) at mserver5.c:656
Note the absurd value for threadid.
Dumping out the memory around GDKvmtrim_id (which is the global variable that absurd value was derived from):
(gdb) x/10a &GDKvmtrim_id-4
0x7f4dd44b1c28 <THRerrorcount+40>: 0x0 0x0
0x7f4dd44b1c38 <THRerrorcount+56>: 0x1 0x100000000
0x7f4dd44b1c48 <GDKvmtrim_id>: 0x100000001 0xffffff8a2e450000
0x7f4dd44b1c58 <GDK_mallocedbytes_estimate>: 0x11546c6f9 0xe31863
0x7f4dd44b1c68 <GDKstopped>: 0x100000001 0x1
The first thing I looked at was the purpose of THRerrorcount and whether it could be overrun and, sure enough, it looks very suspicious to me:
gdk_utils.c:
doGDKaddbuf(const char *prefix, const char *message, size_t messagelen, const char *suffix)
...
THRerrorcount[THRgettid()]++;
Can I ask somebody to eyeball this and see if you agree with my analysis? I'm running a 24 core machine, so an overrun of a 16 element array is very doable. I think this may also be why I've sometimes had mserver5 completely stop doing anything and, upon attaching a debugger, finding that most of the threads have mysteriously disappeared - overwriting GDKstopped would cause this to happen.
FYI, nobody ever reads THRerrorcount so the obvious fix is to remove it entirely.
## Comment 20732
Date: 2015-03-20 16:26:38 +0100
From: @sjoerdmullender
It does indeed look very suspicious. I'll remove this THRerrorcount completely.
## Comment 20733
Date: 2015-03-20 16:27:54 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [346077203679](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=346077203679) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=346077203679](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=346077203679)
Changeset description:
Remove THRerrorcount. It serves no useful purpose that I can see.
And this may well fix bug #3688.
## Comment 20735
Date: 2015-03-20 16:54:37 +0100
From: Richard Hughes <<richard.monetdb>>
Thanks.
BTW, while you were fixing it, I was confirming my theory by adding a printf to doGDKaddbuf which displays the value of THRgettid(). It is indeed easy to provoke a simple overrun of that array by causing a division by zero error:
create table bar as select cast(0 as int) as value with data;
select 1/0 from bar;
The highest value I've seen for gdk_nr_threads=24 is 26. This bug will definitely explain a load of other random glitches I've seen but never told you about (because I've been unable to reproduce).
| Crash at exit (overrun THRerrorcount?) | https://api.github.com/repos/MonetDB/MonetDB/issues/3688/comments | 0 | 2020-11-30T13:00:20Z | 2024-06-27T12:06:43Z | https://github.com/MonetDB/MonetDB/issues/3688 | 753,458,958 | 3,688 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-03-13 12:30:10 +0100
From: Richard Hughes <<richard.monetdb>>
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
CC: @njnes
Last updated: 2019-06-06 13:17:07 +0200
## Comment 20713
Date: 2015-03-13 12:30:10 +0100
From: Richard Hughes <<richard.monetdb>>
Build is Oct2014 [e58372859532](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e58372859532)
sql>set optimizer='minimal_pipe';
operation successful (0.837ms)
sql>create table foo (a int);
operation successful (1.389ms)
sql>insert into foo values (1),(2),(3);
3 affected rows (0.771ms)
sql>select 100*v from (select sum(1) as v from foo group by a) as t;
TypeException:user.s5_1[19]:'bat.insert' undefined in: _62:any := bat.insert(_52
:bat[:oid,:sht], _60:oid, _59:lng)
program contains errors
sql>explain select 100*v from (select sum(1) as v from foo group by a) as t;
TypeException:user.s6_1[19]:'bat.insert' undefined in: _62:any := bat.insert(_52
:bat[:oid,:sht], _60:oid, _59:lng)
+-----------------------------------------------------------------------------+
| mal |
+=============================================================================+
| function user.s6_1{autoCommit=true}(A0:lng,A1:bte):void; |
| X_4 := sql.mvc(); |
| X_5 := A0; |
| X_6 := calc.sht(X_5); |
| X_7:bat[:oid,:oid] := sql.tid(X_4,"sys","foo"); |
| X_10:bat[:oid,:int] := sql.bind(X_4,"sys","foo","a",0); |
| (X_13,r1_13) := sql.bind(X_4,"sys","foo","a",2); |
| X_16:bat[:oid,:int] := sql.bind(X_4,"sys","foo","a",1); |
| X_18 := sql.delta(X_10,X_13,r1_13,X_16); |
| X_19 := algebra.leftfetchjoin(X_7,X_18); |
| X_20 := A1; |
| X_21 := algebra.project(X_19,X_20); |
| (X_22,r1_22,r2_22) := group.subgroupdone(X_19); |
| X_25:bat[:oid,:lng] := aggr.subsum(X_21,X_22,r1_22,true,true); |
| X_50 := bat.reverse(X_25); |
| X_52 := bat.new(nil:oid,nil:sht); |
| barrier (X_56,X_57) := iterator.new(X_25); |
| X_59 := calc.*(X_6,X_57); |
| X_60 := algebra.fetch(X_50,X_56); |
| bat.insert(X_52,X_60,X_59); |
| redo (X_56,X_57) := iterator.next(X_25); |
| exit (X_56,X_57); |
| X_26:bat[:oid,:sht] := X_52; |
| X_29 := sql.resultSet(1,1,X_26); |
| sql.rsColumn(X_29,".L1","sql_mul_single_value","smallint",16,0,X_26); |
| X_35 := io.stdout(); |
| sql.exportResult(X_35,X_29); |
| end s6_1; |
| querylog.define("explain select 100*v from (select sum(1) as v from foo g |
: roup by a) as t;","minimal_pipe") :
+-----------------------------------------------------------------------------+
29 tuples (2.504ms)
That's as reduced as I can get the query while still reproducing it. I haven't done any further investigation because I found a different way of writing the original query.
## Comment 20718
Date: 2015-03-18 15:16:57 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [aa371f63fb0c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=aa371f63fb0c) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=aa371f63fb0c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=aa371f63fb0c)
Changeset description:
fixed bug #3687, ie use correct result type for aggregates
## Comment 20719
Date: 2015-03-18 15:17:55 +0100
From: @njnes
fixed in stable, ie only reset the result type of aggregate functions of special types (like decimal and timestamps)
## Comment 20736
Date: 2015-03-21 17:04:32 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [66fe85a5c3f2](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=66fe85a5c3f2) made by Stefan Manegold <Stefan.Manegold@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=66fe85a5c3f2](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=66fe85a5c3f2)
Changeset description:
incorrect_result_type.Bug-3687: approving 128-bit result:
aggregations (here: sum()) always yield largest available result type
in order to "avoid" (actually: postpone) overflow.
## Comment 27026
Date: 2019-06-06 13:17:07 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [715c9a5ee107](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=715c9a5ee107) made by Martin van Dinther <martin.van.dinther@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=715c9a5ee107](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=715c9a5ee107)
Changeset description:
Eliminated *.int128 output files (by adding cast(... as bigint))
Also extended test: incorrect_result_type.Bug-3687
| 'bat.insert' undefined | https://api.github.com/repos/MonetDB/MonetDB/issues/3687/comments | 0 | 2020-11-30T13:00:18Z | 2024-06-27T12:06:42Z | https://github.com/MonetDB/MonetDB/issues/3687 | 753,458,922 | 3,687 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-03-13 12:16:22 +0100
From: Richard Hughes <<richard.monetdb>>
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
CC: martin.van.dinther, @njnes
Last updated: 2015-08-28 13:42:05 +0200
## Comment 20712
Date: 2015-03-13 12:16:22 +0100
From: Richard Hughes <<richard.monetdb>>
Build is Oct2014 [e58372859532](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e58372859532)
sql>select 12*5/3;
+----------------------+
| sql_mul_single_value |
+======================+
| 12 |
+----------------------+
Expected result is 20.
sql>plan select 12*5/3;
+-----------------------------------------------------------------------------+
| rel |
+=============================================================================+
| [ sys.sql_mul(smallint[tinyint "12"], sys.sql_div(tinyint "5", tinyint "3" |
: )) ] :
+-----------------------------------------------------------------------------+
"select (12*5)/3;" works.
## Comment 20723
Date: 2015-03-19 14:22:33 +0100
From: Martin van Dinther <<martin.van.dinther>>
Also the result of the following is incorrect;
sql>select 12*3/5;
+----------------------+
| sql_mul_single_value |
+======================+
| 0 |
+----------------------+
1 tuple (0.794ms)
expected result: 7.2 such as in:
sql>select 12*3/5.0;
+----------------------+
| sql_mul_single_value |
+======================+
| 7.200 |
+----------------------+
1 tuple (0.875ms)
sql>plan select 12*3/5;
+----------------------------------------------------------------------------------+
| rel |
+==================================================================================+
| [ sys.sql_mul(smallint[tinyint "12"], sys.sql_div(tinyint "3", tinyint "5")) ] |
+----------------------------------------------------------------------------------+
1 tuple (0.466ms)
sql>plan select 12*3/5.0;
+------------------------------------------------------------------------------------------------------------------+
| rel |
+==================================================================================================================+
| [ sys.sql_mul(decimal(15)[tinyint "12"], sys.sql_div(decimal(7,4)[decimal(4,1)[tinyint "3"]], decimal(4,1)[decimal(2,1) "50"])) ] :
+------------------------------------------------------------------------------------------------------------------+
1 tuple (0.559ms)
## Comment 20724
Date: 2015-03-19 16:11:11 +0100
From: @sjoerdmullender
I had a fix ready, but it had other problems, so I haven't committed it.
The problem is one of priorities, not associativity. In the SQL parser we have different priorities for / and *, but they should have the same priority according to the SQL standard. Fixing this uncovered another problem that we should fix as well.
## Comment 20728
Date: 2015-03-20 09:32:24 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [ba9b1cd64364](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ba9b1cd64364) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ba9b1cd64364](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ba9b1cd64364)
Changeset description:
Added test for bug #3686.
## Comment 20872
Date: 2015-05-20 18:31:48 +0200
From: @njnes
fixed in default
## Comment 21201
Date: 2015-08-28 13:42:05 +0200
From: @sjoerdmullender
Jul2015 has been released.
| Wrong associativity of multiply/divide | https://api.github.com/repos/MonetDB/MonetDB/issues/3686/comments | 0 | 2020-11-30T13:00:15Z | 2024-06-27T12:06:41Z | https://github.com/MonetDB/MonetDB/issues/3686 | 753,458,899 | 3,686 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-03-12 19:25:04 +0100
From: Antonio <<antonioknight>>
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
CC: @njnes
Last updated: 2015-03-20 16:44:02 +0100
## Comment 20711
Date: 2015-03-12 19:25:04 +0100
From: Antonio <<antonioknight>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/40.0.2214.115 Safari/537.36
Build Identifier:
This query executed over JDBC driver results into a "Connection to server lost! (mserver still alive?" exception.
select
lower("store_region") as "col3"
from "lineitem_denormalized_first1k"
order by lower("col3") asc
limit 83334 offset 0;
BUT if I just change the ORDER BY condition removing the "lower" function, then all works fine.
Is "lower" function allowed within the ORDER BY clause?
Regards,
Reproducible: Always
### Steps to Reproduce:
select
lower("store_region") as "col3"
from "lineitem_denormalized_first1k"
order by lower("col3") asc
limit 83334 offset 0;
### Actual Results:
SqlException in JDBC driver
This variation works flawlessly
select
lower("store_region") as "col3"
from "lineitem_denormalized_first1k"
order by "col3" asc
limit 83334 offset 0;
## Comment 20721
Date: 2015-03-18 17:38:20 +0100
From: @njnes
I'm unable to reproduce this on the current stable branch. We only allow functions on inner relations, not on the results in the selection list. So
order by lower("store_region") should work.
## Comment 20726
Date: 2015-03-19 19:10:20 +0100
From: Antonio <<antonioknight>>
Forgot to add, that the Windows mserver5 crashes when I execute the query.
Which stable version are you referring to? I am using MonetDB 5 server v11.19.7 "Oct2014-SP1"
## Comment 20727
Date: 2015-03-19 19:10:47 +0100
From: Antonio <<antonioknight>>
(In reply to comment 1)
> I'm unable to reproduce this on the current stable branch. We only allow
> functions on inner relations, not on the results in the selection list. So
> order by lower("store_region") should work.
Forgot to add, that the Windows mserver5 crashes when I execute the query.
Which stable version are you referring to? I am using MonetDB 5 server v11.19.7 "Oct2014-SP1"
## Comment 20734
Date: 2015-03-20 16:44:02 +0100
From: @sjoerdmullender
I can reproduce this on Oct2014-SP1 but not on Oct2014-SP2.
On SP2 I get:
SELECT: identifier 'col3' unknown
So, closing as fixed.
| JDBC connection error (Connection to server lost! (mserver still alive?)) on executing a particular query | https://api.github.com/repos/MonetDB/MonetDB/issues/3685/comments | 0 | 2020-11-30T13:00:13Z | 2024-06-27T12:06:40Z | https://github.com/MonetDB/MonetDB/issues/3685 | 753,458,873 | 3,685 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-03-12 15:27:06 +0100
From: Antonio <<antonioknight>>
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
CC: @njnes
Last updated: 2015-05-07 12:37:52 +0200
## Comment 20710
Date: 2015-03-12 15:27:06 +0100
From: Antonio <<antonioknight>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/40.0.2214.115 Safari/537.36
Build Identifier:
Executing a query that has WHERE condition with NOT IN clause and specifying LIMIT and OFFSET results in different (wrong) result sets compared to:
1) removing the LIMIT condition
2) changing the NOT IN clause into a <> operator
This is an example of the query that fails:
select
"store_id" as "col0" , *
from "lineitem_denormalized_first1k"
where
(
"yyyymmdd" <= 20101213
or "customer_date_of_birth" < date '1977-03-19'
or
(
("customer_state" <> 'mardin')
and
(
"customer_country" = 'usa'
or "customer_country" = 'sweden'
)
and not "year_begin_date" = '2010-01-01'
)
)
order by "col0" asc limit 10000;
that returns 1120 rows (instead of 824 expected records) with part of the result set duplicated.
BUT if I remove the LIMIT 10000 clause, than it correctly returns 824 records.
OR I can achieve the same fix by changing the "not "year_begin_date" = '2010-01-01'" into a "year_begin_date" <> '2010-01-01' and keeping the LIMIT clause, and the result set correctly has 824 records.
I tested the same table on MySql executing the same query and it works without problem.
Is this a know issue?
Reproducible: Always
### Steps to Reproduce:
Use WHERE NOT Field IN ('value') in combination with ORDER BY and LIMIT clause
### Actual Results:
Duplicated part of result set
### Expected Results:
non duplicated rows in result set
## Comment 20739
Date: 2015-03-23 16:25:18 +0100
From: Antonio <<antonioknight>>
Another similar case with this query highlights the problem:
select
Line_Margin
from lineitem_denormalized_first1k
where
(
(
Line_Cost is not null
and
Line_Cost = 4.56
)
or Store_Id <> 35
or
(
Product_Name like ('%' || 'benches' || '%') escape '!'
and
(
Month_Begin_Date is null or
Month_Begin_Date <> date '2010-12-01'
)
)
)
order by Line_Margin limit 68 offset 319
It wrongly returns 0 records, when it should return 68 records.
To solve the issue I have 4 different options:
1) remove the ORDER BY clause
or
2) remove the LIMIT clause
or
3) remove the OFFSET clause
or
4) remove the "is null" from the WHERE conditions.
This seems to highlight the bug where I cannot use ORDER BY + LIMIT + OFFSET clauses in combination when the WHERE conditions have "IS NULL" or "NOT" expressions.
## Comment 20740
Date: 2015-03-25 19:35:24 +0100
From: @njnes
could you also supply the ddl statements for the tables used in your example query?
## Comment 20743
Date: 2015-03-26 16:00:42 +0100
From: Antonio <<antonioknight>>
Created attachment 327
DDL create table
> Attached file: [create first1k table on MonetDb.txt](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3684_create_first1k_table_on_MonetDb.txt_327) (text/plain, 1238 bytes)
> Description: DDL create table
## Comment 20744
Date: 2015-03-26 16:01:58 +0100
From: Antonio <<antonioknight>>
Created attachment 328
ddl create table
> Attached file: [create first1k table on MonetDb.txt](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3684_create_first1k_table_on_MonetDb.txt_328) (text/plain, 1238 bytes)
> Description: ddl create table
## Comment 20745
Date: 2015-03-26 16:19:51 +0100
From: Antonio <<antonioknight>>
Created attachment 329
csv file with data
Used COPY INTO command to import data into table.
> Attached file: [lineitem_denormalized_first1k_headerless.csv](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3684_lineitem_denormalized_first1k_headerless.csv_329) (application/vnd.ms-excel, 319404 bytes)
> Description: csv file with data
## Comment 20746
Date: 2015-03-26 16:21:46 +0100
From: Antonio <<antonioknight>>
(In reply to comment 5)
> Created attachment 329 [details]
> csv file with data
>
> Used COPY INTO command to import data into table.
For convenience, this is the exact command to import data into the table
mclient -d visokio -s "COPY INTO lineitem_denormalized_first1k FROM '/home/antonio/lineitem_denormalized_headerless.csv' USING DELIMITERS ',','\\n','\"'"
## Comment 20774
Date: 2015-04-10 13:27:07 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [ca7a2dd7ec6c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ca7a2dd7ec6c) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ca7a2dd7ec6c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ca7a2dd7ec6c)
Changeset description:
fixed bug #3684, properly pushdown topn's including offsets
## Comment 20775
Date: 2015-04-10 13:27:43 +0200
From: @njnes
fixed by properly pushing down topn's.
| Wrong query result set WHERE "IS NULL" or "NOT IN" clauses uses in combination with ORDER, LIMIT and OFFSET | https://api.github.com/repos/MonetDB/MonetDB/issues/3684/comments | 0 | 2020-11-30T13:00:10Z | 2024-06-27T12:06:39Z | https://github.com/MonetDB/MonetDB/issues/3684 | 753,458,839 | 3,684 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-03-11 15:15:40 +0100
From: @hannesmuehleisen
To: buildtools devs <<bugs-buildtools>>
Version: 11.21.19 (Jul2015-SP4)
Last updated: 2016-04-11 11:22:10 +0200
## Comment 20705
Date: 2015-03-11 15:15:40 +0100
From: @hannesmuehleisen
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2272.76 Safari/537.36
Build Identifier:
If MonetDB is configured with --enable-console=no, Mtest.py hangs on startup (waiting for a prompt from the server). Suggest to remove server console altogether.
Reproducible: Always
| Mtest hangs if server console was disabled | https://api.github.com/repos/MonetDB/MonetDB/issues/3683/comments | 0 | 2020-11-30T13:00:08Z | 2024-06-28T13:15:36Z | https://github.com/MonetDB/MonetDB/issues/3683 | 753,458,806 | 3,683 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-03-11 11:10:59 +0100
From: Pete Hollobon <<pete>>
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.19.9 (Oct2014-SP2)
CC: @mlkersten, pete
Last updated: 2015-11-25 22:18:36 +0100
## Comment 20704
Date: 2015-03-11 11:10:59 +0100
From: Pete Hollobon <<pete>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/40.0.2214.115 Safari/537.36
Build Identifier:
After trying out the query profiler (querylog_enable() / a few queries on querylog_catalog and querylog_calls, then querylog_disable()), MonetDB failed to restart with an error:
!FATAL: BBPinit: duplicate entry in BBP.dir.
Looking at bat/BACKUP/BBP.dir, I found two lines with identical bat cache IDs, apparently for the querylog_calls.id column:
8 32 querylog_calls_id tmpr_10 10 967660 0 2551 0 0 2551 8192 0 0 0 0 void 0 1 1281 0 0 0 0 0 1314840 0 0 0 oid 8 0 1024 0 1 6 3 1314805 1923200 2040
8 65536 2
8 32 querylog_calls_id tmpr_10 10 967660 0 1362 0 0 1363 8192 0 0 0 0 void 0 1 1281 0 0 0 0 0 1314840 0 0 0 oid 8 0 1024 0 1 6 3 1314805 1507983 1090
4 65536 2
The lines were not adjacent.
In order to get the server to start, I deleted both lines.
The bug occurred on our production server, which was having data loaded (with COPY commands) and running large queries at the time I was using the querylog functionality.
I have not been able to reproduce this on a test server.
Reproducible: Couldn't Reproduce
## Comment 21594
Date: 2015-11-25 22:18:36 +0100
From: @mlkersten
This release branch is closed and insuffient data is available to reproduce it.
| Duplicates in BBP.dir | https://api.github.com/repos/MonetDB/MonetDB/issues/3682/comments | 0 | 2020-11-30T13:00:05Z | 2024-06-28T13:15:36Z | https://github.com/MonetDB/MonetDB/issues/3682 | 753,458,772 | 3,682 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-03-11 09:21:05 +0100
From: @dnedev
To: GDK devs <<bugs-common>>
Version: 11.21.19 (Jul2015-SP4)
Last updated: 2016-04-11 11:22:58 +0200
## Comment 20701
Date: 2015-03-11 09:21:05 +0100
From: @dnedev
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_3) AppleWebKit/600.5.6 (KHTML, like Gecko) Version/8.0.5 Safari/600.5.6
Build Identifier:
Using msqldump with a very large dataset (1.2 TB) results in a Segmentation fault.
The last few logged statements are:
[ 7477, "start", "15:01:17.027948", 3, 0, "sql.exportResult(X_96==\"104d2\":streams,X_61=17);", ]
[ 7478, "done" , "15:39:22.033991", 3, 2285006039, "sql.exportResult(X_96==\"104d2\":streams,X_61=17);", ]
[ 7479, "start", "15:39:22.034192", 3, 0, "end s16_1;", ]
[ 7480, "done" , "15:39:22.034311", 3, 117, "end s16_1;", ]
[ 7481, "done" , "15:39:22.034416", 3, 8335400234, "function user.s16_1();", ]
Segmentation fault
I will also attache a complet debug output log.
Reproducible: Always
### Steps to Reproduce:
1. Create a very large database
2. Run msqldump
## Comment 20702
Date: 2015-03-11 09:23:32 +0100
From: @dnedev
Created attachment 325
Debug output
> Attached file: [monetdb-debug-stones12-oct2014-dev-dump.log](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3681_monetdb-debug-stones12-oct2014-dev-dump.log_325) (text/plain, 242964 bytes)
> Description: Debug output
| mserver5 crashes with a segfault when dumping very large databases | https://api.github.com/repos/MonetDB/MonetDB/issues/3681/comments | 0 | 2020-11-30T13:00:02Z | 2024-06-28T13:15:35Z | https://github.com/MonetDB/MonetDB/issues/3681 | 753,458,731 | 3,681 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-03-10 16:56:55 +0100
From: Robin Cijvat <<robin.cijvat>>
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
CC: @njnes
Last updated: 2015-05-07 12:37:43 +0200
## Comment 20700
Date: 2015-03-10 16:56:55 +0100
From: Robin Cijvat <<robin.cijvat>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2272.76 Safari/537.36
Build Identifier:
On the current default branch, the following problem occurs (copied from mclient):
sql>create table t (s string);
operation successful (2.350ms)
sql>prepare select * from t where s = ?;
execute prepared statement using: EXEC 13(...)
+------+--------+-------+--------+-------+--------+
| type | digits | scale | schema | table | column |
+======+========+=======+========+=======+========+
| clob | 0 | 0 | | t | s |
| clob | 0 | 0 | null | null | null |
+------+--------+-------+--------+-------+--------+
2 tuples (0.388ms)
sql>exec 13('foo');
symbol type not found
sql>exec 13(clob 'foo');
syntax error, unexpected CLOB in: "exec 13(clob"
sql>exec 13(string 'foo');
syntax error, unexpected CLOB in: "exec 13(string"
sql>exec 13(cast('foo' as string));
symbol type not found
sql>exec 13(cast('foo' as clob));
symbol type not found
sql>
Reproducible: Always
### Steps to Reproduce:
See SQL in field
### Actual Results:
I get the error messages shown in the details field
## Comment 20707
Date: 2015-03-11 17:02:12 +0100
From: @njnes
fixed, ie don't throw away the prepared statement when recompile is impossible
| Prepared statements fail on execution with message 'Symbol type not found' | https://api.github.com/repos/MonetDB/MonetDB/issues/3680/comments | 0 | 2020-11-30T12:59:59Z | 2024-06-27T12:06:35Z | https://github.com/MonetDB/MonetDB/issues/3680 | 753,458,702 | 3,680 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-03-05 18:30:56 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.17.21 (Jan2014-SP3)
CC: @njnes
Last updated: 2015-08-28 13:42:10 +0200
## Comment 20692
Date: 2015-03-05 18:30:56 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:35.0) Gecko/20100101 Firefox/35.0
Build Identifier:
Also it appears that any specified timezone value (either codes such as PDT, GMT, CET, DST or numeric +/-HH:MM representations such as +01:00 +04:30 -07:00) is not honered during conversion of string to timetz internal value.
Reproducible: Always
### Steps to Reproduce:
1. Start mserver5
2. Start mclient
3. execute following SQL:
CREATE TABLE TIMETZ_TBL (f1 time(2) with time zone);
INSERT INTO TIMETZ_TBL VALUES ('00:00 bla blah');
INSERT INTO TIMETZ_TBL VALUES ('00:01 PDT');
INSERT INTO TIMETZ_TBL VALUES ('00:02 GMT');
INSERT INTO TIMETZ_TBL VALUES ('00:03 CET');
INSERT INTO TIMETZ_TBL VALUES ('00:04 DST');
SELECT f1, cast(f1 as varchar(30)) as time_str FROM TIMETZ_TBL;
INSERT INTO TIMETZ_TBL VALUES ('00:00:20 +00:00 bla blah');
INSERT INTO TIMETZ_TBL VALUES ('01:01:20 +01:00');
INSERT INTO TIMETZ_TBL VALUES ('02:02:20 +04:30');
INSERT INTO TIMETZ_TBL VALUES ('03:03:20 -07:00');
SELECT f1, cast(f1 as varchar(30)) as time_str FROM TIMETZ_TBL;
### Actual Results:
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.20.0 (unreleased), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>CREATE TABLE TIMETZ_TBL (f1 time(2) with time zone);
operation successful (1.909ms)
sql>INSERT INTO TIMETZ_TBL VALUES ('00:00 bla blah');
1 affected row (1.115ms)
sql>INSERT INTO TIMETZ_TBL VALUES ('00:01 PDT');
1 affected row (0.794ms)
sql>INSERT INTO TIMETZ_TBL VALUES ('00:02 GMT');
1 affected row (0.765ms)
sql>INSERT INTO TIMETZ_TBL VALUES ('00:03 CET');
1 affected row (0.779ms)
sql>INSERT INTO TIMETZ_TBL VALUES ('00:04 DST');
1 affected row (0.829ms)
sql>SELECT f1, cast(f1 as varchar(30)) as time_str FROM TIMETZ_TBL;
+-------------------+-------------------+
| f1 | time_str |
+===================+===================+
| 01:00:00.00+01:00 | 01:00:00.00+01:00 |
| 01:01:00.00+01:00 | 01:01:00.00+01:00 |
| 01:02:00.00+01:00 | 01:02:00.00+01:00 |
| 01:03:00.00+01:00 | 01:03:00.00+01:00 |
| 01:04:00.00+01:00 | 01:04:00.00+01:00 |
+-------------------+-------------------+
5 tuples (3.646ms)
sql>INSERT INTO TIMETZ_TBL VALUES ('00:00:20 +00:00 bla blah');
1 affected row (1.160ms)
sql>INSERT INTO TIMETZ_TBL VALUES ('01:01:20 +01:00');
1 affected row (0.864ms)
sql>INSERT INTO TIMETZ_TBL VALUES ('02:02:20 +04:30');
1 affected row (0.729ms)
sql>INSERT INTO TIMETZ_TBL VALUES ('03:03:20 -07:00');
1 affected row (0.783ms)
sql>SELECT f1, cast(f1 as varchar(30)) as time_str FROM TIMETZ_TBL;
+-------------------+-------------------+
| f1 | time_str |
+===================+===================+
| 01:00:00.00+01:00 | 01:00:00.00+01:00 |
| 01:01:00.00+01:00 | 01:01:00.00+01:00 |
| 01:02:00.00+01:00 | 01:02:00.00+01:00 |
| 01:03:00.00+01:00 | 01:03:00.00+01:00 |
| 01:04:00.00+01:00 | 01:04:00.00+01:00 |
| 01:00:20.00+01:00 | 01:00:20.00+01:00 |
| 02:01:20.00+01:00 | 02:01:20.00+01:00 |
| 03:02:20.00+01:00 | 03:02:20.00+01:00 |
| 04:03:20.00+01:00 | 04:03:20.00+01:00 |
+-------------------+-------------------+
9 tuples (1.116ms)
sql>\q
### Expected Results:
errors for the "bla blah" values
correct interpretation and conversion to UTC for the other values
It is probably also a problem for the timestamptz data type
## Comment 20709
Date: 2015-03-11 19:04:56 +0100
From: @njnes
now we return an error on when only a part of the input string is parsed
## Comment 21204
Date: 2015-08-28 13:42:10 +0200
From: @sjoerdmullender
Jul2015 has been released.
| No error is given when incorrect timezone value is specified for a timetz column | https://api.github.com/repos/MonetDB/MonetDB/issues/3679/comments | 0 | 2020-11-30T12:59:57Z | 2024-06-27T12:06:34Z | https://github.com/MonetDB/MonetDB/issues/3679 | 753,458,674 | 3,679 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-03-04 14:43:47 +0100
From: @hannesmuehleisen
To: clients devs <<bugs-clients>>
Version: 11.19.9 (Oct2014-SP2)
Last updated: 2015-05-07 12:37:29 +0200
## Comment 20688
Date: 2015-03-04 14:43:47 +0100
From: @hannesmuehleisen
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/40.0.2214.115 Safari/537.36
Build Identifier:
The ruby driver wants to always install itself into the system-wide ruby directory, which requires root. This should not be the case, if a prefix is given, that should be used/
Reproducible: Always
## Comment 20689
Date: 2015-03-04 15:13:33 +0100
From: @sjoerdmullender
The default location for ruby gems on Darwin is /Library/Ruby/Site/2.0.0. We get this path by asking ruby itself:
ruby -rrbconfig -e "puts RbConfig::CONFIG['sitelibdir']"
The question is, what should we use in stead?
What we do now is try to match ruby's installation prefix with the sitelibdir value, and if the former is a prefix of the latter, we subtract that prefix from sitelibdir and use the remainder and stick it on to our own prefix. On Darwin, there is no match (the prefix as returned by ruby is /System/Library/Frameworks/Ruby.framework/Versions/2.0/usr), so we don't do that. What we could do is just glue the sitelibdir that is returned on to our prefix. You would then get
/usr/local/Library/Ruby/Site/2.0.0 (assuming --prefix=/usr/local).
Is that acceptable?
## Comment 20697
Date: 2015-03-09 17:06:46 +0100
From: @hannesmuehleisen
I think that would be acceptable!
## Comment 20699
Date: 2015-03-09 17:48:13 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [dec9f780e76d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=dec9f780e76d) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=dec9f780e76d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=dec9f780e76d)
Changeset description:
When Ruby's sitelibdir is not inside its own or our prefix, append it.
This has an effect on Darwin where Ruby's prefix is
"/System/Library/Frameworks/Ruby.framework/Versions/2.0/usr" but its
sitelibdir is "/Library/Ruby/Site/2.0.0". So we install our gems in
"$prefix/Library/Ruby/Site/2.0.0" now instead of in
"/Library/Ruby/Site/2.0.0".
This fixes bug #3678.
| Ruby driver installation ignores prefix | https://api.github.com/repos/MonetDB/MonetDB/issues/3678/comments | 0 | 2020-11-30T12:59:54Z | 2024-06-27T12:06:33Z | https://github.com/MonetDB/MonetDB/issues/3678 | 753,458,631 | 3,678 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-03-04 12:17:20 +0100
From: Richard Hughes <<richard.monetdb>>
To: GDK devs <<bugs-common>>
Version: 11.19.9 (Oct2014-SP2)
Last updated: 2015-05-07 12:38:04 +0200
## Comment 20683
Date: 2015-03-04 12:17:20 +0100
From: Richard Hughes <<richard.monetdb>>
An Oct2014 build from 2015-02-13 was working mostly fine. I upgraded to [e58372859532](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e58372859532) and got three crashes in 12 hours.
Snippets of my gdb session showing (what I consider to be) the important information:
Program terminated with signal SIGSEGV, Segmentation fault.
(gdb) bt
0 0x00007fb1a4290225 in BATgroup_internal (groups=0x78a82c,
extents=0x3c5416, histo=0x7da000, b=0x0, g=0x3c5416, e=0x0, h=0x0,
subsorted=0) at gdk_group.c:796
1 0x00007fb1a429469f in BATgroup (groups=<optimized out>,
extents=<optimized out>, histo=<optimized out>, b=<optimized out>,
g=<optimized out>, e=<optimized out>, h=0x0) at gdk_group.c:929
2 0x00007fb1a46e86d3 in GRPsubgroup4 (ngid=0x78a82c, next=0x3c5416,
nhis=0x7da000, bid=0x2, gid=0x0, eid=0x0, hid=0x0) at group.c:48
3 0x00007fb1a46e87c1 in GRPsubgroup1 (ngid=<optimized out>,
next=<optimized out>, nhis=<optimized out>, bid=<optimized out>)
at group.c:75
4 0x00007fb1a464282b in runMALsequence (cntxt=0x78a82c, mb=0x7f9c8ce059d0,
startpc=8232960, stoppc=2, stk=0x7f9c8d0cb990, env=0xc, pcicaller=0x0)
at mal_interpreter.c:654
5 0x00007fb1a46445be in DFLOWworker (T=0x78a82c) at mal_dataflow.c:363
6 0x00007fb1a301a0a4 in start_thread (arg=0x7fb197bfd700)
at pthread_create.c:309
7 0x00007fb1a2d4ecbd in clone ()
at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111
(gdb) disassemble $rip-150,$rip+50
Dump of assembler code from 0x7fb1a429018f to 0x7fb1a4290257:
0x00007fb1a429018f <BATgroup_internal+32255>: (bad)
0x00007fb1a4290190 <BATgroup_internal+32256>: decl -0x75(%rax)
0x00007fb1a4290193 <BATgroup_internal+32259>: jl 0x7fb1a42901b9 <BATgroup_internal+32297>
0x00007fb1a4290195 <BATgroup_internal+32261>: xor %cl,-0x77(%rax)
0x00007fb1a4290198 <BATgroup_internal+32264>: (bad)
0x00007fb1a4290199 <BATgroup_internal+32265>: mov %r11,0x78(%rsp)
0x00007fb1a429019e <BATgroup_internal+32270>: mov %r9,0x70(%rsp)
0x00007fb1a42901a3 <BATgroup_internal+32275>: mov %r8,0x48(%rsp)
0x00007fb1a42901a8 <BATgroup_internal+32280>: callq 0x7fb1a3f718a0 <BATsetcount@plt>
0x00007fb1a42901ad <BATgroup_internal+32285>: mov 0x28(%rsp),%rsi
0x00007fb1a42901b2 <BATgroup_internal+32290>: mov 0x30(%rsp),%rdi
0x00007fb1a42901b7 <BATgroup_internal+32295>: callq 0x7fb1a3f71810 <BATextend@plt>
0x00007fb1a42901bc <BATgroup_internal+32300>: mov 0x10(%rax),%rdx
0x00007fb1a42901c0 <BATgroup_internal+32304>: mov %rax,0x30(%rsp)
0x00007fb1a42901c5 <BATgroup_internal+32309>: mov 0x18(%rax),%rax
0x00007fb1a42901c9 <BATgroup_internal+32313>: mov 0x48(%rsp),%r8
0x00007fb1a42901ce <BATgroup_internal+32318>: mov 0x70(%rsp),%r9
0x00007fb1a42901d3 <BATgroup_internal+32323>: mov 0x78(%rsp),%r11
0x00007fb1a42901d8 <BATgroup_internal+32328>: movsbl 0xb(%rdx),%ecx
0x00007fb1a42901dc <BATgroup_internal+32332>: mov 0x20(%rax),%rax
0x00007fb1a42901e0 <BATgroup_internal+32336>: shl %cl,%rax
0x00007fb1a42901e3 <BATgroup_internal+32339>: add 0x58(%rdx),%rax
0x00007fb1a42901e7 <BATgroup_internal+32343>: mov %rax,%r10
0x00007fb1a42901ea <BATgroup_internal+32346>: mov 0xd8(%rsp),%rax
0x00007fb1a42901f2 <BATgroup_internal+32354>: jmpq 0x7fb1a428cc34 <BATgroup_internal+18596>
0x00007fb1a42901f7 <BATgroup_internal+32359>: mov 0xd8(%rsp),%rsi
0x00007fb1a42901ff <BATgroup_internal+32367>: jmpq 0x7fb1a4289d93 <BATgroup_internal+6659>
0x00007fb1a4290204 <BATgroup_internal+32372>: mov 0xd8(%rsp),%rsi
0x00007fb1a429020c <BATgroup_internal+32380>: jmpq 0x7fb1a4289ee9 <BATgroup_internal+7001>
0x00007fb1a4290211 <BATgroup_internal+32385>: xor %r13d,%r13d
0x00007fb1a4290214 <BATgroup_internal+32388>: lea 0x4cab4(%rip),%r14 0x7fb1a42dcccf
0x00007fb1a429021b <BATgroup_internal+32395>: jmpq 0x7fb1a428911e <BATgroup_internal+3470>
0x00007fb1a4290220 <BATgroup_internal+32400>: mov 0x70(%rsp),%rdi
=> 0x00007fb1a4290225 <BATgroup_internal+32405>: movzwl (%rax,%rdi,1),%eax
0x00007fb1a4290229 <BATgroup_internal+32409>: add $0x2000,%rax
0x00007fb1a429022f <BATgroup_internal+32415>: jmpq 0x7fb1a428c41a <BATgroup_internal+16522>
0x00007fb1a4290234 <BATgroup_internal+32420>: mov 0xd8(%rsp),%rsi
0x00007fb1a429023c <BATgroup_internal+32428>: jmpq 0x7fb1a4289fc4 <BATgroup_internal+7220>
0x00007fb1a4290241 <BATgroup_internal+32433>: mov 0xd8(%rsp),%rsi
0x00007fb1a4290249 <BATgroup_internal+32441>: jmpq 0x7fb1a4289ba9 <BATgroup_internal+6169>
0x00007fb1a429024e <BATgroup_internal+32446>: neg %eax
0x00007fb1a4290250 <BATgroup_internal+32448>: mov %eax,%ecx
0x00007fb1a4290252 <BATgroup_internal+32450>: and $0x3fff,%eax
End of assembler dump.
(gdb) p *bi.b->T
$4 = {id = 0x7fb1a42dcd1d "t", width = 2, type = 12 '\f', shift = 1 '\001',
varsized = 1, key = 0, dense = 0, nonil = 1, nil = 0, sorted = 0,
revsorted = 0, align = 1036545732, nokey = {0, 0}, nosorted = 0,
norevsorted = 0, nodense = 0, seq = 0, heap = {free = 359426,
size = 359426,
base = 0x7f9d7e8ea82c <error: Cannot access memory at address 0x7f9d7e8ea82c>, filename = 0x7fb0c8a82d10 "53/5343.tail", copied = 0, hashash = 0,
forcemap = 0, storage = STORE_MMAP, newstorage = STORE_MMAP,
dirty = 0 '\000', farmid = 0 '\000', parentid = -2787}, vheap = 0x7da100,
hash = 0x0, imprints = 0x0, props = 0x0}
(gdb) p/x $rax
$5 = 0x7f9d7e8ea82c
(gdb) p $rdi
$6 = 7907372
(gdb) p *bi.b->S
$7 = {tid = 140400731870977, stamp = 3763611, copiedtodisk = 0, dirty = 1,
dirtyflushed = 0, descdirty = 1, restricted = 1, persistence = 1, role = 1,
unused = 0, sharecnt = 0, map_head = 0 '\000', map_tail = 0 '\000',
map_hheap = 0 '\000', map_theap = 0 '\000', deleted = 0, first = 0,
inserted = 0, count = 179713, capacity = 179713}
(gdb) p bi.b->T->heap
$8 = {free = 359426, size = 359426,
base = 0x7f9d7e8ea82c <error: Cannot access memory at address 0x7f9d7e8ea82c>,
filename = 0x7fb0c8a82d10 "53/5343.tail", copied = 0, hashash = 0,
forcemap = 0, storage = STORE_MMAP, newstorage = STORE_MMAP,
dirty = 0 '\000', farmid = 0 '\000', parentid = -2787}
Mentally decompiling that assembler points to gdk_group.c:223 (INIT_1;) as the actual location of the crash.
I reckon gdk_group.c:766 is dodgy:
lo = (BUN) ((b->T->heap.base - b2->T->heap.base) >> b->T->shift) + BUNfirst(b);
Redoing your algebra, it looks to me like that subtraction should be the other way round.
## Comment 20684
Date: 2015-03-04 13:44:50 +0100
From: @sjoerdmullender
(In reply to comment 0)
> I reckon gdk_group.c:766 is dodgy:
> lo = (BUN) ((b->T->heap.base - b2->T->heap.base) >> b->T->shift) +
> BUNfirst(b);
>
> Redoing your algebra, it looks to me like that subtraction should be the
> other way round.
I think that code is correct.
b is the view and b2 its parent. A view's heap.base pointer points into its parent's heap, so the view's heap.base minus the parent's heap.base is the offset. That's what's being calculated here (and in gdk_select.c and gdk_unique.c).
What I think might be the problem (and it looks certainly like an error to me) is that when we switch over to using the parent instead of the view in the code you pointed to, we should also update the iterator. So after the statement b = b2; add bi = bat_iterator(b);
Could you try that?
## Comment 20685
Date: 2015-03-04 13:59:16 +0100
From: Richard Hughes <<richard.monetdb>>
You're right. My eyes totally missed the "b = b2;".
I'm rebuilding with the patch below right now. I haven't got a repro script for this, so I'm just going to be evil and put it live to see what happens.
diff -r 71963bf1b19a -r a5b37e0306d4 gdk/gdk_group.c
--- a/gdk/gdk_group.c Tue Mar 03 17:34:05 2015 +0000
+++ b/gdk/gdk_group.c Wed Mar 04 12:48:21 2015 +0000
@@ -766,6 +766,7 @@
lo = (BUN) ((b->T->heap.base - b2->T->heap.base) >> b->T->shift) + BUNfirst(b);
hi = lo + BATcount(b);
b = b2;
+ bi = bat_iterator(b);
} else {
lo = BUNfirst(b);
hi = BUNlast(b);
## Comment 20686
Date: 2015-03-04 14:09:07 +0100
From: @sjoerdmullender
I'm pretty sure the fix is correct. I just don't know whether it'll fix this particular problem.
Looking at the two other places where I used the same trick (using hash on parent of a view), the iterator uses the parent instead of the view.
## Comment 20687
Date: 2015-03-04 14:10:21 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [9b18ef8081e1](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9b18ef8081e1) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=9b18ef8081e1](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=9b18ef8081e1)
Changeset description:
Update bat iterator when switching over to parent bat.
This may fix bug #3677.
## Comment 20690
Date: 2015-03-05 11:11:34 +0100
From: Richard Hughes <<richard.monetdb>>
No crashes since yesterday. Resolved/fixed.
P.S. The HASH parameter to GRP_use_existing_hash_table is now superfluous.
| Crash in BATgroup_internal (caused by 87379087770d?) | https://api.github.com/repos/MonetDB/MonetDB/issues/3677/comments | 0 | 2020-11-30T12:59:51Z | 2024-06-27T12:06:32Z | https://github.com/MonetDB/MonetDB/issues/3677 | 753,458,584 | 3,677 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-03-03 19:11:43 +0100
From: Richard Hughes <<richard.monetdb>>
To: Merovingian devs <<bugs-merovingian>>
Version: 11.19.9 (Oct2014-SP2)
Last updated: 2015-05-07 12:37:39 +0200
## Comment 20682
Date: 2015-03-03 19:11:43 +0100
From: Richard Hughes <<richard.monetdb>>
Steps:
1) Start up quite a few databases (we now have 23 mserver5 instances)
2) service monetdb5-sql stop
Expected:
Everything exits eventually.
Actual:
We're left with 3 defunct mserver5 processes and monetdbd spinning one core.
This is intermittent, but not uncommon (50% of the time?).
0 0x00007fb090a04add in read () at ../sysdeps/unix/syscall-template.S:81
1 0x00000000004073a1 in read (__nbytes=<optimized out>,
__buf=<optimized out>, __fd=<optimized out>)
at /usr/include/x86_64-linux-gnu/bits/unistd.h:44
2 logFD (fd=57,
type=0x7fb08bee7e30 "database 'test' (31079) has exited with exit status
0", type@entry=0x41953a "MSG",
dbname=0x1f9f <error: Cannot access memory at address 0x1f9f>, pid=-1,
stream=0x223f580) at merovingian.c:152
3 0x000000000040766a in logListener (x=<optimized out>) at merovingian.c:232
4 0x00007fb0909fe0a4 in start_thread (arg=0x7fb08beea700)
at pthread_create.c:309
5 0x00007fb090732cbd in clone ()
at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111
I guess what's happening is that the logListener is charging round its do...while loop as fast as possible and never checking _mero_keep_logging because the fd is always readable because it's in the closed state because the child has exited.
I haven't confirmed this hypothesis because that machine is serving actual users so I don't like to shut it down too often. I might try careful application of some big sleep() calls after the select() to see if I can widen the race window on a dev box.
## Comment 20792
Date: 2015-04-14 11:46:16 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [275d5e3b8cf4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=275d5e3b8cf4) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=275d5e3b8cf4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=275d5e3b8cf4)
Changeset description:
Stop logging when we're asked to.
This fixes bug #3676.
## Comment 20793
Date: 2015-04-14 12:34:47 +0200
From: @sjoerdmullender
I was able to reproduce the problem and saw that your assessment was correct. It should now be fixed.
| merovingian hangs trying to exit | https://api.github.com/repos/MonetDB/MonetDB/issues/3676/comments | 0 | 2020-11-30T12:59:49Z | 2024-06-27T12:06:31Z | https://github.com/MonetDB/MonetDB/issues/3676 | 753,458,547 | 3,676 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-02-26 17:24:47 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: MonetDB5 devs <<bugs-monetdb5>>
Version: -- development
CC: @mlkersten, @swingbit, @drstmane
Last updated: 2019-03-19 08:10:21 +0100
## Comment 20679
Date: 2015-02-26 17:24:47 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:35.0) Gecko/20100101 Firefox/35.0
Build Identifier:
Multiple combined calculations quickly result in an output defined as hugeint data type.
For instance: SELECT 1 + 2 + 3 + 4 + 5;
creates an output 15 as hugeint with following mal plan:
function user.s29_2{autoCommit=true}(A0:sht,A1:sht,A2:int,A3:lng,A4:hge):void;
X_23:void := querylog.define("select 1 + 2 + 3 + 4 + 5\n;","default_pipe",11);
X_7 := calc.+(A0,A1);
X_8 := calc.int(X_7);
X_9 := calc.+(X_8,A2);
X_10 := calc.lng(X_9);
X_11 := calc.+(X_10,A3);
X_12 := calc.hge(X_11);
X_13 := calc.+(X_12,A4);
sql.exportValue(1,".L","sql_add_single_value","hugeint",128,0,7,X_13,"");
end user.s29_2;
optimizer.mitosis()
optimizer.dataflow()
Some improvements could be made:
1) for small constants (such as -127 .. 127), the mal generator could maybe use bte instead of sht
2) when adding a sht (or bte) to an int the result will probably fit in an int. So the instruction X_10 := calc.lng(X_9); could potentially be removed (or relaced.
3) idem for adding an int (or sht or bte) to a lng.
So instruction X_12 := calc.hge(X_11); could potentially be removed (or relaced).
4) before doing the sql.exportValue(1,".L","sql_add_single_value","hugeint",128,0,7,X_13,""); instruction, you could check the maximum and minimum value in X_13 bat and determine if a smaller end result column is possible. If so an additional mal X_14 := calc.lng(X_13); or X_14 := calc.int(X_13); could be generated.
Same applies for subtractions, multiplications, etc. calculations.
When we produce less (intermediate) hge bats, less memory is needed, so improves scalability. Potentially also some operations on (e.g. multiplication) on int and lng are faster than on hge bats.
Another benefit is that .int128 specific Mtest output files will be less needed, as in many cases (see improvement step 4) above) the output will be of data type int or lng.
Reproducible: Always
### Steps to Reproduce:
1. Start mserver
2. Start mclient
3. runs SQL:
SELECT 1 + 2 + 3 + 4 + 5 + 6 AS twenty1;
plan SELECT 1 + 2 + 3 + 4 + 5 + 6 AS twenty1;
explain SELECT 1 + 2 + 3 + 4 + 5 + 6 AS twenty1;
SELECT -1 - 2 - 3 - 4 - 5 - 6 AS minus21;
plan SELECT -1 - 2 - 3 - 4 - 5 - 6 AS minus21;
explain SELECT -1 - 2 - 3 - 4 - 5 - 6 AS minus21;
SELECT 1 * 2 * 3 * 4 * 5 * 6 AS seven20;
plan SELECT 1 * 2 * 3 * 4 * 5 * 6 AS seven20;
explain SELECT 1 * 2 * 3 * 4 * 5 * 6 AS seven20;
### Actual Results:
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.20.0 (unreleased), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>SELECT 1 + 2 + 3 + 4 + 5 + 6 AS twenty1;
+---------+
| twenty1 |
+=========+
| 21 |
+---------+
1 tuple (1.300ms)
sql>plan SELECT 1 + 2 + 3 + 4 + 5 + 6 AS twenty1;
+-------------------------------------------------------------------------------------------------------------------------+
| rel |
+=========================================================================================================================+
| [ sys.sql_add(sys.sql_add(hugeint[sys.sql_add(bigint[sys.sql_add(int[sys.sql_add(smallint[tinyint "1"], smallint[tinyi |
: nt "2"])], int[smallint[tinyint "3"]])], bigint[smallint[tinyint "4"]])], hugeint[smallint[tinyint "5"]]), hugeint[smal :
: lint[tinyint "6"]]) as L.twenty1 ] :
+-------------------------------------------------------------------------------------------------------------------------+
1 tuple (0.901ms)
sql>explain SELECT 1 + 2 + 3 + 4 + 5 + 6 AS twenty1;
+--------------------------------------------------------------------------------------------------+
| mal |
+==================================================================================================+
| function user.s2_3{autoCommit=true}(A0:sht,A1:sht,A2:int,A3:lng,A4:hge,A5:hge):void; |
| X_25:void := querylog.define("select 1 + 2 + 3 + 4 + 5 + 6 as twenty1;","default_pipe",12); |
| X_8 := calc.+(A0,A1); |
| X_9 := calc.int(X_8); |
| X_10 := calc.+(X_9,A2); |
| X_11 := calc.lng(X_10); |
| X_12 := calc.+(X_11,A3); |
| X_13 := calc.hge(X_12); |
| X_14 := calc.+(X_13,A4); |
| X_15 := calc.+(X_14,A5); |
| sql.exportValue(1,".L","twenty1","hugeint",128,0,7,X_15,""); |
| end user.s2_3; |
| optimizer.mitosis() |
| optimizer.dataflow() |
+--------------------------------------------------------------------------------------------------+
14 tuples (1.117ms)
sql>SELECT -1 - 2 - 3 - 4 - 5 - 6 AS minus21;
+---------+
| minus21 |
+=========+
| -21 |
+---------+
1 tuple (1.017ms)
sql>plan SELECT -1 - 2 - 3 - 4 - 5 - 6 AS minus21;
+-------------------------------------------------------------------------------------------------------------------------+
| rel |
+=========================================================================================================================+
| [ sys.sql_sub(sys.sql_sub(hugeint[sys.sql_sub(bigint[sys.sql_sub(int[sys.sql_sub(smallint[sys.sql_neg(tinyint "1")], s |
: mallint[tinyint "2"])], int[smallint[tinyint "3"]])], bigint[smallint[tinyint "4"]])], hugeint[smallint[tinyint "5"]]), :
: hugeint[smallint[tinyint "6"]]) as L.minus21 ] :
+-------------------------------------------------------------------------------------------------------------------------+
1 tuple (0.751ms)
sql>explain SELECT -1 - 2 - 3 - 4 - 5 - 6 AS minus21;
+---------------------------------------------------------------------------------------------------+
| mal |
+===================================================================================================+
| function user.s3_3{autoCommit=true}(A0:sht,A1:sht,A2:int,A3:lng,A4:hge,A5:hge):void; |
| X_25:void := querylog.define("select -1 - 2 - 3 - 4 - 5 - 6 as minus21;","default_pipe",12); |
| X_8 := calc.-(A0,A1); |
| X_9 := calc.int(X_8); |
| X_10 := calc.-(X_9,A2); |
| X_11 := calc.lng(X_10); |
| X_12 := calc.-(X_11,A3); |
| X_13 := calc.hge(X_12); |
| X_14 := calc.-(X_13,A4); |
| X_15 := calc.-(X_14,A5); |
| sql.exportValue(1,".L","minus21","hugeint",128,0,7,X_15,""); |
| end user.s3_3; |
| optimizer.mitosis() |
| optimizer.dataflow() |
+---------------------------------------------------------------------------------------------------+
14 tuples (1.045ms)
sql>SELECT 1 * 2 * 3 * 4 * 5 * 6 AS seven20;
+---------+
| seven20 |
+=========+
| 720 |
+---------+
1 tuple (1.241ms)
sql>plan SELECT 1 * 2 * 3 * 4 * 5 * 6 AS seven20;
+-------------------------------------------------------------------------------------------------------------------------+
| rel |
+=========================================================================================================================+
| [ sys.sql_mul(sys.sql_mul(hugeint[sys.sql_mul(bigint[sys.sql_mul(int[sys.sql_mul(smallint[tinyint "1"], tinyint "2")], |
: smallint[tinyint "3"])], int[tinyint "4"])], bigint[tinyint "5"]), hugeint[tinyint "6"]) as L.seven20 ] :
+-------------------------------------------------------------------------------------------------------------------------+
1 tuple (0.849ms)
sql>explain SELECT 1 * 2 * 3 * 4 * 5 * 6 AS seven20;
+--------------------------------------------------------------------------------------------------+
| mal |
+==================================================================================================+
| function user.s4_3{autoCommit=true}(A0:sht,A1:bte,A2:sht,A3:int,A4:lng,A5:hge):void; |
| X_25:void := querylog.define("select 1 * 2 * 3 * 4 * 5 * 6 as seven20;","default_pipe",12); |
| X_8 := calc.*(A0,A1); |
| X_9 := calc.int(X_8); |
| X_10 := calc.*(X_9,A2); |
| X_11 := calc.lng(X_10); |
| X_12 := calc.*(X_11,A3); |
| X_13 := calc.hge(X_12); |
| X_14 := calc.*(X_13,A4); |
| X_15 := calc.*(X_14,A5); |
| sql.exportValue(1,".L","seven20","hugeint",128,0,7,X_15,""); |
| end user.s4_3; |
| optimizer.mitosis() |
| optimizer.dataflow() |
+--------------------------------------------------------------------------------------------------+
14 tuples (1.017ms)
sql>
### Expected Results:
result data types to be not hge but sht in this case (or int or lng in other cases where hge is not needed)
## Comment 20680
Date: 2015-02-26 17:38:29 +0100
From: @drstmane
Thanks, Martin!
This is indeed a problem that has been existing for some time, but has become (more) crucial with the introduction of 128-bit hge / hugeint . We are aware of his and have initial ideas how to potentially tackle the problem.
Best,
Stefan
## Comment 26453
Date: 2018-05-17 11:19:06 +0200
From: @swingbit
Is there any progress on this? I have many plans where several seconds are spent just converting to hge.
## Comment 26929
Date: 2019-03-19 08:10:21 +0100
From: @mlkersten
In general such optimizations call for symbolic reasoning over the MAL plan and then still only holds for constant expressions.
Moreover, up/down casting should respect the final signatures, which then have to be checked for existence as well.
It is better archived as a feature request.
| Request to Optimise the numeric result data type when doing multiple combined calculations | https://api.github.com/repos/MonetDB/MonetDB/issues/3675/comments | 0 | 2020-11-30T12:59:46Z | 2024-06-28T07:19:44Z | https://github.com/MonetDB/MonetDB/issues/3675 | 753,458,508 | 3,675 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-02-22 21:50:18 +0100
From: @mlkersten
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.17.21 (Jan2014-SP3)
Last updated: 2015-08-28 13:42:01 +0200
## Comment 20670
Date: 2015-02-22 21:50:18 +0100
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.10; rv:35.0) Gecko/20100101 Firefox/35.0
Build Identifier:
The performance event stream contains all the details of the actions against the database. Every eligible MonetDB user can grab this stream and thereby compromise privacy.
It would make sense to limit this feature to the database administrator, who, using e.g. the Tachograph web-service makes this information accessible to selective users.
Reproducible: Always
## Comment 20987
Date: 2015-07-09 21:53:44 +0200
From: @mlkersten
The MAL profiler only emits events from the user initiating the profiling request. This greatly reduces the security leak while still allowing users to
study the performance of queries.
Currently, any user can at any time 'steal' the profiling stream.
## Comment 21198
Date: 2015-08-28 13:42:01 +0200
From: @sjoerdmullender
Jul2015 has been released.
| Obfuscate event tracing | https://api.github.com/repos/MonetDB/MonetDB/issues/3674/comments | 0 | 2020-11-30T12:59:43Z | 2024-06-27T12:06:30Z | https://github.com/MonetDB/MonetDB/issues/3674 | 753,458,484 | 3,674 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-02-16 18:15:29 +0100
From: Richard Hughes <<richard.monetdb>>
To: clients devs <<bugs-clients>>
Version: 11.19.9 (Oct2014-SP2)
Last updated: 2015-08-28 13:41:51 +0200
## Comment 20665
Date: 2015-02-16 18:15:29 +0100
From: Richard Hughes <<richard.monetdb>>
Created attachment 322
patch
We have some tables with a lot of columns (that's what column stores are for) and none of mclient's renderers create very easy-to-read output in this situation, so I've added a clone of psql's \x formatter. Sample output:
sql>\f x
sql>select * from sys.functions limit 2;
-[ RECORD 1 ]------------
id | 28
name | not_uniques
func | not_uniques
mod | sql
language | 0
type | 1
side_effect | false
varres | false
vararg | false
schema_id | 0
-[ RECORD 2 ]------------
id | 29
name | not_uniques
func | not_uniques
mod | sql
language | 0
type | 1
side_effect | false
varres | false
vararg | false
schema_id | 0
sql>
I like it, and other people might too.
> Attached file: [mclient-expanded-formatter.patch](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3673_mclient-expanded-formatter.patch_322) (text/plain, 2162 bytes)
> Description: patch
## Comment 20668
Date: 2015-02-19 16:16:18 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [6e2859f319f4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=6e2859f319f4) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=6e2859f319f4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=6e2859f319f4)
Changeset description:
Implemented a new output format in mclient: --format=expanded.
This is based on a suggestion with accompanying code from Richard
Hughes, but with some changes to deal with multi-line values. See
bug #3673.
## Comment 20669
Date: 2015-02-19 16:17:11 +0100
From: @sjoerdmullender
Thanks. I have taken your code and incorporated into the default branch (with some changes).
## Comment 21190
Date: 2015-08-28 13:41:51 +0200
From: @sjoerdmullender
Jul2015 has been released.
| mclient 'expanded' row formatter | https://api.github.com/repos/MonetDB/MonetDB/issues/3673/comments | 0 | 2020-11-30T12:59:41Z | 2024-06-27T12:06:29Z | https://github.com/MonetDB/MonetDB/issues/3673 | 753,458,448 | 3,673 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-02-15 22:07:44 +0100
From: blenoir
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
Last updated: 2015-05-07 12:37:31 +0200
## Comment 20659
Date: 2015-02-15 22:07:44 +0100
From: blenoir
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_5) AppleWebKit/600.2.5 (KHTML, like Gecko) Version/7.1.2 Safari/537.85.11
Build Identifier:
Compilation of MonetDB-11.19.9 generates error in 'make' on Oracle Linux (2.6.18-371.11.1.0.1.el5 - x86_64).
/usr/bin/ld: .libs/libbat_la-gdk_utils.o: relocation R_X86_64_PC32 against `MT_global_exit' can not be used when making a shared object; recompile with -fPIC
Reproducible: Always
### Steps to Reproduce:
1. ./configure --prefix=/usr/local/MonetDB --enable-jdbc --enable-rintegration
2.
3.
### Actual Results:
Compilation fails.
## Comment 20660
Date: 2015-02-15 22:11:21 +0100
From: blenoir
Created attachment 320
output from running configure
> Attached file: [configOutput.txt](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3672_configOutput.txt_320) (text/plain, 24629 bytes)
> Description: output from running configure
## Comment 20661
Date: 2015-02-15 22:15:02 +0100
From: blenoir
Created attachment 321
config.log from running configure
> Attached file: [config.log](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3672_config.log_321) (text/plain, 323850 bytes)
> Description: config.log from running configure
## Comment 20662
Date: 2015-02-16 10:17:20 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [d424dd783e48](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d424dd783e48) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=d424dd783e48](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=d424dd783e48)
Changeset description:
Don't declare a function both as "hidden" and as "extern".
This hopefully fixes bug #3672.
## Comment 20664
Date: 2015-02-16 10:45:13 +0100
From: blenoir
This compiles.
Thanks!
| libbat_la-gdk_utils.o: relocation R_X86_64_PC32 against `MT_global_exit' can not be used when making a shared object | https://api.github.com/repos/MonetDB/MonetDB/issues/3672/comments | 0 | 2020-11-30T12:59:38Z | 2024-06-27T12:06:28Z | https://github.com/MonetDB/MonetDB/issues/3672 | 753,458,419 | 3,672 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-02-12 16:54:50 +0100
From: Dieter <<mayerd>>
To: clients devs <<bugs-clients>>
Version: 11.19.9 (Oct2014-SP2)
CC: @drstmane
Last updated: 2015-05-07 12:37:30 +0200
## Comment 20655
Date: 2015-02-12 16:54:50 +0100
From: Dieter <<mayerd>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/40.0.2214.111 Safari/537.36
Build Identifier:
An error occurs when querying a table multiple times on the same ADO Connection object using a client side Cursor
ErrorCode: E_FAIL from Microsoft Cursor Engine
Error happens on operating system Windows 2012 Server.
MonetDB Database Server: CENTOS 7
The error doesn't occures on other ODBC drivers e.g MSSQL Server or Oracle
nor on older Windows OS versions.
Reproducible: Always
### Steps to Reproduce:
1. Create a Windows ODBC System DSN named 'MonetDB' user 'voc' password 'voc'
2. Create a table in the catalog 'voc' like this and fill it:
create table prices (currency char(3) , valid_from char(8), price decimal(15,2));
insert into prices values('USD', '20130101' , 1.2);
insert into prices values('USD', '20130201' , 1.3);
insert into prices values('USD', '20111201' , 2.3);
3. Execute the script which executes a query multiple times like this (VB-Script):
set con = CreateObject("ADODB.Connection")
conStr = "Provider='MSDASQL.1';Data Source='MonetDB';Initial Catalog='voc';"
'when comment following statement ado uses default server cursor -> works fine
'uncommenting following statement cause a E_FAIL in Microsoft Cursor Engine
'
con.CursorLocation = 3 'client side cursor
'
con.Open conStr, "voc", "voc"
set cmd = CreateObject("ADODB.Command")
cmd.ActiveConnection = con
cmd.CommandType = 1 'text
cmd.CommandText = "select * from prices"
for i = 1 to 5
set rs = cmd.Execute(records, ,-1)
While not rs.EOF
rs.MoveNext
WEnd
rs.Close
set rs = nothing
next
### Actual Results:
Error occurs with message "... Status E_FAIL" when executing the script/ select commands.
### Expected Results:
Select and fetch of database records should be done without problems.
Test environment:
-----------------
Operating System: Windows Server 2012 (using msado15.dll Version 6.2.9200 16384 which comes with Installation of the OS).
MonetDB ODBC Driver Version 11.19.09.01 32Bit or 64 Bit
## Comment 20656
Date: 2015-02-12 16:58:13 +0100
From: Dieter <<mayerd>>
Created attachment 319
VB Script for testing database access on Windows 2012
Running this script results in an error on Windwos 2012.
On older Windows OS versions (Windows 2008, Windows 7) it works.
> Attached file: [monetQueryClientCursor.vbs](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3671_monetQueryClientCursor.vbs_319) (application/octet-stream, 1327 bytes)
> Description: VB Script for testing database access on Windows 2012
## Comment 20671
Date: 2015-02-24 10:15:24 +0100
From: @sjoerdmullender
I don't have access to Windows 2012 (or Windows 8), so it's hard for me to reproduce. What might help is some debug output from the MonetDB ODBC driver.
Can you please create a log (see below) and attach it to this bug report.
To create a log, you need to start the application that uses the MonetDB ODBC driver with an extra environment variable. Probably the easiest way to do that is to start a cmd window (Start -> Run... -> cmd.exe) and in that window type
set ODBCDEBUG=C:\....\odbc.log
and then start your application from this window.
The bit after the = should be the absolute pathname of a file you can write to. After the application is done it should contain a log of the complete interaction with the driver. Before attaching the file, please make sure it doesn't contain any sensitive information.
Instead of attaching, you can also send the file directly to me.
## Comment 20672
Date: 2015-02-24 16:41:23 +0100
From: Dieter <<mayerd>>
Created attachment 323
odbc.log
The last "select * from Prices" is the command wich results the error
> Attached file: [odbc.log](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3671_odbc.log_323) (application/octet-stream, 46993 bytes)
> Description: odbc.log
## Comment 20673
Date: 2015-02-25 15:37:27 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [48623d9c99c9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=48623d9c99c9) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=48623d9c99c9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=48623d9c99c9)
Changeset description:
32/64 bit fixes to the ODBC driver.
It turns out that current documentation on ODBC on msdn.microsoft.com
specify when to use 32 and 64 bits in the interface. That wasn't
always the case.
This should fix bug #3671.
## Comment 20674
Date: 2015-02-25 20:30:45 +0100
From: Dieter <<mayerd>>
Thanks for your engagement on this issue.
Can you provide a compiled Version of the corrected ODBC.dll or whatever I need to test the corrected version as download for my 32bit/64bit Windows - Plattforms. I'd like to test it in my environment and I'll give you a Feedback.
Dieter
## Comment 20675
Date: 2015-02-25 20:37:26 +0100
From: @drstmane
Dieter,
you'll find nightly built installers later tonight / tomorrow morning via our 'TestWeb" (http://monetdb.cwi.nl/testweb/web/status.php) under http://monetdb.cwi.nl/testweb/web/54807:48623d9c99c9/
Stefan
## Comment 20681
Date: 2015-03-02 17:58:45 +0100
From: Dieter <<mayerd>>
I have tested the nightly build ODBC installers (.msi) from today.
The problem is not solved with your changes.
## Comment 20691
Date: 2015-03-05 15:02:21 +0100
From: Dieter <<mayerd>>
Hi Sjoerd,
you've written, that you don't have access to a Windows 2012/Windows 8 system.
If it helps you, I can provide a Windows 8 virtual system (about 15GB),
which you can run using a VMWare player (cost free). It's a Microsoft Developer license which you could use only for this bug issue.
If you need this VMWare file, i would sent you a download link to your personal email address.
Please let me know the further steps.
Dieter
## Comment 20693
Date: 2015-03-06 09:54:49 +0100
From: @sjoerdmullender
I now do have access to a Windows 2012 virtual machine. I was able to download one from Microsoft that I can use for 180 days.
Using it, I can indeed reproduce the problem.
However, I can't currently think of anything else I could try to fix the problem. My problem is, using the debug option in the MonetDB ODBC driver and the trace option in the ODBC Driver Manager, I can't see anything strange happening. When the error occurs, I can see only one call to SQLExtendedFetch instead of three (one for each row in the query result). However, I don't see any errors happening there. It is as if the application just decided one is enough. Without access to any more information about what the application (the VisualBasic interpreter) is doing, it is very hard to figure out why the error is happening.
Is there any trace facility in VB? Is there source code available of the relevant part of the VB interpreter?
## Comment 20695
Date: 2015-03-06 13:01:43 +0100
From: Dieter <<mayerd>>
Hi Sjoerd,
with Windows 8/Sever2012 there is a new tracetool available using powershell commands. See Data Access Tracing in Windows 8
https://msdn.microsoft.com/en-us/library/hh829624(v=vs.85).aspx
Perhaps this may help you.
I also developed a small Windows console program in Visual C++ (odbctest.exe) which do a select against the database like the VBscript does.
I packed the source code and Visual Studio project files in a zip file. Because it uses VC runtime dll's and only release builds
for that dlls are distributed with the operation system, the release build of the testprogam has compiler optimizations disabled but generates
a program database file for debugging.
## Comment 20696
Date: 2015-03-06 13:05:09 +0100
From: Dieter <<mayerd>>
Created attachment 324
ODBC Testprogramm in C with Visual Studio project
for testing with a c programm (alternative to the VBscript)
> Attached file: [odbctest.zip](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3671_odbctest.zip_324) (application/x-zip-compressed, 72282 bytes)
> Description: ODBC Testprogramm in C with Visual Studio project
## Comment 20698
Date: 2015-03-09 17:20:23 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [9eed5049e02a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9eed5049e02a) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=9eed5049e02a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=9eed5049e02a)
Changeset description:
The RowStatusArray pointer in SQLExtendedFetch replace the *I*RD status ptr.
I think this fixes bug #3671.
## Comment 20703
Date: 2015-03-11 10:53:16 +0100
From: Dieter <<mayerd>>
Hi Sjoerd,
with your changes the problem seems to fixed.
My first test on Windows 2012 was successful.
Thanks for your precious engangement.
Dieter
| ODBC-Access on Windows 2012 does not work - E_FAIL | https://api.github.com/repos/MonetDB/MonetDB/issues/3671/comments | 0 | 2020-11-30T12:59:36Z | 2024-06-27T12:06:27Z | https://github.com/MonetDB/MonetDB/issues/3671 | 753,458,379 | 3,671 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-02-09 08:43:42 +0100
From: Asif Khan Suri <<asif.suri>>
To: SQL devs <<bugs-sql>>
Version: -- development
Last updated: 2016-04-11 11:45:26 +0200
## Comment 20642
Date: 2015-02-09 08:43:42 +0100
From: Asif Khan Suri <<asif.suri>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:38.0) Gecko/20100101 Firefox/38.0
Build Identifier:
minor change in sql.c to allow binary load inet type. I am not part of dev group but tried following to get it work. binary load is working after change.
Add new function
int wrap_inet_type(int localtype)
{
static int type_inet = 0;
if( type_inet == 0 || type_inet == TYPE_any) type_inet = ATOMindex("inet");
if( localtype > TYPE_str && localtype == type_inet && type_inet < TYPE_any )
return TYPE_lng;
return localtype;
}
and replace
int tpe = col->type.type->localtype;
with
int tpe = wrap_inet_type(col->type.type->localtype);
in function mvc_bin_import_table_wrap
Reproducible: Always
### Steps to Reproduce:
1.create table inetip (ip inet);
2.COPY BINARY INTO inetip FROM ( '/root/MonetDB/ip.bat' );
3.
### Actual Results:
failed to attach file /root/MonetDB/ip.bat
### Expected Results:
create table inetip (ip inet);
COPY BINARY INTO inetip FROM ( '/root/MonetDB/ip.bat' );
[ 10 ]
select * from inetip;
% sys.inetip table_name
% ip name
% inet type
% 0 length
[ 192.168.100.50 ]
[ 193.168.100.51 ]
[ 193.168.100.52/24 ]
[ 162.168.100.50 ]
[ 163.168.100.51/24 ]
[ 163.168.100.52 ]
[ 162.168.100.50 ]
[ 163.168.100.52 ]
[ NULL ]
[ 163.168.100.52 ]
drop table inetip;
## Comment 20643
Date: 2015-02-09 08:46:14 +0100
From: Asif Khan Suri <<asif.suri>>
Created attachment 318
inet data to do binary load
> Attached file: [ip.bat](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3670_ip.bat_318) (application/octet-stream, 80 bytes)
> Description: inet data to do binary load
| COPY BINARY INTO support for inet type | https://api.github.com/repos/MonetDB/MonetDB/issues/3670/comments | 0 | 2020-11-30T12:59:33Z | 2024-06-28T07:19:44Z | https://github.com/MonetDB/MonetDB/issues/3670 | 753,458,349 | 3,670 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-02-05 18:57:27 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2015-05-07 12:37:34 +0200
## Comment 20637
Date: 2015-02-05 18:57:27 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:35.0) Gecko/20100101 Firefox/35.0
Build Identifier:
>mserver5: ../../dev/gdk/gdk_group.c:386: BATgroup_internal: Assertion `g == ((void *)0) || ((b)->S->count) == ((g)->S->count)' failed.
mserver5.sh: line 5: 13885 Aborted /export/scratch1/dinther/INSTALL/bin/mserver5 -d10 --set embedded_r=true --set mapi_port=41000
Reproducible: Always
### Steps to Reproduce:
1. Start mserver5 (MonetDB v11.20.0 )
2. Start mclient
3. execute SQL:
create table tab (id int not null, nm varchar(30) not null, dt date, qnt dec(30,10), descr text);
-- table created.
insert into tab (id, nm) values (1, 'A');
insert into tab (id, nm, dt, qnt, descr) values (2, 'B', '2015-01-29', 3.1415629, 'iasdfhiasdhagdsnfgankkkjfgjklfgjklsklsklsdfg');
select * from tab;
-- 2 rows.
insert into tab (id, nm, dt, qnt, descr) select id, nm, dt, qnt, descr from tab;
--- 2 rows inserted
select * from tab;
-- 4 rows.
select count(*) as count_dupl_rows, id, nm, dt, qnt, descr from tab
group by id, nm, dt, qnt, descr having count(*) > 1
order by id, nm, dt, qnt, descr;
-- 2 rows (id = 1 and 2) exists twice.
alter table tab add constraint tab_uc6 unique (id, nm, dt, qnt, descr);
### Actual Results:
bash-4.2$ mclient -p 41000
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.20.0 (unreleased), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>create table tab (id int not null, nm varchar(30) not null, dt date, qnt dec(30,10), descr text);
operation successful (7.160ms)
sql>insert into tab (id, nm) values (1, 'A');
1 affected row (1.533ms)
sql>insert into tab (id, nm, dt, qnt, descr) values (2, 'B', '2015-01-29', 3.1415629, 'iasdfhiasdhagdsnfgankkkjfgjklfgjklsklsklsdfg');
1 affected row (1.218ms)
sql>
sql>select * from tab;
+------+------+------------+----------------------------------+----------------------------------------------+
| id | nm | dt | qnt | descr |
+======+======+============+==================================+==============================================+
| 1 | A | null | null | null |
| 2 | B | 2015-01-29 | 3.1415629000 | iasdfhiasdhagdsnfgankkkjfgjklfgjklsklsklsdfg |
+------+------+------------+----------------------------------+----------------------------------------------+
2 tuples (2.444ms)
sql>insert into tab (id, nm, dt, qnt, descr) select id, nm, dt, qnt, descr from tab;
2 affected rows (2.841ms)
sql>select * from tab;
+------+------+------------+----------------------------------+----------------------------------------------+
| id | nm | dt | qnt | descr |
+======+======+============+==================================+==============================================+
| 1 | A | null | null | null |
| 2 | B | 2015-01-29 | 3.1415629000 | iasdfhiasdhagdsnfgankkkjfgjklfgjklsklsklsdfg |
| 1 | A | null | null | null |
| 2 | B | 2015-01-29 | 3.1415629000 | iasdfhiasdhagdsnfgankkkjfgjklfgjklsklsklsdfg |
+------+------+------------+----------------------------------+----------------------------------------------+
4 tuples (3.144ms)
sql>select count(*) as count_dupl_rows, id, nm, dt, qnt, descr from tab
more>group by id, nm, dt, qnt, descr having count(*) > 1
more>order by id, nm, dt, qnt, descr;
+------+------+------+------------+----------------------------------+----------------------------------------+
| coun | id | nm | dt | qnt | descr |
: t_du : : : : : :
: pl_r : : : : : :
: ows : : : : : :
+======+======+======+============+==================================+========================================+
| 2 | 1 | A | null | null | null |
| 2 | 2 | B | 2015-01-29 | 3.1415629000 | iasdfhiasdhagdsnfgankkkjfgjklfgjklskls |
: : : : : : klsdfg :
+------+------+------+------------+----------------------------------+----------------------------------------+
2 tuples (9.792ms)
sql>alter table tab add constraint tab_uc6 unique (id, nm, dt, qnt, descr);
sql>select * from tab;
bash-4.2$
bash-4.2$ ./start_INSTALL_mserver5.sh
builtin opt gdk_dbpath = /export/scratch1/dinther/INSTALL/var/monetdb5/dbfarm/demo
builtin opt gdk_debug = 0
builtin opt gdk_vmtrim = no
builtin opt monet_prompt = >
builtin opt monet_daemon = no
builtin opt mapi_port = 50000
builtin opt mapi_open = false
builtin opt mapi_autosense = false
builtin opt sql_optimizer = default_pipe
builtin opt sql_debug = 0
cmdline opt embedded_r = true
cmdline opt mapi_port = 41000
cmdline opt gdk_debug = 10
MonetDB 5 server v11.20.0
This is an unreleased version
Serving database 'demo', using 8 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs and 128bit integers dynamically linked
Found 15.590 GiB available main-memory.
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2015 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Listening for connection requests on mapi:monetdb://127.0.0.1:41000/
MonetDB/GIS module loaded
MonetDB/SQL module loaded
MonetDB/R module loaded
>mserver5: ../../dev/gdk/gdk_group.c:386: BATgroup_internal: Assertion `g == ((void *)0) || ((b)->S->count) == ((g)->S->count)' failed.
./start_INSTALL_mserver5.sh: line 5: 13885 Aborted /export/scratch1/dinther/INSTALL/bin/mserver5 -d10 --set embedded_r=true --set mapi_port=41000
bash-4.2$
### Expected Results:
no assertion failure
## Comment 20641
Date: 2015-02-08 21:42:17 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [d122ab45f0ec](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d122ab45f0ec) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=d122ab45f0ec](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=d122ab45f0ec)
Changeset description:
fixed bug #3669, correct alignment when removing nulls in ukey checks
| ALTER TABLE <tbl_nm> ADD CONSTRAINT <tbl_uc1> UNIQUE (col1, col2, col3) causes Assertion failure and abort | https://api.github.com/repos/MonetDB/MonetDB/issues/3669/comments | 0 | 2020-11-30T12:59:30Z | 2024-06-27T12:06:25Z | https://github.com/MonetDB/MonetDB/issues/3669 | 753,458,322 | 3,669 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-02-05 18:41:56 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2016-04-11 11:46:31 +0200
## Comment 20636
Date: 2015-02-05 18:41:56 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:35.0) Gecko/20100101 Firefox/35.0
Build Identifier:
The uniqueness constraint defined by a
CREATE UNIQUE INDEX <idx_nm> ON <tbl_nm> (<col_nm_list>)
is not enforced.
However the
ALTER TABLE <tbl_nm> ADD CONSTRAINT <uc_nm> UNIQUE (<col_nm_list>)
does check and enforce uniqueness.
The same logic should be implemented to get the expected semantic behavior after creating a unique index.
The good news is that the code to enforce it already exists (as used by unique constraints).
Reproducible: Always
### Steps to Reproduce:
1. Start mserver5
2. Start mclient
3. Enter SQL commands:
create table tab (id int not null, nm varchar(30) not null, dt date, qnt dec(30,10), descr text);
insert into tab (id, nm) values (1, 'A');
insert into tab (id, nm, dt, qnt, descr) values (2, 'B', '2015-01-29', 3.1415629, 'iasdfhiasdhagdsnfgankkkjfgjklfgjklsklsklsdfg');
select * from tab;
-- now add some unique indexes
create unique index tab_ui1 on tab (id);
create unique index tab_ui2 on tab (nm);
create unique index tab_ui3 on tab (dt);
create unique index tab_ui4 on tab (qnt);
create unique index tab_ui5 on tab (descr);
create unique index tab_ui6 on tab (id, nm, dt, qnt, descr);
-- check that the indexes exist
select i.table_id, t.name as table_name, i.id as index_id, i.name as index_name, i.type from sys.idxs i join sys.tables t on i.table_id = t.id where t.name = 'tab';
-- test if uniqueness is enforced
insert into tab (id, nm, dt, qnt, descr) select id, nm, dt, qnt, descr from tab;
-- 2 rows inserted !!
select * from tab;
-- 4 rows
-- check for duplicates rows:
select count(*) as count_dupl_rows, id, nm, dt, qnt, descr from tab
group by id, nm, dt, qnt, descr having count(*) > 1
order by id, nm, dt, qnt, descr;
-- 2 rows (id = 1 and 2) exists twice.
### Actual Results:
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.20.0 (unreleased), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>create table tab (id int not null, nm varchar(30) not null, dt date, qnt dec(30,10), descr text);
operation successful (2.042ms)
sql>insert into tab (id, nm) values (1, 'A');
1 affected row (1.561ms)
sql>insert into tab (id, nm, dt, qnt, descr) values (2, 'B', '2015-01-29', 3.1415629, 'iasdfhiasdhagdsnfgankkkjfgjklfgjklsklsklsdfg');
1 affected row (1.600ms)
sql>select * from tab;
+------+------+------------+----------------------------------+----------------------------------------------+
| id | nm | dt | qnt | descr |
+======+======+============+==================================+==============================================+
| 1 | A | null | null | null |
| 2 | B | 2015-01-29 | 3.1415629000 | iasdfhiasdhagdsnfgankkkjfgjklfgjklsklsklsdfg |
+------+------+------------+----------------------------------+----------------------------------------------+
2 tuples (1.534ms)
sql>create unique index tab_ui1 on tab (id);
operation successful (1.800ms)
sql>create unique index tab_ui2 on tab (nm);
operation successful (1.599ms)
sql>create unique index tab_ui3 on tab (dt);
operation successful (1.618ms)
sql>create unique index tab_ui4 on tab (qnt);
operation successful (1.526ms)
sql>create unique index tab_ui5 on tab (descr);
operation successful (1.611ms)
sql>create unique index tab_ui6 on tab (id, nm, dt, qnt, descr);
operation successful (4.059ms)
sql>select i.table_id, t.name as table_name, i.id as index_id, i.name as index_name, i.type from sys.idxs i join sys.tables t on i.table_id = t.id where t.name = 'tab';
+----------+------------+----------+------------+------+
| table_id | table_name | index_id | index_name | type |
+==========+============+==========+============+======+
| 7280 | tab | 7282 | tab_ui1 | 0 |
| 7280 | tab | 7284 | tab_ui2 | 0 |
| 7280 | tab | 7286 | tab_ui3 | 0 |
| 7280 | tab | 7288 | tab_ui4 | 0 |
| 7280 | tab | 7290 | tab_ui5 | 0 |
| 7280 | tab | 7292 | tab_ui6 | 0 |
+----------+------------+----------+------------+------+
6 tuples (5.228ms)
sql>select * from tab;+------+------+------------+----------------------------------+----------------------------------------------+
| id | nm | dt | qnt | descr |
+======+======+============+==================================+==============================================+
| 1 | A | null | null | null |
| 2 | B | 2015-01-29 | 3.1415629000 | iasdfhiasdhagdsnfgankkkjfgjklfgjklsklsklsdfg |
+------+------+------------+----------------------------------+----------------------------------------------+
2 tuples (1.459ms)
sql>insert into tab (id, nm, dt, qnt, descr) select id, nm, dt, qnt, descr from tab;
2 affected rows (2.875ms)
sql>select * from tab;
+------+------+------------+----------------------------------+----------------------------------------------+
| id | nm | dt | qnt | descr |
+======+======+============+==================================+==============================================+
| 1 | A | null | null | null |
| 2 | B | 2015-01-29 | 3.1415629000 | iasdfhiasdhagdsnfgankkkjfgjklfgjklsklsklsdfg |
| 1 | A | null | null | null |
| 2 | B | 2015-01-29 | 3.1415629000 | iasdfhiasdhagdsnfgankkkjfgjklfgjklsklsklsdfg |
+------+------+------------+----------------------------------+----------------------------------------------+
4 tuples (0.905ms)
sql>select count(*) as count_dupl_rows, id, nm, dt, qnt, descr from tab
more>group by id, nm, dt, qnt, descr having count(*) > 1
more>order by id, nm, dt, qnt, descr;
+------+------+------+------------+----------------------------------+----------------------------------------+
| coun | id | nm | dt | qnt | descr |
: t_du : : : : : :
: pl_r : : : : : :
: ows : : : : : :
+======+======+======+============+==================================+========================================+
| 2 | 1 | A | null | null | null |
| 2 | 2 | B | 2015-01-29 | 3.1415629000 | iasdfhiasdhagdsnfgankkkjfgjklfgjklskls |
: : : : : : klsdfg :
+------+------+------+------------+----------------------------------+----------------------------------------+
2 tuples (3.299ms)
sql>
### Expected Results:
the insert should fail with an error message that uniqueness constraint xyz is violated
## Comment 20640
Date: 2015-02-08 18:08:11 +0100
From: @njnes
indices form a basis for checking the constraints, not the other way around.
## Comment 20654
Date: 2015-02-12 16:33:31 +0100
From: Martin van Dinther <<martin.van.dinther>>
As discussed the semantics of a UNIQUE INDEX should be consistent with other RDBMS behavior as SQL users expect it.
A "quick" implementation could be to generate the ALTER TABLE <tbl_nm> ADD CONSTRAINT <uc_nm> UNIQUE (<col_nm_list>) statement and execute it as part of the CREATE UNIQUE INDEX execution.
Alternatively if we decide to not enforce UNIQUE INDEX constraint, we should issue a warning when the user does a CREATE UNIQUE INDEX ... stating that the UNIQUE keyword is not enforced and the user should use ALTER TABLE <tbl_nm> ADD CONSTRAINT <uc_nm> UNIQUE (<col_nm_list>) instead.
| UNIQUE INDEX constraint is not enforced during inserts and updates | https://api.github.com/repos/MonetDB/MonetDB/issues/3668/comments | 0 | 2020-11-30T12:59:28Z | 2024-06-28T07:19:44Z | https://github.com/MonetDB/MonetDB/issues/3668 | 753,458,298 | 3,668 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-02-05 16:01:18 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2015-05-07 12:38:06 +0200
## Comment 20634
Date: 2015-02-05 16:01:18 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:35.0) Gecko/20100101 Firefox/35.0
Build Identifier:
mserver5: ../../dev/gdk/gdk_bat.c:2222: BATseqbase: Assertion `o <= oid_nil' failed.
/export/scratch1/dinther/INSTALL/bin/mserver5 -d10 --set embedded_r=true --set mapi_port=41000
Also it appears that after the abort, a restart of the mserver5 keeps on aborting with the assertion failure. Probably the database is corrupted.
Only a complete removal of the INSTALL/var/monetdb5/dbfarm/demo directory allowed me to restart the mserver5.
Reproducible: Always
### Steps to Reproduce:
1. Start mserver5 (MonetDB 5 server v11.20.0, compiled from default branch on 5 febr 2015)
2. Start mclient
3. enter SQL commands:
sql>CREATE TABLE OID_TBL(f1 oid);
operation successful (8.883ms)
sql>INSERT INTO OID_TBL(f1) VALUES (-1040);
sql>SELECT * FROM OID_TBL;
### Actual Results:
bash-4.2$ mclient -p 41000
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.20.0 (unreleased), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>CREATE TABLE OID_TBL(f1 oid);
operation successful (8.883ms)
sql>INSERT INTO OID_TBL(f1) VALUES (-1040);
sql>SELECT * FROM OID_TBL;
bash-4.2$
bash-4.2$ ./start_INSTALL_mserver5.sh
builtin opt gdk_dbpath = /export/scratch1/dinther/INSTALL/var/monetdb5/dbfarm/demo
builtin opt gdk_debug = 0
builtin opt gdk_vmtrim = no
builtin opt monet_prompt = >
builtin opt monet_daemon = no
builtin opt mapi_port = 50000
builtin opt mapi_open = false
builtin opt mapi_autosense = false
builtin opt sql_optimizer = default_pipe
builtin opt sql_debug = 0
cmdline opt embedded_r = true
cmdline opt mapi_port = 41000
cmdline opt gdk_debug = 10
MonetDB 5 server v11.20.0
This is an unreleased version
Serving database 'demo', using 8 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs and 128bit integers dynamically linked
Found 15.590 GiB available main-memory.
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2015 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Listening for connection requests on mapi:monetdb://127.0.0.1:41000/
MonetDB/GIS module loaded
MonetDB/SQL module loaded
MonetDB/R module loaded
> SQL catalog created, loading sql scripts once
loading sql script: 09_like.sql
loading sql script: 10_math.sql
loading sql script: 11_times.sql
loading sql script: 12_url.sql
loading sql script: 13_date.sql
loading sql script: 14_inet.sql
loading sql script: 15_querylog.sql
loading sql script: 16_tracelog.sql
loading sql script: 17_temporal.sql
loading sql script: 19_cluster.sql
loading sql script: 20_vacuum.sql
loading sql script: 21_dependency_functions.sql
loading sql script: 22_clients.sql
loading sql script: 23_skyserver.sql
loading sql script: 24_zorder.sql
loading sql script: 25_debug.sql
loading sql script: 26_sysmon.sql
loading sql script: 39_analytics.sql
loading sql script: 39_analytics_hge.sql
loading sql script: 40_geom.sql
loading sql script: 40_json.sql
loading sql script: 40_json_hge.sql
loading sql script: 41_jsonstore.sql
loading sql script: 45_uuid.sql
loading sql script: 46_gsl.sql
loading sql script: 72_fits.sql
loading sql script: 75_storagemodel.sql
loading sql script: 80_statistics.sql
loading sql script: 80_udf.sql
loading sql script: 80_udf_hge.sql
loading sql script: 85_bam.sql
loading sql script: 89_generator_hge.sql
loading sql script: 90_generator.sql
loading sql script: 99_system.sql
mserver5: ../../dev/gdk/gdk_bat.c:2222: BATseqbase: Assertion `o <= oid_nil' failed.
./start_INSTALL_mserver5.sh: line 5: 9603 Aborted /export/scratch1/dinther/INSTALL/bin/mserver5 -d10 --set embedded_r=true --set mapi_port=41000
When I try to restart the mserver5 I get:
bash-4.2$ ./start_INSTALL_mserver5.sh
builtin opt gdk_dbpath = /export/scratch1/dinther/INSTALL/var/monetdb5/dbfarm/demo
builtin opt gdk_debug = 0
builtin opt gdk_vmtrim = no
builtin opt monet_prompt = >
builtin opt monet_daemon = no
builtin opt mapi_port = 50000
builtin opt mapi_open = false
builtin opt mapi_autosense = false
builtin opt sql_optimizer = default_pipe
builtin opt sql_debug = 0
cmdline opt embedded_r = true
cmdline opt mapi_port = 41000
cmdline opt gdk_debug = 10
MonetDB 5 server v11.20.0
This is an unreleased version
Serving database 'demo', using 8 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs and 128bit integers dynamically linked
Found 15.590 GiB available main-memory.
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2015 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Listening for connection requests on mapi:monetdb://127.0.0.1:41000/
MonetDB/GIS module loaded
mserver5: ../../dev/gdk/gdk_bat.c:2222: BATseqbase: Assertion `o <= oid_nil' failed.
./start_INSTALL_mserver5.sh: line 5: 9750 Aborted /export/scratch1/dinther/INSTALL/bin/mserver5 -d10 --set embedded_r=true --set mapi_port=41000
bash-4.2$
### Expected Results:
no assertion failure
also it appears that after the abort, a restart of the mserver5 keeps on aborting with the assertion failure
negative oid values should not be accepted, an error msg should be returned
## Comment 20646
Date: 2015-02-09 22:39:37 +0100
From: @njnes
fixed, added new class of positive numeric types (oid).
| insert of negative value for oid column aborts mserver5 process with assertion failure | https://api.github.com/repos/MonetDB/MonetDB/issues/3667/comments | 0 | 2020-11-30T12:59:26Z | 2024-06-27T12:06:23Z | https://github.com/MonetDB/MonetDB/issues/3667 | 753,458,275 | 3,667 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-02-05 12:26:39 +0100
From: Asif Khan Suri <<asif.suri>>
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
Last updated: 2015-05-07 12:38:08 +0200
## Comment 20632
Date: 2015-02-05 12:26:39 +0100
From: Asif Khan Suri <<asif.suri>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:38.0) Gecko/20100101 Firefox/38.0
Build Identifier:
select cast(textip as inet) from iptable; destroying the the table by truncating the column textip to first part of the ip. Additionally from same ip stored in text column, only first occurrence results into inet rest appears as nil.
I am doing the binary copy from the files where no support for inet column there fore binary loading the ips as text into landing table. while inserting to final table where ip is defined as inet same behavior is happening. in SP2 transfer is not happening any more. and giving parser error of empty string where no text column is not empty.
Reproducible: Always
### Steps to Reproduce:
1. create table iptable with textip column as varchar(20).
2. insert '192.168.0.1' three times and 192.168.10.2 one times.
3. select cast(textip as inet) from iptable
### Actual Results:
192.168.0.1
nil
nil
192.168.10.2
### Expected Results:
192.168.0.1
192.168.0.1
192.168.0.1
192.168.10.2
loss of original column values
## Comment 20633
Date: 2015-02-05 12:33:14 +0100
From: Asif Khan Suri <<asif.suri>>
in SP2 its giving an error "ERROR: GDKerror:!ERROR: Error while parsing, unexpected string ''" and truncating the column.
## Comment 20635
Date: 2015-02-05 17:22:04 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [dedb4c1b64ce](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=dedb4c1b64ce) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=dedb4c1b64ce](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=dedb4c1b64ce)
Changeset description:
Do not modify input arguments.
This fixes bug #3666.
| casting text column to inet truncating text column and resulting inet for first occurrence only | https://api.github.com/repos/MonetDB/MonetDB/issues/3666/comments | 0 | 2020-11-30T12:59:23Z | 2024-06-27T12:06:22Z | https://github.com/MonetDB/MonetDB/issues/3666 | 753,458,247 | 3,666 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-02-05 08:02:39 +0100
From: sorear
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.19.7 (Oct2014-SP1)
CC: @njnes
Last updated: 2015-05-07 12:37:28 +0200
## Comment 20630
Date: 2015-02-05 08:02:39 +0100
From: sorear
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/40.0.2214.94 Safari/537.36
Build Identifier:
If I run a long query in one terminal session, and a very short one in another, MonetDB seems to wait for the long query to finish before running the short query. In particular, I can't get useful results from sys.queue() because by the time sys.queue starts, it's the only running query and reports only on itself(?).
Reproducible: Always
### Steps to Reproduce:
1. start two mclients in different terminals. In the first:
Welcome to mclient, the MonetDB/SQL interactive terminal (Oct2014-SP1)
Database: MonetDB v11.19.7 (Oct2014-SP1), '...'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>create table blah as select * from sys.generate_series(10000,13000) with data;
operation successful (2.098ms)
second:
Welcome to mclient, the MonetDB/SQL interactive terminal (Oct2014-SP1)
Database: MonetDB v11.19.7 (Oct2014-SP1), 'mapi:monetdb://sorear1.dev.gudtech.com:50000/gt'
Type \q to quit, \? for a list of available commands
auto commit mode: on
first, issue a long-running command:
sql>select count(distinct a.value * b.value) from blah a, blah b;
second attempt to see this command:
sql>select * from sys.queue();
several seconds later, this appears in the first:
+---------+
| L1 |
+=========+
| 4072685 |
+---------+
1 tuple (9.0s)
simultaneously in the second:
+------+---------+----------------------------+----------------------------+----------+---------+------------+----------------------------+
| qtag | user | started | estimate | progress | status | tag | query |
+======+=========+============================+============================+==========+=========+============+============================+
| 40 | monetdb | 2015-02-05 06:55:21.000000 | null | null | running | 12783204@0 | select * from sys.queue(); |
+------+---------+----------------------------+----------------------------+----------+---------+------------+----------------------------+
1 tuple (2.5s)
when not busy, queue requests finish instantly:
sql>select * from sys.queue();
+------+---------+----------------------------+----------------------------+----------+---------+------------+----------------------------+
| qtag | user | started | estimate | progress | status | tag | query |
+======+=========+============================+============================+==========+=========+============+============================+
| 41 | monetdb | 2015-02-05 06:55:24.000000 | null | null | running | 12783204@0 | select * from sys.queue(); |
+------+---------+----------------------------+----------------------------+----------+---------+------------+----------------------------+
1 tuple (0.638ms)
### Expected Results:
I'm not certain this is truly a bug, but it'd be very very useful if I could observe concurrently running queries with select * from sys.queue() or something functionally equivalent.
## Comment 20631
Date: 2015-02-05 08:04:52 +0100
From: sorear
This particular example is on a VM with 2 cores; I suspect that makes a difference here.
## Comment 20714
Date: 2015-03-13 14:08:52 +0100
From: @sjoerdmullender
The problem is caused by changeset [e79b2edeba71](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e79b2edeba71) from January 2014.
That changeset introduced a semaphore to limit concurrency. But this was done in such a way that no new queries enter the fray if the maximum level of concurrency (currently set at half the number of cores) is reached.
## Comment 20715
Date: 2015-03-13 14:34:24 +0100
From: @sjoerdmullender
(In reply to comment 2)
> The problem is caused by changeset [e79b2edeba71](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e79b2edeba71) from January 2014.
Make that Jan 2013.
## Comment 20716
Date: 2015-03-13 18:14:32 +0100
From: sorear
Interesting. We've been seeing behavior on a couple of 16 core production machines where the system is under sustained MonetDB load (mserver5 consistently > 500% CPU), but when I run "select * from sys.queue" it waits several seconds and then reports that the select is the only thing running. It sounds like the example I included may not have been an accurate reproduction then?
## Comment 20717
Date: 2015-03-16 14:51:31 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [68156d38a8c4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=68156d38a8c4) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=68156d38a8c4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=68156d38a8c4)
Changeset description:
Don't limit incoming queries based on number of cores.
This fixes bug #3665.
## Comment 20794
Date: 2015-04-14 12:35:56 +0200
From: @sjoerdmullender
The limiting semaphore has been removed. The problem should not occur anymore.
| inter-session starvation issue, particularly affects sys.queue | https://api.github.com/repos/MonetDB/MonetDB/issues/3665/comments | 0 | 2020-11-30T12:59:21Z | 2024-06-27T12:06:21Z | https://github.com/MonetDB/MonetDB/issues/3665 | 753,458,221 | 3,665 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-02-04 15:52:24 +0100
From: beatnut <<dwustlich>>
To: buildtools devs <<bugs-buildtools>>
Version: 11.19.9 (Oct2014-SP2)
Last updated: 2015-05-07 12:37:38 +0200
## Comment 20627
Date: 2015-02-04 15:52:24 +0100
From: beatnut <<dwustlich>>
a make with jsonstore-feature enabled breaks build-process with:
rest_jsonstore_handle_get.c:72:2: error: too few arguments to function 'SQLstatementIntern'
suggested fix (a quick guess only):
change line 72 from:
qmsg = SQLstatementIntern(c, &query, "rest", TRUE, TRUE, resultbuffer);
to
qmsg = SQLstatementIntern(c, &query, "rest", TRUE, TRUE, resultbuffer);
then it compiles without error.
## Comment 20628
Date: 2015-02-04 16:19:10 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [10db29917017](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=10db29917017) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=10db29917017](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=10db29917017)
Changeset description:
SQLstatementIntern grew an extra parameter at some point.
This fixes bug #3664.
I don't have the necessary files to test the mseed change.
## Comment 20629
Date: 2015-02-04 16:19:36 +0100
From: @sjoerdmullender
Thanks for reporting.
| SQLstatementIntern missing parameter when using jsonstore | https://api.github.com/repos/MonetDB/MonetDB/issues/3664/comments | 0 | 2020-11-30T12:59:18Z | 2024-06-27T12:06:20Z | https://github.com/MonetDB/MonetDB/issues/3664 | 753,458,189 | 3,664 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-02-03 00:47:20 +0100
From: @yzchang
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
CC: @njnes
Last updated: 2015-05-07 12:37:57 +0200
## Comment 20621
Date: 2015-02-03 00:47:20 +0100
From: @yzchang
In the following query, two approached, which essentially does the same computation, are used to update the values of table t2. The 1st approach (see the 1st UPDATE statement) uses a subquery, and produces the wrong result. The 2nd approach (see the 2nd UPDATE statement) first stored the results of the subquery in table t3, and produces the correct result.
CREATE TABLE t1 (id VARCHAR(48), col1 VARCHAR(32), col2 VARCHAR(8), excepted BOOLEAN);
INSERT INTO t1 (id, col1, col2, excepted) VALUES ('12', 'col1.A.99.code', '.03', 'false'), ('12', 'col1.A.99.code', '.02', 'false');
CREATE TABLE t2 (id VARCHAR(48), col3 VARCHAR(32), col2 VARCHAR(8), row int);
INSERT INTO t2 (id, col3, col2, row) VALUES ('12',null,null,null);
UPDATE t2 SET (col3, col2, row) =
(SELECT col1, col2, row
FROM (
SELECT id, col1, col2,
ROW_NUMBER() OVER (PARTITION BY id ORDER BY (col1 LIKE '%.%.99.%') ASC,
col1 ASC, col2 ASC) AS row
FROM t1
WHERE excepted = false
AND col1 LIKE '%.A.%'
) AS t3
WHERE t3.row = 1
AND t2.id= t3.id
);
SELECT * FROM t2;
DROP TABLE t2;
CREATE TABLE t2 (id VARCHAR(48), col3 VARCHAR(32), col2 VARCHAR(8), row int);
INSERT INTO t2 (id, col3, col2, row) VALUES ('12',null,null,null);
CREATE TABLE t3 AS (
SELECT id, col1, col2,
ROW_NUMBER() OVER (PARTITION BY id ORDER BY (col1 LIKE '%.%.99.%') ASC,
col1 ASC,col2 ASC) AS row
FROM t1
WHERE excepted = false AND col1 LIKE '%.A.%'
) WITH DATA;
UPDATE t2 SET (col3, col2, row) = (SELECT col1, col2, row FROM t3 WHERE t2.id= t3.id AND t3.row = 1);
SELECT * FROM t2;
DROP TABLE t1;
DROP TABLE t2;
DROP TABLE t3;
## Comment 20622
Date: 2015-02-03 00:51:27 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [c90ba8342e40](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=c90ba8342e40) made by Jennie Zhang <y.zhang@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=c90ba8342e40](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=c90ba8342e40)
Changeset description:
Added test and expected output for bug #3663
## Comment 20638
Date: 2015-02-08 17:15:27 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [e7f28b96f180](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e7f28b96f180) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=e7f28b96f180](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=e7f28b96f180)
Changeset description:
fix bug #3663, ie keep the order information in the expression
## Comment 20639
Date: 2015-02-08 18:04:44 +0100
From: @njnes
fixed issue with relational optimizer losing the ordering information
| Incorrect result ROW_NUMBER in subquery | https://api.github.com/repos/MonetDB/MonetDB/issues/3663/comments | 0 | 2020-11-30T12:59:16Z | 2024-06-27T12:06:19Z | https://github.com/MonetDB/MonetDB/issues/3663 | 753,458,169 | 3,663 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-02-02 23:39:48 +0100
From: @yzchang
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
CC: b.kaptijn, @njnes
Last updated: 2015-05-07 12:38:12 +0200
## Comment 20620
Date: 2015-02-02 23:39:48 +0100
From: @yzchang
The following query crashes mserver:
CREATE TABLE een (a INTEGER, b INTEGER, C integer);
UPDATE een SET (A,B,C) = (SELECT 2,2,2);
## Comment 20623
Date: 2015-02-03 00:58:07 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [39ba5999a5ba](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=39ba5999a5ba) made by Jennie Zhang <y.zhang@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=39ba5999a5ba](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=39ba5999a5ba)
Changeset description:
added test and expected output for bug #3662
## Comment 20626
Date: 2015-02-04 14:29:18 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [79619c1f070b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=79619c1f070b) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=79619c1f070b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=79619c1f070b)
Changeset description:
fixed bug #3662, ie multi value updates
| UPDATE row with row value constructor crashes monetdb server | https://api.github.com/repos/MonetDB/MonetDB/issues/3662/comments | 0 | 2020-11-30T12:59:13Z | 2024-06-27T12:06:17Z | https://github.com/MonetDB/MonetDB/issues/3662 | 753,458,148 | 3,662 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-02-02 14:15:31 +0100
From: Richard Hughes <<richard.monetdb>>
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
Last updated: 2015-05-07 12:37:47 +0200
## Comment 20619
Date: 2015-02-02 14:15:31 +0100
From: Richard Hughes <<richard.monetdb>>
Created attachment 317
Generate monetdb-dbg.deb package
It's generally helpful for users to have access to debug symbols for the pre-built binaries downloaded from the MonetDB web site. It allows not-particularly-technical users to generate more useful crash reports without needing to go through the hassle of building from source. Debug symbols are also good for profiling tools.
The attached patch does this for the .deb packages, following the instructions at https://wiki.debian.org/DebugPackage. I'm not personally familiar with the other deployment platforms you offer.
> Attached file: [monetdb-dbg-package.diff](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3661_monetdb-dbg-package.diff_317) (text/plain, 1245 bytes)
> Description: Generate monetdb-dbg.deb package
## Comment 20624
Date: 2015-02-03 11:12:26 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [b79f3dc67f49](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=b79f3dc67f49) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=b79f3dc67f49](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=b79f3dc67f49)
Changeset description:
Also provide debug packages for Debian and Ubuntu.
This fixes bug #3661.
## Comment 20625
Date: 2015-02-03 11:13:24 +0100
From: @sjoerdmullender
Thanks for the suggestion and the patch. I have applied the patch.
| Ship debug symbols for pre-built binaries | https://api.github.com/repos/MonetDB/MonetDB/issues/3661/comments | 0 | 2020-11-30T12:59:11Z | 2024-06-27T12:06:16Z | https://github.com/MonetDB/MonetDB/issues/3661 | 753,458,117 | 3,661 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-02-02 00:56:26 +0100
From: K <<mdb>>
To: SQL devs <<bugs-sql>>
Version: 11.19.9 (Oct2014-SP2)
Last updated: 2015-05-07 12:38:09 +0200
## Comment 20615
Date: 2015-02-02 00:56:26 +0100
From: K <<mdb>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:35.0) Gecko/20100101 Firefox/35.0
Build Identifier:
A problem remains in the implementation of comparison operators on the inet datatype. Depending on the actual values for the data, the results may not be correct. Perhaps this is related to the same issue as 3646?
Reproducible: Always
### Steps to Reproduce:
create table "bug"
(
address inet
);
insert into "bug" values('1.0.0.11');
insert into "bug" values('2.0.0.10');
insert into "bug" values('3.0.0.9');
insert into "bug" values('4.0.0.8');
insert into "bug" values('5.0.0.7');
insert into "bug" values('6.0.0.6');
insert into "bug" values('7.0.0.5');
insert into "bug" values('8.0.0.4');
insert into "bug" values('9.0.0.3');
insert into "bug" values('10.0.0.2');
insert into "bug" values('11.0.0.1');
select count(*) from bug where address>=inet'4.0.0.0';
select count(*) from bug where address<inet'8.0.0.0';
select count(*) from bug where address>=inet'4.0.0.0' and address<inet'8.0.0.0';
### Actual Results:
8
6 <-- wrong
3 <-- wrong
### Expected Results:
8
7
4
Running the same test case on Oct2014-SP1 gave different (but still wrong) results:
11
0
0
Also note that the results do depend on the data. If instead of the values given above, you use:
1.0.0.0
2.0.0.0
3.0.0.0
4.0.0.0
5.0.0.0
6.0.0.0
7.0.0.0
8.0.0.0
9.0.0.0
10.0.0.0
11.0.0.0
the results are correct!
8
7
4
So as with 3646 I suspect that this may be a byte order problem.
I'm marking this "normal" because I'm not sure if it qualifies as "major", but this bug is fatal for anyone working with the 'inet' datatype because it causes monetdb to return incorrect results. It would almost be better if the server crashed, since at least then you'd know something was wrong! It would be great if a patch were made available, since this otherwise no queries (at least using comparisons on this datatype) can be done until the next release.
As requested here's the output from mserver5 --version:
MonetDB 5 server v11.19.9 "Oct2014-SP2" (64-bit, 64-bit oids)
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2015 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 31.3GiB available memory, 8 available cpu cores
Libraries:
libpcre: 8.33 2013-05-28 (compiled with 8.33)
openssl: OpenSSL 1.0.1e 11 Feb 2013 (compiled with )
libxml2: 2.9.1 (compiled with 2.9.1)
Compiled by: mockbuild@ (x86_64-redhat-linux-gnu)
Compilation: gcc -O3 -fomit-frame-pointer -pipe -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches -m64 -mtune=generic -D_FORTIFY_SOURCE=2
Linking : /usr/bin/ld -m elf_x86_64 -Wl,-z,relro
## Comment 20616
Date: 2015-02-02 00:59:08 +0100
From: K <<mdb>>
Created attachment 316
Test case (same as described in bug report)
> Attached file: [bug3660.testcase](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3660_bug3660.testcase_316) (text/plain, 641 bytes)
> Description: Test case (same as described in bug report)
## Comment 20617
Date: 2015-02-02 09:34:58 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [5890ce4c8682](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=5890ce4c8682) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=5890ce4c8682](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=5890ce4c8682)
Changeset description:
Don't use ATOMstorage if nil or compare function is different.
This fixes bug #3660.
Also added test for the bug.
| Incorrect Results for Comparison Operators on inet Datatype | https://api.github.com/repos/MonetDB/MonetDB/issues/3660/comments | 0 | 2020-11-30T12:59:08Z | 2024-06-27T12:06:15Z | https://github.com/MonetDB/MonetDB/issues/3660 | 753,458,089 | 3,660 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-01-31 14:32:44 +0100
From: yzan
To: Documentation maintainers <<bugs-docs>>
Version: unspecified
CC: @njnes
Last updated: 2015-04-14 13:07:30 +0200
## Comment 20614
Date: 2015-01-31 14:32:44 +0100
From: yzan
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:24.0) Gecko/20100101 Firefox/24.0
Build Identifier:
I downloaded VOC dataset from:
https://www.monetdb.org/Documentation/Manuals/SQLreference/Programming/JDBC
and I noticed that there are dates from future.
Reproducible: Always
### Steps to Reproduce:
select arrival_date
from "voyages"
where arrival_date is not null
order by arrival_date desc
limit 5
### Actual Results:
21762-06-26
11788-04-29
1937-09-17
1795-10-31
1795-10-03
### Expected Results:
1762-06-26
1788-04-29
1737-09-17
1795-10-31
1795-10-03
## Comment 20618
Date: 2015-02-02 10:08:31 +0100
From: @njnes
indeed the two incorrect dates are in the data, we should correct the voc example
## Comment 20795
Date: 2015-04-14 13:07:30 +0200
From: @sjoerdmullender
I have replaced all versions of the VOC data that I could find with corrected versions.
| Low data quality of the example database | https://api.github.com/repos/MonetDB/MonetDB/issues/3659/comments | 0 | 2020-11-30T12:59:06Z | 2024-06-27T12:06:14Z | https://github.com/MonetDB/MonetDB/issues/3659 | 753,458,057 | 3,659 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-01-30 09:03:07 +0100
From: Dann Corbit <<danncorbit>>
To: clients devs <<bugs-clients>>
Version: 11.19.9 (Oct2014-SP2)
CC: @hannesmuehleisen
Last updated: 2015-01-30 10:59:25 +0100
## Comment 20612
Date: 2015-01-30 09:03:07 +0100
From: Dann Corbit <<danncorbit>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:25.2) Gecko/20150112 Firefox/31.9 PaleMoon/25.2.0
Build Identifier:
Filename: libmonetodbcs.dll
Threat name: Suspicious.Cloud.7.F
Full Path: c:\program files (x86)\monetdb\monetdb odbc driver\lib\libmonetodbcs.dll
____________________________
Details
Very Few Users, Mature, Risk High
Origin
Downloaded from
Unknown
Activity
Actions performed: Actions performed: 1
____________________________
On computers as of
1/29/2015 at 11:50:08 PM
Last Used
1/29/2015 at 11:50:08 PM
Startup Item
No
Launched
No
____________________________
Very Few Users
Fewer than 5 users in the Norton Community have used this file.
Mature
This file was released more than 31 days 1 month ago.
High
This file risk is high.
Threat type: Heuristic Virus. Detection of a threat based on malware heuristics.
____________________________
Source: External Media
____________________________
File Actions
File: c:\program files (x86)\MonetDB\monetdb odbc driver\lib\ libmonetodbcs.dll Blocked
____________________________
File Thumbprint - SHA:
b3a8683a6a6e5982d32b2c63b8ef89547ffdf18eaa75fd078400c1e2947b53e1
File Thumbprint - MD5:
Not available
Reproducible: Always
### Steps to Reproduce:
Using the MSI file MonetDB-ODBC-Installer-i386-20150123.msi, install onto Windows 7 Home Premium Version 6.1 (Build 7601: Service Pack 1) with Norton Security Suite:
Version: 21.6.0.32
End Point ID: {14F256C8-63EA-47A1-9138-154A5361ACA8}
LiveUpdate Channel: Traditional
Current SKU: 21317287
Family SKU: 21291129
Media SKU: 21317287
Vendor Name: Comcast
### Actual Results:
Auto Protect is processing security risk Suspicious.Cloud.7.f.
Auto Protect deleted security risk Suspicious.Cloud.7.f. Your computer is secure.
Error message from the MSI installer:
There is a problem with this Windows Installer Package. A program run as part of the setup did not finish as expected. Contact your support personnel or package vendor.
### Expected Results:
Correctly installed the driver.
The 64 bit driver does install as expected.
## Comment 20613
Date: 2015-01-30 10:59:25 +0100
From: @hannesmuehleisen
Reported to Norton as a false positive.
| Norton Security Suite thinks that the 32 bit odbc installer is a virus. Report included below | https://api.github.com/repos/MonetDB/MonetDB/issues/3658/comments | 0 | 2020-11-30T12:59:03Z | 2024-06-28T13:15:34Z | https://github.com/MonetDB/MonetDB/issues/3658 | 753,458,026 | 3,658 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-01-28 05:20:20 +0100
From: Roman <<kislenok.roman>>
To: Merovingian devs <<bugs-merovingian>>
Version: 11.21.19 (Jul2015-SP4)
Last updated: 2018-04-24 20:59:36 +0200
## Comment 20588
Date: 2015-01-28 05:20:20 +0100
From: Roman <<kislenok.roman>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/39.0.2171.95 Safari/537.36
Build Identifier:
There is cool parameter for mserver5 called --dbextra, but there is no way to set this parameter using monetdbd or monetdb.
Reproducible: Always
### Steps to Reproduce:
1. try to set --dbextra using monetdbd
### Actual Results:
no such parameter.
### Expected Results:
dbextra for mserver5 is used
## Comment 26406
Date: 2018-04-24 20:59:36 +0200
From: @sjoerdmullender
This feature was implemented in the Jul2017 (11.27.1) release.
| There is no way to pass --dbextra parameter from monetdbd to mserver5 | https://api.github.com/repos/MonetDB/MonetDB/issues/3657/comments | 0 | 2020-11-30T12:59:01Z | 2024-06-27T12:06:12Z | https://github.com/MonetDB/MonetDB/issues/3657 | 753,457,997 | 3,657 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-01-27 08:30:17 +0100
From: @bartscheers
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
CC: @njnes
Last updated: 2015-05-07 12:37:51 +0200
## Comment 20581
Date: 2015-01-27 08:30:17 +0100
From: @bartscheers
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:34.0) Gecko/20100101 Firefox/34.0
Build Identifier:
When one executes the fitsload() function, the server returns with an (error) message:
invalid/unknown response from server, ignoring output
following 5 lines of comments about loaded columns and finishes with:
(remaining output omitted, use \fraw to examine in detail).
However, the loading was successful (and very fast!), and the loaded table can be queried etc.
Reproducible: Always
### Steps to Reproduce:
1. Use example fits file: call fitsattach('/myfits/r564532_cat.fits');
2. call fitsload('r564532_cat_3');
3.
### Actual Results:
invalid/unknown response from server, ignoring output
? Loading 1864 rows in table apm-binarytable
? Column number loaded for 1 ms Total 1 ms
? Column isophotal_flux loaded for 0 ms Total 0 ms
? Column total_flux loaded for 0 ms Total 0 ms
? Column core_flux loaded for 1 ms Total 1 ms
(remaining output omitted, use \fraw to examine in detail)
### Expected Results:
Maybe the number of loaded records.
Relation with bug #3654.
https://www.monetdb.org/bugzilla/show_bug.cgi?id=3654
## Comment 20582
Date: 2015-01-27 08:40:16 +0100
From: @bartscheers
Using this from within a python call, it breaks the program:
File "lib/python2.7/site-packages/monetdb/mapi.py", line 196, in cmd
raise ProgrammingError("unknown state: %s" % response)
monetdb.exceptions.ProgrammingError: unknown state: Loading 1864 rows in table apm-binarytable
Column number loaded for 0 ms Total 0 ms
Column isophotal_flux loaded for 0 ms Total 0 ms
Column total_flux loaded for 1 ms Total 1 ms
etc.
## Comment 20644
Date: 2015-02-09 10:31:31 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [5bd33e3ec650](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=5bd33e3ec650) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=5bd33e3ec650](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=5bd33e3ec650)
Changeset description:
Don't send debugging output to client.
This should fix bug #3656.
## Comment 20649
Date: 2015-02-11 16:33:46 +0100
From: @njnes
Bart could you verify the fix works for you.
## Comment 20650
Date: 2015-02-11 17:37:54 +0100
From: @bartscheers
Yes, this works !
| error message after calling fitsload() | https://api.github.com/repos/MonetDB/MonetDB/issues/3656/comments | 0 | 2020-11-30T12:58:58Z | 2024-06-27T12:06:11Z | https://github.com/MonetDB/MonetDB/issues/3656 | 753,457,971 | 3,656 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-01-26 19:42:21 +0100
From: Anthony Damico <<ajdamico>>
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
CC: @njnes
Last updated: 2015-05-07 12:38:02 +0200
## Comment 20578
Date: 2015-01-26 19:42:21 +0100
From: Anthony Damico <<ajdamico>>
User-Agent: Mozilla/5.0 (Windows NT 6.3; WOW64; rv:35.0) Gecko/20100101 Firefox/35.0
Build Identifier:
fully documented here
https://stackoverflow.com/questions/28120224/monetdb-or-clause?noredirect=1comment44677898_28120224
Reproducible: Always
### Steps to Reproduce:
Welcome to mclient, the MonetDB/SQL interactive terminal (Jan2014-SP3)
Database: MonetDB v11.17.21 (Jan2014-SP3), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>
sql>
sql>CREATE TABLE employee
more>(
more> employee_id INTEGER NOT NULL,
more> manager_id INTEGER NULL
more>);
operation successful (38.382ms)
sql>
sql>INSERT INTO employee (employee_id,manager_id) values (50,70);
1 affected rows (18.525ms)
sql>INSERT INTO employee (employee_id,manager_id) values (60,70);
1 affected rows (1.100ms)
sql>INSERT INTO employee (employee_id,manager_id) values (70,null);
1 affected rows (1.083ms)
sql>
sql>select employee_id, manager_id
more>from employee
more>where employee_id = 60
more>and (manager_id in (-1) or -1 in (-1));
+-------------+------------+
| employee_id | manager_id |
+=============+============+
| 50 | 70 |
| 60 | 70 |
| 70 | null |
+-------------+------------+
3 tuples (1.628ms)
sql>
### Actual Results:
all three lines
### Expected Results:
i think it should just be the employee id == 60?
## Comment 20583
Date: 2015-01-27 08:52:14 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [c106814afbd5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=c106814afbd5) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=c106814afbd5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=c106814afbd5)
Changeset description:
fixed bug in in / or handling. Fixes bug #3655.
## Comment 20584
Date: 2015-01-27 08:53:32 +0100
From: @njnes
fixed, the -1 in (-1) wasn't handled properly
| SQL WHERE -1 in (-1) issue? | https://api.github.com/repos/MonetDB/MonetDB/issues/3655/comments | 0 | 2020-11-30T12:58:56Z | 2024-06-27T12:06:10Z | https://github.com/MonetDB/MonetDB/issues/3655 | 753,457,944 | 3,655 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-01-26 16:02:02 +0100
From: @bartscheers
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
Last updated: 2015-08-28 13:43:05 +0200
## Comment 20571
Date: 2015-01-26 16:02:02 +0100
From: @bartscheers
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:34.0) Gecko/20100101 Firefox/34.0
Build Identifier:
cfitsio-devel libraries exist on the system. After installation and creating a new database (with monetdb create), one still needs to execute extra commands to enable the fits functionality.
The commands to be executed are:
create procedure listdir(dirname string) external name fits.listdir;
create procedure fitsattach(fname string) external name fits.attach;
create procedure fitsload(tname string) external name fits.load;
create procedure listdirpat(dirname string,pat string) external name fits.listdirpattern;
After this it works (from within mclient).
Reproducible: Always
### Steps to Reproduce:
1. monetdb create bg; monetdb release bg; monetdb start bg
2. mclient -dbg
3. sql> call fitsattach('/home/bscheers/Desktop/r564532_cat.fits');
SELECT: no such unary operator 'fitsattach(char)'
### Actual Results:
See above:
### Expected Results:
Fits file to be attached.
## Comment 20645
Date: 2015-02-09 10:35:25 +0100
From: @sjoerdmullender
This was fixed in changeset [048f9571b293](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=048f9571b293)
## Comment 21230
Date: 2015-08-28 13:43:05 +0200
From: @sjoerdmullender
Jul2015 has been released.
| configure --enable-fits requires extra commands after creating a database instance | https://api.github.com/repos/MonetDB/MonetDB/issues/3654/comments | 0 | 2020-11-30T12:58:53Z | 2024-06-27T12:06:09Z | https://github.com/MonetDB/MonetDB/issues/3654 | 753,457,917 | 3,654 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-01-20 11:21:12 +0100
From: @hannesmuehleisen
To: SQL devs <<bugs-sql>>
Version: 11.19.3 (Oct2014)
CC: @njnes
Last updated: 2015-05-07 12:38:13 +0200
## Comment 20562
Date: 2015-01-20 11:21:12 +0100
From: @hannesmuehleisen
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/39.0.2171.99 Safari/537.36
Build Identifier:
PREPARE SELECT id FROM tables WHERE name = ?;
works fine, output:
+---------+--------+-------+--------+--------+--------+
| type | digits | scale | schema | table | column |
+=========+========+=======+========+========+========+
| int | 32 | 0 | | tables | id |
| varchar | 1024 | 0 | null | null | null |
+---------+--------+-------+--------+--------+--------+
PREPARE SELECT id FROM tables WHERE name = LOWER(?);
crashes mserver (!),
Reproducible: Always
### Steps to Reproduce:
1. run PREPARE SELECT id FROM tables WHERE name = LOWER(?);
### Actual Results:
crash
### Expected Results:
No crash
## Comment 20564
Date: 2015-01-20 17:59:07 +0100
From: @hannesmuehleisen
Log says:
Assertion failed: (t1), function bind_func, file rel_select.c, line 1437.
## Comment 20573
Date: 2015-01-26 18:51:57 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [82edf3c91080](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=82edf3c91080) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=82edf3c91080](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=82edf3c91080)
Changeset description:
added test for bug #3653
## Comment 20574
Date: 2015-01-26 18:52:32 +0100
From: @njnes
fixed, ie added function lookup only on the function name
| PREPARE crashes mserver if unbound variable is function parameter | https://api.github.com/repos/MonetDB/MonetDB/issues/3653/comments | 0 | 2020-11-30T12:58:51Z | 2024-06-27T12:06:08Z | https://github.com/MonetDB/MonetDB/issues/3653 | 753,457,884 | 3,653 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-01-20 10:19:53 +0100
From: @drstmane
To: Documentation maintainers <<bugs-docs>>
Version: unspecified
Last updated: 2015-07-15 16:22:53 +0200
## Comment 20560
Date: 2015-01-20 10:19:53 +0100
From: @drstmane
The "copied" installation documentation under
https://www.monetdb.org/Documentation/UserGuide/Downloads/Fedora
https://www.monetdb.org/Documentation/UserGuide/Downloads/UbuntuDebian
https://www.monetdb.org/Documentation/UserGuide/Downloads/FreeBSD
https://www.monetdb.org/Documentation/UserGuide/Downloads/Mac
https://www.monetdb.org/Documentation/UserGuide/Downloads/Solaris
https://www.monetdb.org/Documentation/UserGuide/Downloads/Windows
as linked to from
https://www.monetdb.org/Downloads
and
https://www.monetdb.org/Documentation/Guide/Installation
is largely outdated, among others mentioning obsolete release "Jan2014-SP1" rather than the latest "Oct2014_SP-1" release, as well as listing old OS versions that are no longer supported, but lacking up-to-date OS versions that are supported.
The "genuine" (automatically generated and maintained) installation information under
http://dev.monetdb.org/downloads/
in detail under
http://dev.monetdb.org/downloads/Fedora/
http://dev.monetdb.org/downloads/deb/
http://dev.monetdb.org/downloads/epel/
http://dev.monetdb.org/downloads/FreeBSD/
http://dev.monetdb.org/downloads/MacOSX/
http://dev.monetdb.org/downloads/Solaris/
http://dev.monetdb.org/downloads/Windows/
is up-to-date.
IMHO, we should either find a way to update to former (copies) automatically with every release, or replace the former by direct links to the latter.
In the second case, we might want to consider, though to update the information under the latter with some of the extra information from the former.
## Comment 20647
Date: 2015-02-10 16:47:00 +0100
From: @sjoerdmullender
I have updated the left menu of the main Downloads page. I have also added text to the Windows download page from the page that now used to be referred to by the Windows link in said menu.
The MacOSX link still points to the old page since it has a lot of information, but all the others now point directly to the downloads area. So no longer to an old copy.
Please check.
## Comment 20648
Date: 2015-02-10 17:11:43 +0100
From: @sjoerdmullender
(In reply to comment 1)
> The MacOSX link still points to the old page since it has a lot of
> information, but all the others now point directly to the downloads area.
Now the MacOSX information has been copied to the downloads site and the link in the menu has been changed to point there.
## Comment 20658
Date: 2015-02-15 11:58:59 +0100
From: @drstmane
Hi Sjoerd,
thank you very much!
https://www.monetdb.org/Downloads look great to me, now, linking directly to the genuine download site & pages.
However,
the links under https://www.monetdb.org/Documentation/Guide/Installation --- both in the left-hand side context menu as well as at the bottom of the pape --- still point to the outdate copies of the download pages,
and "RHEL & CentOS" & "Java (JDBC)" are missing, there.
Best,
Stefan
## Comment 20663
Date: 2015-02-16 10:24:31 +0100
From: @sjoerdmullender
(In reply to comment 3)
> However,
> the links under https://www.monetdb.org/Documentation/Guide/Installation ---
> both in the left-hand side context menu as well as at the bottom of the pape
> --- still point to the outdate copies of the download pages,
> and "RHEL & CentOS" & "Java (JDBC)" are missing, there.
That's all because these pages are still in Drupal and are all part of a "book". I guess the next step is to actually remove the pages.
## Comment 21003
Date: 2015-07-15 16:22:53 +0200
From: @sjoerdmullender
(In reply to Sjoerd Mullender from comment 4)
> That's all because these pages are still in Drupal and are all part of a
> "book". I guess the next step is to actually remove the pages.
I have removed the book pages.
| Installation documentation outdated | https://api.github.com/repos/MonetDB/MonetDB/issues/3652/comments | 0 | 2020-11-30T12:58:48Z | 2024-06-27T12:06:07Z | https://github.com/MonetDB/MonetDB/issues/3652 | 753,457,855 | 3,652 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-01-20 10:09:14 +0100
From: @drstmane
To: Documentation maintainers <<bugs-docs>>
Version: unspecified
Last updated: 2015-07-15 17:04:10 +0200
## Comment 20559
Date: 2015-01-20 10:09:14 +0100
From: @drstmane
The list of provided binary packages under
https://www.monetdb.org/Documentation/Guide/Installation
is incomplete.
It lack (a link to) "RHEL & CentOS", which is present under
https://www.monetdb.org/Downloads
as well as under
http://dev.monetdb.org/downloads/
## Comment 21006
Date: 2015-07-15 17:04:10 +0200
From: @sjoerdmullender
With the removal of the subpages for bug #3652, this is no longer an issue.
| https://www.monetdb.org/Documentation/Guide/Installation: installation list incomplete | https://api.github.com/repos/MonetDB/MonetDB/issues/3651/comments | 0 | 2020-11-30T12:58:46Z | 2024-06-27T12:06:06Z | https://github.com/MonetDB/MonetDB/issues/3651 | 753,457,829 | 3,651 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-01-15 01:34:50 +0100
From: sorear
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
CC: @njnes
Last updated: 2015-05-07 12:38:03 +0200
## Comment 20557
Date: 2015-01-15 01:34:50 +0100
From: sorear
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/39.0.2171.95 Safari/537.36
Build Identifier:
$ mclient -d test -u monetdb
password:
Welcome to mclient, the MonetDB/SQL interactive terminal (Oct2014-SP1)
Database: MonetDB v11.19.7 (Oct2014-SP1), 'mapi:monetdb://redacted:50000/test'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>create table foo (id integer, bar_id integer);
operation successful (3.868ms)
sql>create table foo_bar (id integer);
operation successful (2.002ms)
sql>insert into foo (id, bar_id) values (1,2), (3,4);
2 affected rows (4.187ms)
sql>insert into foo_bar (id) values (5), (6);
2 affected rows (7.795ms)
sql>select * from foo;
+------+--------+
| id | bar_id |
+======+========+
| 1 | 2 |
| 3 | 4 |
+------+--------+
2 tuples (2.818ms)
sql>select * from foo_bar;
+------+
| id |
+======+
| 5 |
| 6 |
+------+
2 tuples (3.284ms)
sql>
$ monetdb stop test
stopping database 'test'... done
$ monetdb start test
starting database 'test'... done
$ mclient -d test -u monetdb
password:
Welcome to mclient, the MonetDB/SQL interactive terminal (Oct2014-SP1)
Database: MonetDB v11.19.7 (Oct2014-SP1), 'mapi:monetdb://redacted:50000/test'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>select * from foo;
+------+--------+
| id | bar_id |
+======+========+
| 1 | 2 |
| 3 | 4 |
+------+--------+
2 tuples (2.990ms)
sql>select * from foo_bar;
+------+
| id |
+======+
| 2 |
| 4 |
| 5 |
| 6 |
+------+
4 tuples (2.093ms)
sql>
There are two colums associated to the "sys_foo_bar_id" BAT, because _ is not a reserved character in identifiers. I'm working around this by being careful not to create columns with related names, but this behavior still strikes me as a bug. BAT name generation should concatenate with a reserved character, or escape underscores.
Possibly the best approach would be to not code the SQL-level column and table name at all but rather use the OID for the column - this could potentially allow for ALTER TABLE ... RENAME TO and ALTER TABLE ... RENAME COLUMN without having to get the GDK logger involved in the business of renaming BATs.
Reproducible: Always
### Actual Results:
Note that the contents of the foo_bar table have changed after the restart.
### Expected Results:
Kept the tables separate.
## Comment 20567
Date: 2015-01-25 18:26:07 +0100
From: @njnes
For the stable release added a test too prevent bat level name conflicts. For the new feature release we will look into using the sql object numbers.
## Comment 20568
Date: 2015-01-25 19:01:20 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [324d579fa720](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=324d579fa720) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=324d579fa720](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=324d579fa720)
Changeset description:
work-around for a bug #3635
(temporary) fix / improved check for bug #3650
## Comment 20572
Date: 2015-01-26 18:50:43 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [609f87915225](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=609f87915225) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=609f87915225](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=609f87915225)
Changeset description:
fixed bug #3650, lookup the unary function based on the function name only.
## Comment 20576
Date: 2015-01-26 19:12:07 +0100
From: sorear
I think that changeset was supposed to reference bug #3653, not this one.
| Naming of persistent BATs is fragile | https://api.github.com/repos/MonetDB/MonetDB/issues/3650/comments | 0 | 2020-11-30T12:58:44Z | 2024-06-27T12:06:05Z | https://github.com/MonetDB/MonetDB/issues/3650 | 753,457,806 | 3,650 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-01-14 23:41:36 +0100
From: sorear
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.19.7 (Oct2014-SP1)
CC: @mlkersten
Last updated: 2015-01-29 14:07:28 +0100
## Comment 20556
Date: 2015-01-14 23:41:36 +0100
From: sorear
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/39.0.2171.95 Safari/537.36
Build Identifier:
There was a crash in RECYCLEdrop on a mserver5 instance today, and I noticed that it's checking that recycleBlk is not null, taking a lock, and then assuming that recycleBlk is still not null - it could have been nulled before the lock-take succeeds.
There appear to be other locking issues in the recycler as the code in the reproduction section crashes in many places.
Reproducible: Always
### Steps to Reproduce:
Start and release a database (assumed "test" below), and:
for x in 1 2 ; do perl -MDBI -e 'my $db = DBI->connect("dbi:monetdb:database=test;host=localhost;port=50000", "monetdb", "monetdb"); for (1 .. 1000) { $db->do("SET OPTIMIZER='\'recycler_pipe\''"); $db->do("SELECT * FROM sys.schemas LIMIT 1"); $db->do("START TRANSACTION"); $db->do("COMMIT") }' & done
### Actual Results:
Segfaults almost immediately, but in an unpredictable place.
### Expected Results:
No segfaults.
## Comment 20558
Date: 2015-01-15 15:01:36 +0100
From: @mlkersten
Thank you for reporting. The Recycler optimizer is alpha-release code and any real-life use case would help its further maturing.
Martin
## Comment 20563
Date: 2015-01-20 14:18:36 +0100
From: @mlkersten
Applied the patch in RECYCLEDROP.
## Comment 20565
Date: 2015-01-20 18:14:23 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [22c1bf98c903](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=22c1bf98c903) made by Martin Kersten <mk@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=22c1bf98c903](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=22c1bf98c903)
Changeset description:
Bug #3649 reported fix.
Don't assume variables won't change in a concurrent situation.
## Comment 20597
Date: 2015-01-29 14:07:28 +0100
From: @sjoerdmullender
Oct2014-SP2 has been released.
| recycler crashes with concurrent transactions | https://api.github.com/repos/MonetDB/MonetDB/issues/3649/comments | 0 | 2020-11-30T12:58:41Z | 2024-06-27T12:06:04Z | https://github.com/MonetDB/MonetDB/issues/3649 | 753,457,779 | 3,649 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-01-12 21:20:39 +0100
From: sorear
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
CC: @njnes
Last updated: 2015-05-07 12:38:06 +0200
## Comment 20552
Date: 2015-01-12 21:20:39 +0100
From: sorear
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/39.0.2171.95 Safari/537.36
Build Identifier:
If a connection is dropped while temp tables are active, trans_drop_tmp will remove them from the active transaction without resetting tables.nelm on the relevant sql_schema object.
The next time that transaction is cleaned, reset_schema will call list_remove_node on the (dangling) table node; not finding it, list_remove_node will decrement tables->set.cnt and perform no other action.
If as a result of this tables->set.cnt is -1 but h and t are null, the *next* call to list_append will dereference a null pointer.
The attached patch fixes the initial issue in trans_drop_tmp and adds asserts to catch the cases where nelm points outside the list and the count goes below zero. It's not clear to me what list_remove_node should do if passed a pointer which is not a node in the list; it's possible there should be an assert(p != NULL) in list_remove_node somewhere.
What I did with "cs_check" is a little weird too. Ideally we'd avoid the function call and the iteration under -DNDEBUG; perhaps also there should be more cs_foo functions added so that nobody outside sql_changeset.c ever has to modify the set lists?
Reproducible: Always
### Steps to Reproduce:
Start a MonetDB instance, create and release a database called "test". Run the following:
for i in $(seq 1 100); do echo $i; perl -MDBI -e 'my $db = DBI->connect("dbi:monetdb:database=test;host=localhost;port=50000", "monetdb", "monetdb"); $db->do("START TRANSACTION"); $db->do("CREATE LOCAL TEMPORARY TABLE foo ( bar int ) ON COMMIT DROP")'; done
### Actual Results:
mserver5 segfaults once every 8 connections.
### Expected Results:
No segfaults.
The bug was first observed with Jan2014-SP1. It may be older than that.
Not certain about the severity setting - it crashes frequently for us, but most people probably aren't using local temporary tables.
## Comment 20553
Date: 2015-01-12 21:22:43 +0100
From: sorear
Created attachment 314
Patch which appears to fix the bug and adds assertions to avoid a 3rd-level crash
> Attached file: [temp_table_crash.patch](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3648_temp_table_crash.patch_314) (text/plain, 2623 bytes)
> Description: Patch which appears to fix the bug and adds assertions to avoid a 3rd-level crash
## Comment 20577
Date: 2015-01-26 19:15:26 +0100
From: @njnes
added a new cs_remove_node function which includes (re)setting the nelm pointer.
## Comment 20579
Date: 2015-01-26 20:14:48 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [9335d29210f2](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9335d29210f2) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=9335d29210f2](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=9335d29210f2)
Changeset description:
fixed bug #3648, ie properly set nelm after removing a node.
Thanks you sorear@gudtech.com for reporting the bug and supppling the patch.
## Comment 20580
Date: 2015-01-27 02:05:57 +0100
From: sorear
Isn't the use of n->next in cs_remove_node a use after free, since list_remove_node calls node_destroy which (may) call GDKfree?
## Comment 20585
Date: 2015-01-27 09:28:42 +0100
From: @njnes
indeep patch will follow soon..
| memory corruption on unclean connection shutdown with local temporary tables | https://api.github.com/repos/MonetDB/MonetDB/issues/3648/comments | 0 | 2020-11-30T12:58:39Z | 2024-06-27T12:06:03Z | https://github.com/MonetDB/MonetDB/issues/3648 | 753,457,758 | 3,648 |
[
"MonetDB",
"MonetDB"
] | Date: 2015-01-10 00:21:37 +0100
From: sorear
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
CC: @njnes
Last updated: 2015-05-07 12:37:43 +0200
## Comment 20551
Date: 2015-01-10 00:21:37 +0100
From: sorear
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/39.0.2171.95 Safari/537.36
Build Identifier:
One of our MonetDB servers recently entered a state where a particular column existed in sys._columns but was missing from sql_catalog_nme. This lead to several crashes, as follows. create_col (sql/storage/bat/bat_storage.c:965) will return LOG_ERROR and create an invalid (?) sql_delta record with a zero bat it. load_column (sql/storage/store.c:517; line numbers for Oct2014SP1) does not check this error message, and the server starts normally.
However, any attempt to read the batless column results in a server segfault, and (much more problematically) all checkpoints will crash, since gtr_update walks over all columns and gtr_update_delta (./sql/storage/bat/bat_storage.c:1475) does not check whether the inserts-BAT actually exists before checking to see if it has elements.
After crashing, MonetDB reloads the same invalid state, and crashes in the same way on the next attempt to checkpoint (translating into crashes every 30 seconds triggered by the store_update timer).
I was able to get the system back into a stable state by running DROP COLUMN on the offending columns.
Detailed instructions for corrupting a MonetDB database in this way with gdb will be created on request.
I do not currently understand how the corruption was created in the first place; if I can reproduce the corruption I will open a separate ticket for it. If it helps, the column in question had just been dropped and the transaction where the ALTER TABLE DROP COLUMN was run had committed successfully. There may have been an unrelated crash during the gtr_update run that was supposed to flush the DROP COLUMN which set this in motion.
Reproducible: Always
## Comment 20575
Date: 2015-01-26 18:54:30 +0100
From: @njnes
the bat should exist. A possible cause for loosing the bat was fixed.
| missing BAT for a column leads to crash in gtr_update_delta | https://api.github.com/repos/MonetDB/MonetDB/issues/3647/comments | 0 | 2020-11-30T12:58:37Z | 2024-06-27T12:06:02Z | https://github.com/MonetDB/MonetDB/issues/3647 | 753,457,741 | 3,647 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-12-25 00:33:14 +0100
From: K <<mdb>>
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
Last updated: 2015-01-29 14:07:16 +0100
## Comment 20548
Date: 2014-12-25 00:33:14 +0100
From: K <<mdb>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:34.0) Gecko/20100101 Firefox/34.0
Build Identifier:
Given a table with an 'inet' datatype column, displaying the rows using the ORDER BY clause results in incorrect ordering. This appears to be an endian issue.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE "testing" ( "addr" inet );
2. INSERT INTO "testing" VALUES('192.168.0.1');
3. INSERT INTO "testing" VALUES('255.255.255.0');
4. SELECT * FROM "testing" ORDER BY addr;
### Actual Results:
255.255.255.0
192.168.0.1
i.e. the addresses are being sorted on the bases of the LOW byte first, rather than the HIGH byte.
### Expected Results:
192.168.0.1
255.255.255.0
The expected ordering
What the example does not make clear (because it's so short) is that if you have a table with 1000 addresses, the ORDER BY result will appear essentially random because the sorting for an address like A.B.C.D sorts by D first, then C, then B, and then A.
This is what you might expect if the addresses are being stored as simple unsigned integers in host byte order (since x86_64 is little-endian). A quick solution might be to always store network addresses in network byte order and make sure to use the proper comparison.
## Comment 20554
Date: 2015-01-14 14:25:33 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [f81a9bd3789b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=f81a9bd3789b) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=f81a9bd3789b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=f81a9bd3789b)
Changeset description:
Added inet.cmp function. We cannot rely on inherited lngCmp.
Unfortunately, we cannot trust the (rev)sorted flags on inet columns
in older databases, so we need to increment the BBP version number and
squash those flags when the database is upgraded.
This fixes bug #3646.
## Comment 20594
Date: 2015-01-29 14:07:16 +0100
From: @sjoerdmullender
Oct2014-SP2 has been released.
| ORDER BY clause does not produce proper results on 'inet' datatype | https://api.github.com/repos/MonetDB/MonetDB/issues/3646/comments | 0 | 2020-11-30T12:58:34Z | 2024-06-27T12:06:01Z | https://github.com/MonetDB/MonetDB/issues/3646 | 753,457,706 | 3,646 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-12-25 00:21:21 +0100
From: K <<mdb>>
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
CC: @njnes
Last updated: 2015-05-07 12:37:54 +0200
## Comment 20547
Date: 2014-12-25 00:21:21 +0100
From: K <<mdb>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:34.0) Gecko/20100101 Firefox/34.0
Build Identifier:
The 'inet' data type is documented as supporting certain binary operators such as << or <<= (to indicate subnet membership) in WHERE clauses, however these operators are not recognized as such by the command parser. Instead the << is interpreted as a left shift operator, and <<= is flagged as a syntax error.
Reproducible: Always
### Steps to Reproduce:
1. SELECT inet'192.168.0.1' << inet'192.168.0.0/24';
2. SELECT inet'192.168.0.1' <<= inet'192.168.0.0.24';
3.
### Actual Results:
SELECT: no such binary operator 'left_shift(inet,inet)'
syntax error, unexpected '=' in: "select inet'192.168.0.1' <<=
### Expected Results:
true
true
The code for these operators does appear in monetdb5/modules/atoms/inet.c, but somehow it seems that calls to the functions are missing in the parser code. Note however that other operators (such as =, <, >, etc) are handled correctly.
## Comment 20651
Date: 2015-02-11 18:35:38 +0100
From: @njnes
added support for scanning <<= and >>=. Added the create function statements for <<(inet,int), >>(inet,inet) etc.
## Comment 20652
Date: 2015-02-11 18:37:54 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [7bd0698c0cdd](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=7bd0698c0cdd) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=7bd0698c0cdd](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=7bd0698c0cdd)
Changeset description:
added <<=,>>= scanning to the sql lexer
added the create function's for <<(int,int) etc.
(added upgrade code for this)
Added test for bug #3645
## Comment 20667
Date: 2015-02-19 14:28:56 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [e14ed64aa17e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e14ed64aa17e) made by Martin van Dinther <martin.van.dinther@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=e14ed64aa17e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=e14ed64aa17e)
Changeset description:
Approved new output after implementation of <<= and >>= as fix for bug #3645
| Network address operators such as << and <<= do not work | https://api.github.com/repos/MonetDB/MonetDB/issues/3645/comments | 0 | 2020-11-30T12:58:32Z | 2024-06-27T12:06:00Z | https://github.com/MonetDB/MonetDB/issues/3645 | 753,457,685 | 3,645 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-12-25 00:11:46 +0100
From: K <<mdb>>
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
Last updated: 2015-01-29 14:07:42 +0100
## Comment 20546
Date: 2014-12-25 00:11:46 +0100
From: K <<mdb>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:34.0) Gecko/20100101 Firefox/34.0
Build Identifier:
The INSERT INTO command allows networks in CIDR notation (xxx.xxx.xxx.xxx/yy) to be inserted into 'inet' datatype fields, however COPY INTO does not recognize such data values as valid. This has serious implications since it means that large tables cannot be efficiently loaded, and the default output of msqldump (without the -N option) cannot be re-imported back into the database.
Reproducible: Always
### Steps to Reproduce:
1. Create a table with an 'inet' data type column
2. Use INSERT to add a row with the 'inet' field as a network in CIDR notation
3. msqldump the table
4. Try to re-load the table from the msqldump output
### Actual Results:
Fails with what appears to be a syntax/parsing error.
### Expected Results:
Successful loading of the data
Note that if the 'inet' datatype field is a host address (no /nn prefix length), all is well, so the trouble seems to be the /nn specification. I tried many different ways to quote the data field but none seemed to work, so I perhaps the the parsing routine for COPY INTO is different from that used for INSERT?
## Comment 20549
Date: 2014-12-26 12:29:43 +0100
From: @sjoerdmullender
This bug has been fixed with changeset [5b37b295fcac](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=5b37b295fcac)
A test has been added to sql/test/testdb/testdb-dump.
## Comment 20609
Date: 2015-01-29 14:07:42 +0100
From: @sjoerdmullender
Oct2014-SP2 has been released.
| COPY INTO fails to import "inet" data type when value has prefix length in CIDR notation | https://api.github.com/repos/MonetDB/MonetDB/issues/3644/comments | 0 | 2020-11-30T12:58:28Z | 2024-06-27T12:05:59Z | https://github.com/MonetDB/MonetDB/issues/3644 | 753,457,661 | 3,644 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-12-19 18:44:25 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
CC: @njnes
Last updated: 2015-01-29 14:07:37 +0100
## Comment 20543
Date: 2014-12-19 18:44:25 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:34.0) Gecko/20100101 Firefox/34.0
Build Identifier:
missing a few
sql_sub(<timetz>, arg2)
function implementations which do exist for
sql_sub(<time>, arg2).
See also sql/test/pg_regress/timetz.sql
Would be good if they could be added.
Reproducible: Always
### Steps to Reproduce:
CREATE TABLE TIMETZ_TBL (f1 time(2) with time zone);
INSERT INTO TIMETZ_TBL VALUES ('11:59:59.99 PM PDT');
SELECT f1 AS "Time TZ" FROM TIMETZ_TBL;
SELECT f1, sql_sub(f1, 12*60*60.0 + 66) FROM TIMETZ_TBL;
SELECT f1, sql_sub(f1, cast(2*60*60 as interval second)) FROM TIMETZ_TBL;
SELECT f1, sql_sub(f1, cast('23:22:21' as timetz)) FROM TIMETZ_TBL;
SELECT f1, sql_sub(f1, cast('23:22:21' as time)) FROM TIMETZ_TBL;
SELECT f1, sql_sub(cast('23:22:21' as timetz), f1) FROM TIMETZ_TBL;
DROP TABLE TIMETZ_TBL;
### Actual Results:
mclient
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.20.0 (unreleased), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>CREATE TABLE TIMETZ_TBL (f1 time(2) with time zone);
operation successful (2.778ms)
sql>INSERT INTO TIMETZ_TBL VALUES ('11:59:59.99 PM PDT');
1 affected row (1.988ms)
sql>SELECT f1 AS "Time TZ" FROM TIMETZ_TBL;
+-------------------+
| Time TZ |
+===================+
| 12:59:59.99+01:00 |
+-------------------+
1 tuple (3.382ms)
sql>SELECT f1, sql_sub(f1, 12*60*60.0 + 66) FROM TIMETZ_TBL;
SELECT: no such binary operator 'sql_sub(timetz,decimal)'
sql>SELECT f1, sql_sub(f1, cast(2*60*60 as interval second)) FROM TIMETZ_TBL;
SELECT: no such binary operator 'sql_sub(timetz,sec_interval)'
sql>SELECT f1, sql_sub(f1, cast('23:22:21' as timetz)) FROM TIMETZ_TBL;
SELECT: no such binary operator 'sql_sub(timetz,timetz)'
sql>SELECT f1, sql_sub(f1, cast('23:22:21' as time)) FROM TIMETZ_TBL;
SELECT: no such binary operator 'sql_sub(timetz,time)'
sql>SELECT f1, sql_sub(cast('23:22:21' as timetz), f1) FROM TIMETZ_TBL;
SELECT: no such binary operator 'sql_sub(timetz,timetz)'
sql>DROP TABLE TIMETZ_TBL;
operation successful (2.137ms)
### Expected Results:
no errors
## Comment 20544
Date: 2014-12-22 09:17:58 +0100
From: @njnes
function added
## Comment 20604
Date: 2015-01-29 14:07:37 +0100
From: @sjoerdmullender
Oct2014-SP2 has been released.
| Missing implementations of scalar function: sql_sub(<timetz>, arg2) | https://api.github.com/repos/MonetDB/MonetDB/issues/3643/comments | 0 | 2020-11-30T12:58:27Z | 2024-06-27T12:05:58Z | https://github.com/MonetDB/MonetDB/issues/3643 | 753,457,644 | 3,643 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-12-18 12:37:13 +0100
From: @yzchang
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
CC: b.kaptijn, @njnes
Last updated: 2015-01-29 14:07:33 +0100
## Comment 20538
Date: 2014-12-18 12:37:13 +0100
From: @yzchang
Please try the queries below. All three variations of Q1 produce incorrect result, i.e., they return one tuple, while they should return zero tuple. The problem seems to be cause by the combination of '<' and '=' (see Q3), but somehow subquery helps to avoid this problem (see Q2).
Thanks in advance!
Jennie
======================
CREATE TABLE t1 (rec CLOB, datum DATE, ref CLOB, nr CLOB);
CREATE TABLE t2 (rec CLOB, begindatum DATE, einddatum DATE, ref CLOB, nr CLOB);
INSERT INTO t1 VALUES ('1','2012-03-14','8','1234');
INSERT INTO t2 VALUES ('2','2012-03-01','2012-03-31','8','1234');
-- Q1: the following query and its two variations return incorrect result
SELECT DISTINCT t1.rec FROM t1 INNER JOIN t2 ON t1.datum >= t2.begindatum AND t1.datum <= t2.einddatum WHERE t1.ref < t2.ref AND t1.nr = t2.nr;
SELECT DISTINCT t1.rec FROM t1 INNER JOIN t2 ON t1.datum >= t2.begindatum AND t1.datum <= t2.einddatum AND t1.ref < t2.ref WHERE t1.nr = t2.nr;
SELECT DISTINCT t1.rec FROM t1 INNER JOIN t2 ON t1.datum >= t2.begindatum AND t1.datum <= t2.einddatum AND t1.ref < t2.ref AND t1.nr = t2.nr;
-- Q2: alternatives of Q1, but returns correct result, using subqueries to avoid the problem
SELECT DISTINCT rec FROM ( SELECT t1.rec as rec, t1.ref as refA, t2.ref as refB FROM t1 INNER JOIN t2 ON t1.nr = t2.nr WHERE t1.datum >= t2.begindatum AND t1.datum <= t2.einddatum) as tmp WHERE tmp.refA < tmp.refB;
SELECT DISTINCT rec FROM ( SELECT t1.rec as rec, t1.ref as refA, t2.ref as refB, t1.nr AS nrA, t2.nr AS nrB FROM t1 INNER JOIN t2 ON t1.datum >= t2.begindatum AND t1.datum <= t2.einddatum) as tmp WHERE tmp.refA < tmp.refB AND nrA = nrB;
-- Q3: with one condiction less than Q1, but returns correct result
SELECT DISTINCT t1.rec FROM t1 INNER JOIN t2 ON t1.datum >= t2.begindatum AND t1.datum <= t2.einddatum WHERE t1.ref < t2.ref;
DROP TABLE t1;
DROP TABLE t2;
## Comment 20539
Date: 2014-12-18 12:42:42 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [1302c3d9425e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=1302c3d9425e) made by Jennie Zhang <y.zhang@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=1302c3d9425e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=1302c3d9425e)
Changeset description:
Added test for Bug #3642
## Comment 20540
Date: 2014-12-19 14:18:18 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [646c11e1fdac](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=646c11e1fdac) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=646c11e1fdac](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=646c11e1fdac)
Changeset description:
add selections in range - multi column selects, fixes Bug #3642
## Comment 20541
Date: 2014-12-19 14:18:53 +0100
From: @njnes
fixed by adding the selection (when this is not the first) in range (multicolumn)selects
## Comment 20601
Date: 2015-01-29 14:07:33 +0100
From: @sjoerdmullender
Oct2014-SP2 has been released.
| Combined WHERE conditions less-than plus equals-to produce incorrect results | https://api.github.com/repos/MonetDB/MonetDB/issues/3642/comments | 0 | 2020-11-30T12:58:25Z | 2024-06-27T12:05:57Z | https://github.com/MonetDB/MonetDB/issues/3642 | 753,457,616 | 3,642 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-12-17 00:35:11 +0100
From: sorear
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
Last updated: 2015-01-29 14:07:41 +0100
## Comment 20533
Date: 2014-12-17 00:35:11 +0100
From: sorear
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/39.0.2171.95 Safari/537.36
Build Identifier:
If the last character of a string literal is a U+FEFF, then the lexer will spuriously generate a "waiting for more input" response.
Reproducible: Always
### Steps to Reproduce:
1. start mclient
2. enter "select ' ';", run it, see it return a value
3. paste a zero-width no-break space (aka a "byte order mark") immediately before the closing quote, and try to run it again
### Actual Results:
MonetDB considers the version with just spaces to be a complete SQL statement, but responds to the version with the ZWNBS with a secondary prompt (my actual code uses MAPI, and receives a \1\2\n).
### Expected Results:
Both versions are complete SQL statements.
MonetDB 5 server v11.19.7 "Oct2014-SP1" (64-bit, 64-bit oids)
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2014 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 1.8GiB available memory, 2 available cpu cores
Libraries:
libpcre: 7.8 2008-09-05 (compiled with 7.8)
openssl: OpenSSL 1.0.1e 11 Feb 2013 (compiled with OpenSSL 1.0.1e-fips 11 Feb 2013)
libxml2: 2.7.6 (compiled with 2.7.6)
Compiled by: gt@sorear1.dev.gudtech.com (x86_64-unknown-linux-gnu)
Compilation: gcc -O3 -fomit-frame-pointer -pipe -march=native -D_FORTIFY_SOURCE=2
Linking : /usr/bin/ld -m elf_x86_64
## Comment 20545
Date: 2014-12-23 09:51:22 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [72699051b01e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=72699051b01e) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=72699051b01e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=72699051b01e)
Changeset description:
The server should not treat U+FEFF specially, that's an interface issue.
This fixes bug #3641.
Also added test.
## Comment 20607
Date: 2015-01-29 14:07:41 +0100
From: @sjoerdmullender
Oct2014-SP2 has been released.
| SQL lexer fails to detect string end if it the last character is U+FEFF ZERO WIDTH NO-BREAK SPACE | https://api.github.com/repos/MonetDB/MonetDB/issues/3641/comments | 0 | 2020-11-30T12:58:22Z | 2024-06-27T12:05:56Z | https://github.com/MonetDB/MonetDB/issues/3641 | 753,457,589 | 3,641 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-12-12 17:23:51 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
CC: @njnes
Last updated: 2015-01-29 14:07:43 +0100
## Comment 20529
Date: 2014-12-12 17:23:51 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:34.0) Gecko/20100101 Firefox/34.0
Build Identifier:
sql_add(<date>, <month interval>) is implemented. Also
sql_sub(<timestamp>, <month interval>) and
sql_sub(<timestamptz>, <month interval>) are implemented but
sql_sub(<date>, <month interval>) is not!
For orthogonality and consistency
sql_sub(<date>, <month interval>)
should be implemented also.
Reproducible: Always
### Steps to Reproduce:
1. start mserver5
2. start mclient
3. run sql commands:
CREATE TABLE DATE_TBL (f1 date);
INSERT INTO DATE_TBL VALUES ('2014-12-12');
SELECT f1, sql_sub(f1, 365*24*60*60.0) FROM DATE_TBL;
SELECT f1, sql_add(f1, cast(365*24*60*60 as interval second)) FROM DATE_TBL;
SELECT f1, sql_sub(f1, cast(12 as interval month)) FROM DATE_TBL;
SELECT f1, sql_sub(f1, cast(-18 as interval month)) FROM DATE_TBL;
SELECT f1, sql_add(f1, cast(12 as interval month)) FROM DATE_TBL;
SELECT f1, sql_add(f1, -cast(12 as interval month)) FROM DATE_TBL;
DROP TABLE DATE_TBL;
### Actual Results:
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.20.0 (unreleased), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>CREATE TABLE DATE_TBL (f1 date);
operation successful (2.730ms)
sql>INSERT INTO DATE_TBL VALUES ('2014-12-12');
1 affected row (1.058ms)
sql>SELECT f1, sql_sub(f1, 365*24*60*60.0) FROM DATE_TBL;
+------------+------------+
| f1 | sql_sub_f1 |
+============+============+
| 2014-12-12 | 2013-12-12 |
+------------+------------+
1 tuple (1.703ms)
sql>SELECT f1, sql_add(f1, cast(365*24*60*60 as interval second)) FROM DATE_TBL;
+------------+------------+
| f1 | sql_add_f1 |
+============+============+
| 2014-12-12 | 2015-12-12 |
+------------+------------+
1 tuple (1.811ms)
sql>SELECT f1, sql_sub(f1, cast(12 as interval month)) FROM DATE_TBL;
TypeException:user.s6_4[14]:'mtime.date_sub_month_interval' undefined in: _55:any := mtime.date_sub_month_interval(_53:date, _17:int)
program contains errors
sql>SELECT f1, sql_sub(f1, cast(-18 as interval month)) FROM DATE_TBL;
TypeException:user.s7_4[14]:'mtime.date_sub_month_interval' undefined in: _55:any := mtime.date_sub_month_interval(_53:date, _17:int)
program contains errors
sql>SELECT f1, sql_add(f1, cast(12 as interval month)) FROM DATE_TBL;
+------------+------------+
| f1 | sql_add_f1 |
+============+============+
| 2014-12-12 | 2015-12-12 |
+------------+------------+
1 tuple (1.257ms)
sql>SELECT f1, sql_add(f1, -cast(12 as interval month)) FROM DATE_TBL;
+------------+------------+
| f1 | sql_add_f1 |
+============+============+
| 2014-12-12 | 2013-12-12 |
+------------+------------+
1 tuple (1.393ms)
sql>DROP TABLE DATE_TBL;
operation successful (2.162ms)
### Expected Results:
no errors
## Comment 20534
Date: 2014-12-17 12:02:42 +0100
From: @njnes
function added
## Comment 20535
Date: 2014-12-17 12:04:00 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [ae180fad42c8](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ae180fad42c8) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ae180fad42c8](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ae180fad42c8)
Changeset description:
added missing substract month interval function, see Bug #3640.
## Comment 20610
Date: 2015-01-29 14:07:43 +0100
From: @sjoerdmullender
Oct2014-SP2 has been released.
| Missing implementation of scalar function: sql_sub(<date>, <month interval>) | https://api.github.com/repos/MonetDB/MonetDB/issues/3640/comments | 0 | 2020-11-30T12:58:20Z | 2024-06-27T12:05:55Z | https://github.com/MonetDB/MonetDB/issues/3640 | 753,457,570 | 3,640 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-12-12 17:04:56 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
Last updated: 2014-12-19 17:56:56 +0100
## Comment 20528
Date: 2014-12-12 17:04:56 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:34.0) Gecko/20100101 Firefox/34.0
Build Identifier:
System table sys.functions lists scalar functions: second, minute, hour, day, month and year as supported functions for when the first argument is type timestamp or timestamptz.
Using these scalar functions in SQL queries results in an error,
while other scalar timestamp functions such as week(), weekofyear(), dayofmonth(), dayofweek() and dayofyear() run fine.
Reproducible: Always
### Steps to Reproduce:
1. Start mserver5
2. start mclient
3. run sql commands:
CREATE TABLE TIMESTAMP_TBL ( d1 timestamp(2) );
INSERT INTO TIMESTAMP_TBL VALUES ('2014-12-12 23:22:21');
-- the next return a parse error
SELECT d1, hour(d1) FROM TIMESTAMP_TBL;
SELECT d1, minute(d1) FROM TIMESTAMP_TBL;
SELECT d1, second(d1) FROM TIMESTAMP_TBL;
SELECT d1, day(d1) FROM TIMESTAMP_TBL;
SELECT d1, month(d1) FROM TIMESTAMP_TBL;
SELECT d1, year(d1) FROM TIMESTAMP_TBL;
-- the next queries run without error
SELECT d1, week(d1) FROM TIMESTAMP_TBL;
SELECT d1, weekofyear(d1) FROM TIMESTAMP_TBL;
SELECT d1, dayofmonth(d1) FROM TIMESTAMP_TBL;
SELECT d1, dayofweek(d1) FROM TIMESTAMP_TBL;
SELECT d1, dayofyear(d1) FROM TIMESTAMP_TBL;
DROP TABLE TIMESTAMP_TBL;
### Actual Results:
bash-4.2$ mclient
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.20.0 (unreleased), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>CREATE TABLE TIMESTAMP_TBL ( d1 timestamp(2) );
operation successful (2.778ms)
sql>INSERT INTO TIMESTAMP_TBL VALUES ('2014-12-12 23:22:21');
1 affected row (2.035ms)
sql>SELECT d1, hour(d1) FROM TIMESTAMP_TBL;
syntax error, unexpected HOUR in: "select d1, hour"
sql>SELECT d1, minute(d1) FROM TIMESTAMP_TBL;
syntax error, unexpected MINUTE in: "select d1, minute"
sql>SELECT d1, second(d1) FROM TIMESTAMP_TBL;
syntax error, unexpected SECOND in: "select d1, second"
sql>SELECT d1, day(d1) FROM TIMESTAMP_TBL;
syntax error, unexpected DAY in: "select d1, day"
sql>SELECT d1, month(d1) FROM TIMESTAMP_TBL;
syntax error, unexpected MONTH in: "select d1, month"
sql>SELECT d1, year(d1) FROM TIMESTAMP_TBL;
syntax error, unexpected YEAR in: "select d1, year"
sql>SELECT d1, week(d1) FROM TIMESTAMP_TBL;
+------------------------+---------+
| d1 | week_d1 |
+========================+=========+
| 2014-12-12 23:22:21.00 | 50 |
+------------------------+---------+
1 tuple (0.864ms)
sql>SELECT d1, weekofyear(d1) FROM TIMESTAMP_TBL;
+------------------------+---------------+
| d1 | weekofyear_d1 |
+========================+===============+
| 2014-12-12 23:22:21.00 | 50 |
+------------------------+---------------+
1 tuple (1.525ms)
sql>SELECT d1, dayofmonth(d1) FROM TIMESTAMP_TBL;
+------------------------+---------------+
| d1 | dayofmonth_d1 |
+========================+===============+
| 2014-12-12 23:22:21.00 | 12 |
+------------------------+---------------+
1 tuple (1.796ms)
sql>SELECT d1, dayofweek(d1) FROM TIMESTAMP_TBL;
+------------------------+--------------+
| d1 | dayofweek_d1 |
+========================+==============+
| 2014-12-12 23:22:21.00 | 5 |
+------------------------+--------------+
1 tuple (1.978ms)
sql>SELECT d1, dayofyear(d1) FROM TIMESTAMP_TBL;
+------------------------+--------------+
| d1 | dayofyear_d1 |
+========================+==============+
| 2014-12-12 23:22:21.00 | 346 |
+------------------------+--------------+
1 tuple (1.064ms)
sql>DROP TABLE TIMESTAMP_TBL;
operation successful (2.217ms)
### Expected Results:
No error when second(), minute(), hour(), day(), month() or year() are used in SQL on a timestamp or timestaptz column.
When implementing you may consider to add more scalar timestamp and timestamptz functions such as:
quarter(<timestamp>), halfyear(<timestamp>), century(<timestamp>), millenium(<timestamp>), epoch(<timestamp>)
## Comment 20531
Date: 2014-12-12 17:46:29 +0100
From: @sjoerdmullender
Same as bug #3638: you need to quote the function names:
SELECT d1, "hour"(d1) FROM TIMESTAMP_TBL;
SELECT d1, "minute"(d1) FROM TIMESTAMP_TBL;
SELECT d1, "second"(d1) FROM TIMESTAMP_TBL;
SELECT d1, "day"(d1) FROM TIMESTAMP_TBL;
SELECT d1, "month"(d1) FROM TIMESTAMP_TBL;
SELECT d1, "year"(d1) FROM TIMESTAMP_TBL;
## Comment 20542
Date: 2014-12-19 17:56:56 +0100
From: Martin van Dinther <<martin.van.dinther>>
If function names as stored in sys.functions can not be used as functions in SQL, we may need to change the names in sys.functions to the supported names, so with the double quotes. E.g.:
UPDATE sys.functions SET name = '"'||name||'"'
WHERE name in ('second', 'minute', 'hour', 'day', 'month', 'year')
AND mod = 'mtime'
-- 24 rows updated.
This way the function names are in-sync with the SQL parser behaviour.
| SQL parser does not accept scalar functions second(), minute(), hour(), day(), month(), year() for timestamp or timestamptz column | https://api.github.com/repos/MonetDB/MonetDB/issues/3639/comments | 0 | 2020-11-30T12:58:17Z | 2024-06-28T13:15:33Z | https://github.com/MonetDB/MonetDB/issues/3639 | 753,457,544 | 3,639 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-12-12 16:51:43 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
Last updated: 2014-12-12 17:43:49 +0100
## Comment 20527
Date: 2014-12-12 16:51:43 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:34.0) Gecko/20100101 Firefox/34.0
Build Identifier:
System table sys.functions lists scalar functions: day, month and year as supported functions for when the first argument is type date.
Using these scalar functions in SQL queries results in an error,
while other scalar date functions such as week(), weekofyear(), dayofmonth(), dayofweek() and dayofyear() run fine.
Reproducible: Always
### Steps to Reproduce:
1. Start mserver5
2. start mclient
3. run sql commands:
CREATE TABLE DATE_TBL (f1 date);
INSERT INTO DATE_TBL VALUES ('2014-12-12');
-- the next return a parse error
SELECT f1, day(f1) FROM DATE_TBL;
SELECT f1, month(f1) FROM DATE_TBL;
SELECT f1, year(f1) FROM DATE_TBL;
-- the next queries run without error
SELECT f1, week(f1) FROM DATE_TBL;
SELECT f1, weekofyear(f1) FROM DATE_TBL;
SELECT f1, dayofmonth(f1) FROM DATE_TBL;
SELECT f1, dayofweek(f1) FROM DATE_TBL;
SELECT f1, dayofyear(f1) FROM DATE_TBL;
DROP TABLE DATE_TBL;
### Actual Results:
bash-4.2$ mclient
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.20.0 (unreleased), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>CREATE TABLE DATE_TBL (f1 date);
operation successful (2.421ms)
sql>INSERT INTO DATE_TBL VALUES ('2014-12-12');
1 affected row (1.940ms)
sql>SELECT f1, day(f1) FROM DATE_TBL;
syntax error, unexpected DAY in: "select f1, day"
sql>SELECT f1, month(f1) FROM DATE_TBL;
syntax error, unexpected MONTH in: "select f1, month"
sql>SELECT f1, year(f1) FROM DATE_TBL;
syntax error, unexpected YEAR in: "select f1, year"
sql>SELECT f1, week(f1) FROM DATE_TBL;
+------------+---------+
| f1 | week_f1 |
+============+=========+
| 2014-12-12 | 50 |
+------------+---------+
1 tuple (1.269ms)
sql>SELECT f1, weekofyear(f1) FROM DATE_TBL;
+------------+---------------+
| f1 | weekofyear_f1 |
+============+===============+
| 2014-12-12 | 50 |
+------------+---------------+
1 tuple (1.186ms)
sql>SELECT f1, dayofmonth(f1) FROM DATE_TBL;
+------------+---------------+
| f1 | dayofmonth_f1 |
+============+===============+
| 2014-12-12 | 12 |
+------------+---------------+
1 tuple (1.882ms)
sql>SELECT f1, dayofweek(f1) FROM DATE_TBL;
+------------+--------------+
| f1 | dayofweek_f1 |
+============+==============+
| 2014-12-12 | 5 |
+------------+--------------+
1 tuple (1.369ms)
sql>SELECT f1, dayofyear(f1) FROM DATE_TBL;
+------------+--------------+
| f1 | dayofyear_f1 |
+============+==============+
| 2014-12-12 | 346 |
+------------+--------------+
1 tuple (0.838ms)
sql>DROP TABLE DATE_TBL;
operation successful (2.241ms)
### Expected Results:
No error when day(), month() or year() are used in SQL.
When implementing you may consider to add more scalar date functions such as:
quarter(<date>), halfyear(<date>), century(<date>), millenium(<date>), epoch(<date>)
## Comment 20530
Date: 2014-12-12 17:43:49 +0100
From: @sjoerdmullender
You need to quote the names of these functions:
SELECT f1, "day"(f1) FROM DATE_TBL;
SELECT f1, "month"(f1) FROM DATE_TBL;
SELECT f1, "year"(f1) FROM DATE_TBL;
This is because the tokens day, month and year are SQL keywords (used in e.g. interval types).
| SQL parser does not accept scalar functions day(), month(), year() for date column | https://api.github.com/repos/MonetDB/MonetDB/issues/3638/comments | 0 | 2020-11-30T12:58:15Z | 2024-06-28T13:15:33Z | https://github.com/MonetDB/MonetDB/issues/3638 | 753,457,518 | 3,638 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-12-05 16:51:47 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
CC: @njnes
Last updated: 2014-12-10 13:34:33 +0100
## Comment 20519
Date: 2014-12-05 16:51:47 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:33.0) Gecko/20100101 Firefox/33.0
Build Identifier:
It appears any string is accepted.
When the string value starts with an integer number (such a 1 or +2 or -3) that number will be stored
else 0 will be stored, even if the string represent non-sense.
For intervals there are special string interpretation semantics such as:
3 4:05:06 SQL standard format: 3 days 4 hours 5 minutes 6 seconds.
1-2 SQL standard format: 1 year 2 months
Postgresql has additional string interpretation semantics, see http://www.postgresql.org/docs/9.3/interactive/datatype-datetime.htmlDATATYPE-INTERVAL-INPUT
Reproducible: Always
### Steps to Reproduce:
1. start mserver5
2. start mclient
3. run SQL:
CREATE TABLE INTERVAL_TBL (f1 interval second);
INSERT INTO INTERVAL_TBL (f1) VALUES ('1.2345');
INSERT INTO INTERVAL_TBL (f1) VALUES (60 * 60 * 24 * 365 * 2014);
INSERT INTO INTERVAL_TBL (f1) VALUES ('123456789012345678');
INSERT INTO INTERVAL_TBL (f1) VALUES ('1234567890123456789');
INSERT INTO INTERVAL_TBL (f1) VALUES ('12345678901234567890');
SELECT * FROM INTERVAL_TBL;
INSERT INTO INTERVAL_TBL (f1) VALUES ('1 day 2 hours 3 minutes 4 seconds');
INSERT INTO INTERVAL_TBL (f1) VALUES ('6 years');
INSERT INTO INTERVAL_TBL (f1) VALUES ('5 months');
INSERT INTO INTERVAL_TBL (f1) VALUES ('5 months 12 hours');
SELECT * FROM INTERVAL_TBL;
INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 1 minute');
INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 5 hour');
INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 10 day');
INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 34 year');
INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 3 months');
INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 14 seconds ago');
SELECT * FROM INTERVAL_TBL;
INSERT INTO INTERVAL_TBL (f1) VALUES ('badly formatted interval');
INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 30 eons ago');
SELECT * FROM INTERVAL_TBL;
SELECT cast('01:00' as interval second) AS "One hour";
SELECT cast('+02:00' as interval second) AS "Two hours";
SELECT cast('-08:00' as interval second) AS "Eight hours";
SELECT cast('-05' as interval second) AS "Five hours";
SELECT cast('-1 +02:03' as interval second) AS "22 hours ago...";
SELECT cast('-1 days +02:03' as interval second) AS "22 hours ago...";
SELECT cast('10 years -11 month -12 days +13:14' as interval second) AS "9 years...";
### Actual Results:
bash-4.2$ mclient
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.20.0 (unreleased), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>select * from sys.types where sqlname like '%interval%';
+------+------------+----------------+--------+-------+-------+--------+-----------+
| id | systemname | sqlname | digits | scale | radix | eclass | schema_id |
+======+============+================+========+=======+=======+========+===========+
| 21 | int | month_interval | 32 | 0 | 2 | 7 | 0 |
| 22 | lng | sec_interval | 19 | 1 | 10 | 7 | 0 |
+------+------------+----------------+--------+-------+-------+--------+-----------+
2 tuples (5.555ms)
sql>\q;
bash-4.2$
bash-4.2$
bash-4.2$
bash-4.2$
bash-4.2$ mclient
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.20.0 (unreleased), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>CREATE TABLE INTERVAL_TBL (f1 interval second);
operation successful (35.846ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 1 minute');
1 affected row (2.026ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 5 hour');
1 affected row (1.933ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 10 day');
1 affected row (1.587ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 34 year');
1 affected row (1.698ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 3 months');
1 affected row (1.694ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 14 seconds ago');
1 affected row (2.609ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('1 day 2 hours 3 minutes 4 seconds');
1 affected row (2.445ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('6 years');
1 affected row (2.325ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('5 months 12 hours');
1 affected row (3.322ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 30 eons ago');
1 affected row (21.895ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('badly formatted interval');
1 affected row (2.169ms)
sql>SELECT * FROM INTERVAL_TBL;
+----------------------------------------------------------------------------------------------------------------+
| f1 |
+================================================================================================================+
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 1.000 |
| 6.000 |
| 5.000 |
| 0.000 |
| 0.000 |
+----------------------------------------------------------------------------------------------------------------+
11 tuples (1.145ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('1.2345');
1 affected row (2.327ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('123456789012345');
1 affected row (1.943ms)
sql>SELECT * FROM INTERVAL_TBL;
+----------------------------------------------------------------------------------------------------------------+
| f1 |
+================================================================================================================+
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 1.000 |
| 6.000 |
| 5.000 |
| 0.000 |
| 0.000 |
| 1.234 |
| 1.234 |
| 123456789012345.000 |
| 123456789012345.000 |
| 123456789012345.000 |
+----------------------------------------------------------------------------------------------------------------+
16 tuples (0.817ms)
sql>SELECT * FROM INTERVAL_TBL;
+----------------------------------------------------------------------------------------------------------------+
| f1 |
+================================================================================================================+
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 1.000 |
| 6.000 |
| 5.000 |
| 0.000 |
| 0.000 |
| 1.234 |
| 123456789012345.000 |
+----------------------------------------------------------------------------------------------------------------+
13 tuples (0.708ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES (60 * 60 * 24 * 365 * 2014);
1 affected row (3.274ms)
sql>SELECT * FROM INTERVAL_TBL;
+----------------------------------------------------------------------------------------------------------------+
| f1 |
+================================================================================================================+
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 1.000 |
| 6.000 |
| 5.000 |
| 0.000 |
| 0.000 |
| 1.234 |
| 123456789012345.000 |
| 63513504000.000 |
| 63513504000.000 |
+----------------------------------------------------------------------------------------------------------------+
15 tuples (0.757ms)
sql>\q;
bash-4.2$ mclient
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.20.0 (unreleased), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>CREATE TABLE INTERVAL_TBL (f1 interval second);
operation successful (39.253ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('1.2345');
1 affected row (2.976ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES (60 * 60 * 24 * 365 * 2014);
1 affected row (19.355ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('123456789012345678');
1 affected row (3.279ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('1234567890123456789');
1 affected row (2.442ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('12345678901234567890');
1 affected row (4.519ms)
sql>SELECT * FROM INTERVAL_TBL;
+----------------------------------------------------------------------------------------------------------------+
| f1 |
+================================================================================================================+
| 1.234 |
| 63513504000.000 |
| -5670419503621183.312 |
| -1363962815083169.272 |
| -1.000 |
+----------------------------------------------------------------------------------------------------------------+
5 tuples (1.075ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('1 day 2 hours 3 minutes 4 seconds');
1 affected row (1.904ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('6 years');
1 affected row (2.571ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('5 months');
1 affected row (3.148ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('5 months 12 hours');
1 affected row (2.330ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 1 minute');
1 affected row (3.047ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 5 hour');
1 affected row (2.088ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 10 day');
1 affected row (2.369ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 34 year');
1 affected row (5.788ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 3 months');
1 affected row (2.013ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 14 seconds ago');
1 affected row (5.293ms)
sql>SELECT * FROM INTERVAL_TBL;
+----------------------------------------------------------------------------------------------------------------+
| f1 |
+================================================================================================================+
| 1.234 |
| 63513504000.000 |
| -5670419503621183.312 |
| -1363962815083169.272 |
| -1.000 |
| 1.000 |
| 6.000 |
| 5.000 |
| 5.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
+----------------------------------------------------------------------------------------------------------------+
15 tuples (0.710ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('badly formatted interval');
1 affected row (2.276ms)
sql>INSERT INTO INTERVAL_TBL (f1) VALUES ('@ 30 eons ago');
1 affected row (10.857ms)
sql>SELECT * FROM INTERVAL_TBL;
+----------------------------------------------------------------------------------------------------------------+
| f1 |
+================================================================================================================+
| 1.234 |
| 63513504000.000 |
| -5670419503621183.312 |
| -1363962815083169.272 |
| -1.000 |
| 1.000 |
| 6.000 |
| 5.000 |
| 5.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
| 0.000 |
+----------------------------------------------------------------------------------------------------------------+
17 tuples (0.624ms)
sql>SELECT cast('01:00' as interval second) AS "One hour";
+----------------------------------------------------------------------------------------------------------------+
| One hour |
+================================================================================================================+
| 1.000 |
+----------------------------------------------------------------------------------------------------------------+
1 tuple (0.588ms)
sql>SELECT cast('+02:00' as interval second) AS "Two hours";
+----------------------------------------------------------------------------------------------------------------+
| Two hours |
+================================================================================================================+
| 2.000 |
+----------------------------------------------------------------------------------------------------------------+
1 tuple (0.664ms)
sql>SELECT cast('-08:00' as interval second) AS "Eight hours";
+----------------------------------------------------------------------------------------------------------------+
| Eight hours |
+================================================================================================================+
| -8.000 |
+----------------------------------------------------------------------------------------------------------------+
1 tuple (0.412ms)
sql>SELECT cast('-05' as interval second) AS "Five hours";
+----------------------------------------------------------------------------------------------------------------+
| Five hours |
+================================================================================================================+
| -5.000 |
+----------------------------------------------------------------------------------------------------------------+
1 tuple (0.430ms)
sql>SELECT cast('-1 +02:03' as interval second) AS "22 hours ago...";
+----------------------------------------------------------------------------------------------------------------+
| 22 hours ago... |
+================================================================================================================+
| -1.000 |
+----------------------------------------------------------------------------------------------------------------+
1 tuple (0.638ms)
sql>SELECT cast('-1 days +02:03' as interval second) AS "22 hours ago...";
+----------------------------------------------------------------------------------------------------------------+
| 22 hours ago... |
+================================================================================================================+
| -1.000 |
+----------------------------------------------------------------------------------------------------------------+
1 tuple (0.629ms)
sql>SELECT cast('10 years -11 month -12 days +13:14' as interval second) AS "9 years...";
+----------------------------------------------------------------------------------------------------------------+
| 9 years... |
+================================================================================================================+
| 10.000 |
+----------------------------------------------------------------------------------------------------------------+
1 tuple (0.651ms)
sql>
### Expected Results:
do proper validation of the string value and return an error when it does not comply to an "interval of seconds" value or get outside the range of allowed values.
See also: sql/test/pg_regress/Tests/interval.sql
## Comment 20524
Date: 2014-12-10 13:34:33 +0100
From: @njnes
we do not support the postgres extensions. Sofar we require proper bounds (ie hour to minute etc) to be given within the query.
Added (on default only) a error when the string is not fully parsed.
| conversion of string values to an interval second column is implemented incorrect and incomplete | https://api.github.com/repos/MonetDB/MonetDB/issues/3637/comments | 0 | 2020-11-30T12:58:12Z | 2024-06-27T12:05:51Z | https://github.com/MonetDB/MonetDB/issues/3637 | 753,457,474 | 3,637 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-12-05 15:52:59 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
CC: @njnes
Last updated: 2014-12-10 10:52:06 +0100
## Comment 20518
Date: 2014-12-05 15:52:59 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:33.0) Gecko/20100101 Firefox/33.0
Build Identifier:
In system tables sys.types all supported data types are listed.
Use query: select * from sys.types;
The returned sqlname of each type returns the name to be used in SQL (such as CREATE TABLE or cast(a as sqltypename))
For types with id 21 and 22 the sqlname contains "month_interval" and "sec_interval". Those names are not accepted in SQL CREATE TABLE or cast() functions. They should be changed in to "interval month" and "interval second"
Reproducible: Always
### Steps to Reproduce:
1. start mserver5
2. start mclient
3. execute sql queries:
select * from sys.types where sqlname like '%interval%';
-- shows 2 rows with sqlnames: "month_interval" and "sec_interval"
CREATE TABLE sec_interval (col1 sec_interval);
-- this fails
CREATE TABLE interval_second (col1 interval second);
-- this succeeds
CREATE TABLE month_interval (col1 month_interval);
-- this fails: type (month_interval) unknown in: "create table month_interval (col1 month_interval)"
CREATE TABLE interval_month (col1 interval month);
-- this succeeds
SELECT cast(123 as sec_interval) as col1;
-- this fails: type (sec_interval) unknown in: "select cast(123 as sec_interval)"
SELECT cast(5 as month_interval) as col1;
-- this fails: type (month_interval) unknown in: "select cast(5 as month_interval)"
SELECT cast(123 as interval second) as col1;
SELECT cast('123' as interval second) as col1;
SELECT cast(5 as interval month) as col1;
SELECT cast('5' as interval month) as col1;
DROP TABLE interval_second;
DROP TABLE interval_month;
### Actual Results:
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.20.0 (unreleased), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>select * from sys.types where sqlname like '%interval%';
+------+------------+----------------+--------+-------+-------+--------+-----------+
| id | systemname | sqlname | digits | scale | radix | eclass | schema_id |
+======+============+================+========+=======+=======+========+===========+
| 21 | int | month_interval | 32 | 0 | 2 | 7 | 0 |
| 22 | lng | sec_interval | 19 | 1 | 10 | 7 | 0 |
+------+------------+----------------+--------+-------+-------+--------+-----------+
2 tuples (5.555ms)
sql>CREATE TABLE sec_interval (col1 sec_interval);
type (sec_interval) unknown in: "create table sec_interval (col1 sec_interval)"
sql>CREATE TABLE interval_second (col1 interval second);
operation successful (41.799ms)
sql>CREATE TABLE month_interval (col1 month_interval);
type (month_interval) unknown in: "create table month_interval (col1 month_interval)"
sql>CREATE TABLE interval_month (col1 interval month);
operation successful (4.424ms)
sql>SELECT cast(123 as sec_interval) as col1;
type (sec_interval) unknown in: "select cast(123 as sec_interval)"
syntax error, unexpected AS in: "as"
sql>SELECT cast(5 as month_interval) as col1;
type (month_interval) unknown in: "select cast(5 as month_interval)"
syntax error, unexpected AS in: "as"
sql>SELECT cast(123 as interval second) as col1;
+----------------------------------------------------------------------------------------------------------------+
| col1 |
+================================================================================================================+
| 123.000 |
+----------------------------------------------------------------------------------------------------------------+
1 tuple (0.378ms)
sql>SELECT cast('123' as interval second) as col1;
+----------------------------------------------------------------------------------------------------------------+
| col1 |
+================================================================================================================+
| 123.000 |
+----------------------------------------------------------------------------------------------------------------+
1 tuple (0.627ms)
sql>SELECT cast(5 as interval month) as col1;
+----------------------------------------------------------------------------------------------------------------+
| col1 |
+================================================================================================================+
| 5 |
+----------------------------------------------------------------------------------------------------------------+
1 tuple (0.646ms)
sql>SELECT cast('5' as interval month) as col1;
+----------------------------------------------------------------------------------------------------------------+
| col1 |
+================================================================================================================+
| 5 |
+----------------------------------------------------------------------------------------------------------------+
1 tuple (0.608ms)
sql>DROP TABLE interval_second;
operation successful (2.119ms)
sql>DROP TABLE interval_month;
operation successful (3.232ms)
sql>\q;
### Expected Results:
please correct the sqlname values for these 2 data types to "interval month" and "interval second" to bring it in line with the sql parser.
## Comment 20523
Date: 2014-12-10 10:52:06 +0100
From: @njnes
the internal names of the sql types are month_/sec_interval. The combination of sqlname and digits/scale columns gives a unique description of the types. There is indeed (why would there be) link to the sql keyword(s)
| wrong sqlname for data types: "month_interval" and "sec_interval" in sys.types | https://api.github.com/repos/MonetDB/MonetDB/issues/3636/comments | 0 | 2020-11-30T12:58:09Z | 2024-06-28T13:15:32Z | https://github.com/MonetDB/MonetDB/issues/3636 | 753,457,443 | 3,636 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-12-04 18:35:46 +0100
From: Richard Hughes <<richard.monetdb>>
To: SQL devs <<bugs-sql>>
Version: 11.19.3 (Oct2014)
CC: @njnes
Last updated: 2015-01-28 18:32:54 +0100
## Comment 20511
Date: 2014-12-04 18:35:46 +0100
From: Richard Hughes <<richard.monetdb>>
Created attachment 313
Reproduction script
To reproduce:
1) monetdb create acidityfail
2) monetdb start acidityfail
3) ./acidityfail.py `pidof mserver5`
Important: The attached acidityfail.py will kill -9 the PID it is given. If you have more than one mserver5 instance on the machine you're running it on, then the last line given above may kill the wrong one. Be careful.
The script has some comments in it explaining what's going on. The expected behaviour is that the contents of the table 'foo' before and after the kill+restart should be the same. The actual behaviour is that the data has been rolled back to pre-'delete from foo', despite all statements running in autocommit mode.
To run the script a second time you'll need to "drop table foo;drop table bar;" manually (or recreate the entire database).
I believe this bug has existed for as long as I've been using MonetDB.
> Attached file: [acidityfail.py](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3635_acidityfail.py_313) (application/octet-stream, 2316 bytes)
> Description: Reproduction script
## Comment 20514
Date: 2014-12-04 22:25:21 +0100
From: @sjoerdmullender
From my initial very quick investigation, I think it may be going wrong already before the server get killed. The WAL is empty when the server gets killed and the new BAT is not mentioned in the BBP. The old one is, though.
## Comment 20526
Date: 2014-12-12 15:52:28 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [6be6b54a52d9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=6be6b54a52d9) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=6be6b54a52d9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=6be6b54a52d9)
Changeset description:
Added test for bug #3635.
Richard Hughes' test, but modified to fit into the testing infrastructure.
## Comment 20532
Date: 2014-12-15 14:17:51 +0100
From: Richard Hughes <<richard.monetdb>>
I made my test overcomplicated: the shutdown doesn't need to be unclean, a normal stop/start cycle will also reproduce this issue.
Updated summary line to reflect this.
## Comment 20550
Date: 2015-01-08 09:33:11 +0100
From: @njnes
working on a solution to re-register the bat after a 'clear' (delete all).
My current patch seems to work, but sure needs more cleaning and thought.
## Comment 20569
Date: 2015-01-25 19:01:46 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [324d579fa720](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=324d579fa720) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=324d579fa720](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=324d579fa720)
Changeset description:
work-around for a bug #3635
(temporary) fix / improved check for bug #3650
## Comment 20570
Date: 2015-01-26 12:51:31 +0100
From: Richard Hughes <<richard.monetdb>>
Excellent. Thanks Niels.
I haven't spotted any regressions with this patch yet, and I've given it a fairly good work-out. I'll let you know if anything does come up.
## Comment 20586
Date: 2015-01-27 14:05:28 +0100
From: Richard Hughes <<richard.monetdb>>
Bad news - I broke it:
mserver5: bat_storage.c:1407: empty_col: Assertion `c->data && c->base.allocated
&& bat->bid == 0' failed.
Program received signal SIGABRT, Aborted.
[Switching to Thread 0x7fffebcd3700 (LWP 6378)]
0x00007ffff5d31107 in __GI_raise (sig=sig@entry=6)
at ../nptl/sysdeps/unix/sysv/linux/raise.c:56
56 ../nptl/sysdeps/unix/sysv/linux/raise.c: No such file or directory.
(gdb) bt
0 0x00007ffff5d31107 in __GI_raise (sig=sig@entry=6)
at ../nptl/sysdeps/unix/sysv/linux/raise.c:56
1 0x00007ffff5d324e8 in __GI_abort () at abort.c:89
2 0x00007ffff5d2a226 in __assert_fail_base (
fmt=0x7ffff5e60948 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n",
assertion=assertion@entry=0x7fffeff80028 "c->data && c->base.allocated && ba
t->bid == 0", file=file@entry=0x7fffeff7f740 "bat_storage.c",
line=line@entry=1407,
function=function@entry=0x7fffeff80683 <__PRETTY_FUNCTION__.12010> "empty_co
l") at assert.c:92
3 0x00007ffff5d2a2d2 in __GI___assert_fail (
assertion=0x7fffeff80028 "c->data && c->base.allocated && bat->bid == 0",
file=0x7fffeff7f740 "bat_storage.c", line=1407,
function=0x7fffeff80683 <__PRETTY_FUNCTION__.12010> "empty_col")
at assert.c:101
4 0x00007fffeff314db in empty_col (c=0x7ffe180203d0) at bat_storage.c:1407
5 0x00007fffeff33014 in update_table (tr=0x7ffe18007f80, ft=0x7ffe18020280,
tt=0x1ffc560) at bat_storage.c:1823
6 0x00007fffeff1ea62 in rollforward_update_table (tr=0x7ffe18007f80,
ft=0x7ffe18020280, tt=0x1ffc560, mode=3) at store.c:2761
7 0x00007fffeff1d679 in rollforward_changeset_updates (tr=0x7ffe18007f80,
fs=0x7ffe180082b0, ts=0x20fdfb0, b=0x20fdf80,
rollforward_updates=0x7fffeff1e6f9 <rollforward_update_table>,
rollforward_creates=0x7fffeff1de2e <rollforward_create_table>,
rollforward_deletes=0x7fffeff1e469 <rollforward_drop_table>,
fd=0x7fffeff1eb78 <conditional_table_dup>, mode=3) at store.c:2374
8 0x00007fffeff1ee95 in rollforward_update_schema (tr=0x7ffe18007f80,
fs=0x7ffe18008280, ts=0x20fdf80, mode=3) at store.c:2837
9 0x00007fffeff1d679 in rollforward_changeset_updates (tr=0x7ffe18007f80,
fs=0x7ffe18007fb0, ts=0x1cc8100, b=0x1cc80d0,
rollforward_updates=0x7fffeff1ec8d <rollforward_update_schema>,
rollforward_creates=0x7fffeff1e692 <rollforward_create_schema>,
rollforward_deletes=0x7fffeff1e602 <rollforward_drop_schema>,
fd=0x7fffeff1cbb5 <schema_dup>, mode=3) at store.c:2374
10 0x00007fffeff1f038 in rollforward_trans (tr=0x7ffe18007f80, mode=3)
at store.c:2859
11 0x00007fffeff20552 in sql_trans_commit (tr=0x7ffe18007f80) at store.c:3275
12 0x00007fffefe9330a in mvc_commit (m=0x7ffdc8004a90, chain=0, name=0x0)
at sql_mvc.c:269
13 0x00007fffefde0151 in SQLtransaction (cntxt=0x680750, mb=0x7ffe200077c0,
stk=0x7ffdc812ea40, pci=0x7ffdc80391f0) at sql.c:282
14 0x00007ffff78c7c38 in runMALsequence (cntxt=0x680750, mb=0x7ffe200077c0,
startpc=1, stoppc=0, stk=0x7ffdc812ea40, env=0x0, pcicaller=0x0)
at mal_interpreter.c:651
15 0x00007ffff78c686b in runMAL (cntxt=0x680750, mb=0x7ffe200077c0,
mbcaller=0x0, env=0x0) at mal_interpreter.c:374
16 0x00007fffefe0b0ad in SQLengineIntern (c=0x680750, be=0x7ffdc8007ca0)
at sql_scenario.c:2704
17 0x00007fffefe0b74f in SQLengine (c=0x680750) at sql_scenario.c:2797
18 0x00007ffff78f7d85 in runPhase (c=0x680750, phase=4) at mal_scenario.c:526
19 0x00007ffff78f7fb3 in runScenarioBody (c=0x680750) at mal_scenario.c:570
20 0x00007ffff78f80c3 in runScenario (c=0x680750) at mal_scenario.c:590
21 0x00007ffff78f9a80 in MSserveClient (dummy=0x680750) at mal_session.c:459
22 0x00007ffff7395964 in thread_starter (arg=0x7ffe200079a0)
at gdk_system.c:535
23 0x00007ffff60ad0a4 in start_thread (arg=0x7fffebcd3700)
at pthread_create.c:309
24 0x00007ffff5de1cbd in clone ()
at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111
(gdb) f 4
4 0x00007fffeff314db in empty_col (c=0x7ffe180203d0) at bat_storage.c:1407
1407 assert(c->data && c->base.allocated && bat->bid == 0);
(gdb) p *c
$1 = {base = {wtime = 13876, rtime = 0, allocated = 1, flag = 0, id = 6872,
name = 0x7ffe18020440 "id"}, type = {type = 0x1c55670, digits = 64,
scale = 0}, colnr = 0, null = 0 '\000', def = 0x0, unique = 0 '\000',
drop_action = 0, storage_type = 0x0, sorted = 0, t = 0x7ffe18020280,
data = 0x7ffdc80b2510}
(gdb) p *bat
$2 = {name = 0x7ffdc8048be0 "sys_data_staging_id", bid = 2718,
ibase = 3948, ibid = 1968, ubid = 1908, cnt = 12825, ucnt = 0,
cached = 0x0, wtime = 13876, next = 0x0}
I'm still try to get you a proper repro, but I've managed the same crash twice so far and they both followed a rolled-back transaction, one step of which was the "delete from data_staging;". In both cases the transaction rolled back because of an error in a later step rather than an explicit rollback statement (once was a failure to create a new table, the other was a concurrency conflict in the catalog).
Yesterday's test was with your patch grafted on to my old working copy from 2014-11-10. Today's testing has been on Oct2014 head, however yesterday's test didn't happen to trip over any rollbacks so I can't say for sure whether it's this patch or anything else in the past 2.5 months that has caused the problem.
## Comment 20587
Date: 2015-01-27 15:01:34 +0100
From: Richard Hughes <<richard.monetdb>>
Created attachment 315
Reproduction script, causes SIGABRT
As promised, here's a repro for you. I haven't tried it on a release build, but on a debug build it asserts in tr_update_delta(). This is different to the stack trace in my previous comment, but I hope they have the same root cause.
mserver5: bat_storage.c:1722: tr_update_delta: Assertion `(((cur)->S->count) + (
(ins)->S->count)) == cbat->cnt' failed.
[Thread 0x7fffee372700 (LWP 15886) exited]
Program received signal SIGABRT, Aborted.
[Switching to Thread 0x7fffee774700 (LWP 15861)]
0x00007ffff5d31107 in __GI_raise (sig=sig@entry=6)
at ../nptl/sysdeps/unix/sysv/linux/raise.c:56
56 ../nptl/sysdeps/unix/sysv/linux/raise.c: No such file or directory.
(gdb) bt
0 0x00007ffff5d31107 in __GI_raise (sig=sig@entry=6)
at ../nptl/sysdeps/unix/sysv/linux/raise.c:56
1 0x00007ffff5d324e8 in __GI_abort () at abort.c:89
2 0x00007ffff5d2a226 in __assert_fail_base (
fmt=0x7ffff5e60948 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n",
assertion=assertion@entry=0x7ffff0101280 "(((cur)->S->count) + ((ins)->S->co
unt)) == cbat->cnt", file=file@entry=0x7ffff0100740 "bat_storage.c",
line=line@entry=1722,
function=function@entry=0x7ffff01016d0 <__PRETTY_FUNCTION__.12126> "tr_updat
e_delta") at assert.c:92
3 0x00007ffff5d2a2d2 in __GI___assert_fail (
assertion=0x7ffff0101280 "(((cur)->S->count) + ((ins)->S->count)) == cbat->c
nt", file=0x7ffff0100740 "bat_storage.c", line=1722,
function=0x7ffff01016d0 <__PRETTY_FUNCTION__.12126> "tr_update_delta")
at assert.c:101
4 0x00007ffff00b3865 in tr_update_delta (tr=0x1ab7010, obat=0x7fffe013e500,
cbat=0x7fffe022e9b0, unique=0) at bat_storage.c:1722
5 0x00007ffff00b4319 in update_table (tr=0x1ab7010, ft=0x7fffe0135ef0,
tt=0x7fffe02d37b0) at bat_storage.c:1871
6 0x00007ffff009fa62 in rollforward_update_table (tr=0x1ab7010,
ft=0x7fffe0135ef0, tt=0x7fffe02d37b0, mode=3) at store.c:2761
7 0x00007ffff009e679 in rollforward_changeset_updates (tr=0x1ab7010,
fs=0x1ab7340, ts=0x1aaeaa0, b=0x1aaea70,
rollforward_updates=0x7ffff009f6f9 <rollforward_update_table>,
rollforward_creates=0x7ffff009ee2e <rollforward_create_table>,
rollforward_deletes=0x7ffff009f469 <rollforward_drop_table>,
fd=0x7ffff009fb78 <conditional_table_dup>, mode=3) at store.c:2374
8 0x00007ffff009fe95 in rollforward_update_schema (tr=0x1ab7010,
fs=0x1ab7310, ts=0x1aaea70, mode=3) at store.c:2837
9 0x00007ffff009e679 in rollforward_changeset_updates (tr=0x1ab7010,
fs=0x1ab7040, ts=0x1ab6fb0, b=0x1ab6f80,
rollforward_updates=0x7ffff009fc8d <rollforward_update_schema>,
rollforward_creates=0x7ffff009f692 <rollforward_create_schema>,
rollforward_deletes=0x7ffff009f602 <rollforward_drop_schema>,
fd=0x7ffff009dbb5 <schema_dup>, mode=3) at store.c:2374
10 0x00007ffff00a0038 in rollforward_trans (tr=0x1ab7010, mode=3)
at store.c:2859
11 0x00007ffff00a1552 in sql_trans_commit (tr=0x1ab7010) at store.c:3275
12 0x00007ffff001430a in mvc_commit (m=0x7fffe0005430, chain=0, name=0x0)
at sql_mvc.c:269
13 0x00007fffeff813c5 in SQLautocommit (c=0x6425e0, m=0x7fffe0005430)
at sql_scenario.c:375
14 0x00007fffeff89252 in SQLreader (c=0x6425e0) at sql_scenario.c:2064
15 0x00007ffff78f7d85 in runPhase (c=0x6425e0, phase=0) at mal_scenario.c:526
16 0x00007ffff78f7ecd in runScenarioBody (c=0x6425e0) at mal_scenario.c:556
17 0x00007ffff78f80c3 in runScenario (c=0x6425e0) at mal_scenario.c:590
18 0x00007ffff78f9a80 in MSserveClient (dummy=0x6425e0) at mal_session.c:459
19 0x00007ffff7395964 in thread_starter (arg=0x7fffe8000a20)
at gdk_system.c:535
20 0x00007ffff60ad0a4 in start_thread (arg=0x7fffee774700)
at pthread_create.c:309
21 0x00007ffff5de1cbd in clone ()
at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111
(gdb) f 4
4 0x00007ffff00b3865 in tr_update_delta (tr=0x1ab7010, obat=0x7fffe013e500,
cbat=0x7fffe022e9b0, unique=0) at bat_storage.c:1722
1722 assert((BATcount(cur) + BATcount(ins)) == cbat->cnt);
(gdb) p *cur->S
$1 = {tid = 140737194182401, stamp = 3475, copiedtodisk = 0, dirty = 1,
dirtyflushed = 0, descdirty = 1, restricted = 1, persistence = 1, role = 0,
unused = 0, sharecnt = 0, map_head = 0 '\000', map_tail = 0 '\000',
map_hheap = 0 '\000', map_theap = 0 '\000', deleted = 0, first = 0,
inserted = 0, count = 0, capacity = 1024}
(gdb) p *ins->S
$2 = {tid = 140737194182401, stamp = 3783, copiedtodisk = 0, dirty = 1,
dirtyflushed = 0, descdirty = 1, restricted = 0, persistence = 1, role = 0,
unused = 0, sharecnt = 0, map_head = 0 '\000', map_tail = 0 '\000',
map_hheap = 0 '\000', map_theap = 0 '\000', deleted = 0, first = 0,
inserted = 0, count = 3, capacity = 256}
(gdb) p *cbat
$3 = {name = 0x7fffe0233c40 "sys_foo_a", bid = 681, ibase = 3, ibid = 676,
ubid = 17, cnt = 6, ucnt = 0, cached = 0x0, wtime = 21, next = 0x0}
> Attached file: [acidityfail2.py](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3635_acidityfail2.py_315) (application/octet-stream, 995 bytes)
> Description: Reproduction script, causes SIGABRT
## Comment 20589
Date: 2015-01-28 14:13:12 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [1c94812f26f6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=1c94812f26f6) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=1c94812f26f6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=1c94812f26f6)
Changeset description:
make sure we reset the cleared flag also on failed transactions (fixes Bug #3635 part 2)
## Comment 20590
Date: 2015-01-28 14:17:23 +0100
From: @njnes
fixed the abort, ie not sure this fixes your new crash.
## Comment 20592
Date: 2015-01-28 18:32:54 +0100
From: Richard Hughes <<richard.monetdb>>
Looking good. I've re-run everything and haven't managed to crash it again; thanks for your patches.
| data loss: table reverts to an older version after restart | https://api.github.com/repos/MonetDB/MonetDB/issues/3635/comments | 0 | 2020-11-30T12:58:06Z | 2024-06-27T12:05:49Z | https://github.com/MonetDB/MonetDB/issues/3635 | 753,457,414 | 3,635 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-12-02 19:05:19 +0100
From: Jerry Evans <<g4>>
To: buildtools devs <<bugs-buildtools>>
Version: 11.19.7 (Oct2014-SP1)
CC: g4
Last updated: 2014-12-05 12:47:39 +0100
## Comment 20505
Date: 2014-12-02 19:05:19 +0100
From: Jerry Evans <<g4>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/39.0.2171.65 Safari/537.36
Build Identifier:
-------------------------snip--------------------------
copy ".\..\..\..\monetdb5\scheduler\run_isolate.mal"
"run_isolate.mal"
1 file(s) copied.
if not exist "tools" mkdir "tools"
copy ".\..\..\monetdb5\tools\Makefile.msc" "tools\Makefile"
1 file(s) copied.
cd "tools" && U:\vs2013\VC\BIN\nmake.exe /nologo DEBUG=1
"prefix=R:\src\
libraries\MonetDB-11.19.3\bin" "bits=32" all
NMAKE : fatal error U1073: don't know how to make '..\..\gdk\libbat.lib'
Stop.
NMAKE : fatal error U1077: 'cd' : return code '0x2'
Stop.
NMAKE : fatal error U1077: 'cd' : return code '0x2'
Stop.
NMAKE : fatal error U1077: 'U:\vs2013\VC\BIN\nmake.exe' : return code '0x2'
Stop.
R:\src\libraries\MonetDB-11.19.3\bin>
-------------------------snip--------------------------
Reproducible: Always
### Steps to Reproduce:
The following is contained in a buildit.cmd file so all dependent library paths and include folders are present. Then it invokes nmake:
SET LIBOPENSSL=U:\pkg\OpenSSL\Win32
SET LIBICONV=U:\pkg\monet\iconv-1.9.2.win32
SET LIBZLIB=U:\pkg\monet\zlib-1.2.5
SET LIBXML2=U:\pkg\monet\libxml2-2.7.8.win32
SET LIBPCRE=U:\pkg\monet\pcre-7.0-bin
SET INCLUDE=%INCLUDE%;U:\pkg\monet\iconv-1.9.2.win32\include;U:\pkg\monet\pcre-7.0-bin\include
nmake /D -f ..\NT\makefile DEBUG=1 HAVE_OPENSSL=1 HAVE_SQL=1 HAVE_MONETDB5=1 all
### Actual Results:
NMAKE : fatal error U1073: don't know how to make '..\..\gdk\libbat.lib'
### Expected Results:
nmake runs to completion
Root folder is R:\src\libraries\MonetDB-11.19.3.
Building in R:\src\libraries\MonetDB-11.19.3\bin
The build is failing at the point shown below. What have I missed here? The
build folder is a sibling of /NT as indicated in the windowsbuild.rst ...?
I would really appreciate it if the team could offer some guidance here. I am trying to build a version from the very latest code to avoid the last bug I reported here: https://www.monetdb.org/bugzilla/show_bug.cgi?id=3628
TIA
Jerry (g4@novadsp.com)
## Comment 20506
Date: 2014-12-03 09:05:41 +0100
From: @sjoerdmullender
You need to have HAVE_GDK=1 on the nmake command line.
Given the libraries you list, you probably also want HAVE_ICONV=1 HAVE_LIBZ=1 HAVE_PCRE=1
## Comment 20512
Date: 2014-12-04 19:34:11 +0100
From: Jerry Evans <<g4>>
Thank you sjoerd. Your suggestion is indeed the right one. What is the best forum to discuss further issues with compilation? Does anyone use the developers mailing list?
TIA.
## Comment 20513
Date: 2014-12-04 22:23:04 +0100
From: @sjoerdmullender
(In reply to comment 2)
> Thank you sjoerd. Your suggestion is indeed the right one. What is the best
> forum to discuss further issues with compilation? Does anyone use the
> developers mailing list?
>
> TIA.
Yes, that's an active list and appropriate for this kind of thing.
## Comment 20515
Date: 2014-12-05 12:47:39 +0100
From: Jerry Evans <<g4>>
Hello Sjoerd.
All of my posts are apparently being rejected as spam.
-------------------------
Your message did not reach some or all of the intended recipients.
Sent: Thu, 4 Dec 2014 19:02:20 -0000
Subject: Apologies: testing
The following recipient(s) could not be reached:
developers-list@monetdb.org
Error Type: SMTP
Remote server (192.16.197.211) issued an error.
hMailServer sent: .
Remote server replied: 550 5.7.1 Blocked by SpamAssassin
---------------------------
and
---------------------------
Sent: Sat, 29 Nov 2014 15:22:48 -0000
Subject: Minor documentation suggestion ...
The following recipient(s) could not be reached:
users-list@monetdb.org
Error Type: SMTP
Remote server (192.16.197.211) issued an error.
hMailServer sent: .
Remote server replied: 550 5.7.1 Blocked by SpamAssassin
| nmake fails attempting to build 32 bit server | https://api.github.com/repos/MonetDB/MonetDB/issues/3634/comments | 0 | 2020-11-30T12:58:03Z | 2024-06-28T13:15:31Z | https://github.com/MonetDB/MonetDB/issues/3634 | 753,457,386 | 3,634 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-12-01 18:11:27 +0100
From: Richard Hughes <<richard.monetdb>>
To: SQL devs <<bugs-sql>>
Version: 11.19.3 (Oct2014)
CC: @njnes
Last updated: 2015-01-29 14:07:29 +0100
## Comment 20502
Date: 2014-12-01 18:11:27 +0100
From: Richard Hughes <<richard.monetdb>>
sql>set optimizer='minimal_pipe';
operation successful (1.050ms)
sql>create table foo (a int, b int);
operation successful (1.886ms)
sql>insert into foo values (1,1);
1 affected row (1.452ms)
sql>select a from foo group by a having sum(b)>count(*)*.5;
+---+
| a |
+===+
+---+
0 tuples (2.002ms)
Expected result:
+---+
| a |
+===+
| 1 |
+---+
The same calculation in the select clause gets it right:
sql>select a,sum(b)>count(*)*.5 from foo group by a;
+------+-------+
| a | >_L1 |
+======+=======+
| 1 | true |
+------+-------+
1 tuple (2.039ms)
sql>explain select a from foo group by a having sum(b)>count(*)*.5;
+-----------------------------------------------------------------------------+
| mal |
+=============================================================================+
| function user.s46_1{autoCommit=true}(A0:bte):void; |
| X_3 := sql.mvc(); |
| X_4:bat[:oid,:oid] := sql.tid(X_3,"sys","foo"); |
| X_7:bat[:oid,:int] := sql.bind(X_3,"sys","foo","b",0); |
| (X_10,r1_10) := sql.bind(X_3,"sys","foo","b",2); |
| X_13:bat[:oid,:int] := sql.bind(X_3,"sys","foo","b",1); |
| X_15 := sql.delta(X_7,X_10,r1_10,X_13); |
| X_16 := algebra.leftfetchjoin(X_4,X_15); |
| X_17:bat[:oid,:int] := sql.bind(X_3,"sys","foo","a",0); |
| (X_19,r1_19) := sql.bind(X_3,"sys","foo","a",2); |
| X_21:bat[:oid,:int] := sql.bind(X_3,"sys","foo","a",1); |
| X_22 := sql.delta(X_17,X_19,r1_19,X_21); |
| X_23 := algebra.leftfetchjoin(X_4,X_22); |
| (X_24,r1_24,r2_24) := group.subgroupdone(X_23); |
| X_27:bat[:oid,:lng] := aggr.subsum(X_16,X_24,r1_24,true,true); |
| X_28:bat[:oid,:wrd] := aggr.subcount(X_24,X_24,r1_24,false); |
| X_30:bat[:oid,:lng] := batcalc.lng(X_28,19,1); |
| X_34 := A0; |
| X_35 := calc.lng(1,X_34,19,1); |
| X_36:bat[:oid,:lng] := batcalc.*(X_30,X_35); |
| X_38:bat[:oid,:lng] := batcalc.lng(2,X_36); |
| X_39:bat[:oid,:bit] := batcalc.>(X_27,X_38); |
| X_41 := algebra.subselect(X_39,true,true,true,true,false); |
| X_43 := algebra.leftfetchjoin(r1_24,X_23); |
| X_44 := algebra.leftfetchjoin(X_41,X_43); |
| X_45 := sql.resultSet(1,1,X_44); |
| sql.rsColumn(X_45,"sys.foo","a","int",32,0,X_44); |
| X_51 := io.stdout(); |
| sql.exportResult(X_51,X_45); |
| end s46_1; |
| querylog.define("select a from foo group by a having sum(b)>count(*)*.5;" |
: ,"minimal_pipe") :
+-----------------------------------------------------------------------------+
The batcalc.lng calls don't look quite right to me.
## Comment 20503
Date: 2014-12-01 19:03:17 +0100
From: Richard Hughes <<richard.monetdb>>
Created attachment 312
Proposed fix
How's this for a fix? That code is scary-complicated.
> Attached file: [having-float-wrong-result.patch](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3633_having-float-wrong-result.patch_312) (text/plain, 510 bytes)
> Description: Proposed fix
## Comment 20507
Date: 2014-12-03 13:13:36 +0100
From: @njnes
Added a test. incorrect_floating_point_cast.Bug-3633
Fixed by proper type checks
## Comment 20508
Date: 2014-12-03 13:15:35 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [d9da821f8731](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d9da821f8731) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=d9da821f8731](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=d9da821f8731)
Changeset description:
fixed bug #3633
ie properly check types on compare expressions
made string a higher type in auto coersion.
## Comment 20598
Date: 2015-01-29 14:07:29 +0100
From: @sjoerdmullender
Oct2014-SP2 has been released.
| Wrong result for HAVING with floating-point constant | https://api.github.com/repos/MonetDB/MonetDB/issues/3633/comments | 0 | 2020-11-30T12:58:00Z | 2024-06-27T12:05:47Z | https://github.com/MonetDB/MonetDB/issues/3633 | 753,457,349 | 3,633 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-28 12:56:15 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: buildtools devs <<bugs-buildtools>>
Version: -- development
Last updated: 2015-01-29 14:07:32 +0100
## Comment 20475
Date: 2014-11-28 12:56:15 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:33.0) Gecko/20100101 Firefox/33.0
Build Identifier:
Making clean in ruby
make[2]: Entering directory `/ufs/dinther/BUILD/clients/ruby'
Making clean in adapter
make[3]: Entering directory `/ufs/dinther/BUILD/clients/ruby/adapter'
rm -rf .libs _libs
test -z "activerecord-monetdb-adapter-0.1.gem" || rm -f activerecord-monetdb-adapter-0.1.gem
rm -f *.lo
[ '/ufs/dinther/dev/dev/clients/ruby/adapter' -ef . ] || rm -f 'lib/active_record/connection_adapters/monetdb_adapter.rb'
[ '/ufs/dinther/dev/dev/clients/ruby/adapter' -ef . ] || rm -f 'activerecord-monetdb-adapter-0.1.gemspec'
[ '/ufs/dinther/dev/dev/clients/ruby/adapter' -ef . ] || rmdir 'lib/active_record/connection_adapters'
rmdir: failed to remove ‘lib/active_record/connection_adapters’: No such file or directory
make[3]: *** [mostlyclean-local-activerecord-monetdb-adapter-0_1_gemspec] Error 1
make[3]: Leaving directory `/ufs/dinther/BUILD/clients/ruby/adapter'
make[2]: *** [clean-recursive] Error 1
make[2]: Leaving directory `/ufs/dinther/BUILD/clients/ruby'
make[1]: *** [clean-recursive] Error 1
make[1]: Leaving directory `/ufs/dinther/BUILD/clients'
make: *** [clean-recursive] Error 1
bash-4.2$
Reproducible: Always
### Steps to Reproduce:
1. cd BUILD
2. make clean
3. make clean
### Actual Results:
first run of make clean finished normally.
second run:
bash-4.2$ make clean
Making clean in buildtools
make[1]: Entering directory `/ufs/dinther/BUILD/buildtools'
Making clean in conf
make[2]: Entering directory `/ufs/dinther/BUILD/buildtools/conf'
rm -rf .libs _libs
test -z "" || rm -f
rm -f *.lo
make[2]: Leaving directory `/ufs/dinther/BUILD/buildtools/conf'
make[2]: Entering directory `/ufs/dinther/BUILD/buildtools'
rm -rf .libs _libs
test -z "" || rm -f
rm -f *.lo
make[2]: Leaving directory `/ufs/dinther/BUILD/buildtools'
make[1]: Leaving directory `/ufs/dinther/BUILD/buildtools'
Making clean in common
make[1]: Entering directory `/ufs/dinther/BUILD/common'
Making clean in stream
make[2]: Entering directory `/ufs/dinther/BUILD/common/stream'
rm -rf .libs _libs
test -z "libstream.la" || rm -f libstream.la
rm -f ./so_locations
rm -f *.o
test -z "monetdb-stream.pc" || rm -f monetdb-stream.pc
rm -f *.lo
make[2]: Leaving directory `/ufs/dinther/BUILD/common/stream'
Making clean in options
make[2]: Entering directory `/ufs/dinther/BUILD/common/options'
rm -rf .libs _libs
test -z "libmoptions.la" || rm -f libmoptions.la
rm -f ./so_locations
rm -f *.o
test -z "" || rm -f
rm -f *.lo
make[2]: Leaving directory `/ufs/dinther/BUILD/common/options'
Making clean in utils
make[2]: Entering directory `/ufs/dinther/BUILD/common/utils'
rm -rf .libs _libs
test -z "libmcrypt.la libmutils.la libmsabaoth.la libmuuid.la" || rm -f libmcrypt.la libmutils.la libmsabaoth.la libmuuid.la
rm -f ./so_locations
rm -f *.o
test -z "" || rm -f
rm -f *.lo
make[2]: Leaving directory `/ufs/dinther/BUILD/common/utils'
make[2]: Entering directory `/ufs/dinther/BUILD/common'
rm -rf .libs _libs
test -z "" || rm -f
rm -f *.lo
make[2]: Leaving directory `/ufs/dinther/BUILD/common'
make[1]: Leaving directory `/ufs/dinther/BUILD/common'
Making clean in clients
make[1]: Entering directory `/ufs/dinther/BUILD/clients'
Making clean in mapilib
make[2]: Entering directory `/ufs/dinther/BUILD/clients/mapilib'
rm -rf .libs _libs
test -z "libmapi.la" || rm -f libmapi.la
rm -f ./so_locations
rm -f *.o
test -z "monetdb-mapi.pc" || rm -f monetdb-mapi.pc
rm -f *.lo
make[2]: Leaving directory `/ufs/dinther/BUILD/clients/mapilib'
Making clean in mapiclient
make[2]: Entering directory `/ufs/dinther/BUILD/clients/mapiclient'
rm -f mclient msqldump stethoscope tomograph
rm -rf .libs _libs
test -z "libmcutil.la" || rm -f libmcutil.la
rm -f ./so_locations
rm -f *.o
test -z "" || rm -f
rm -f *.lo
make[2]: Leaving directory `/ufs/dinther/BUILD/clients/mapiclient'
Making clean in odbc
make[2]: Entering directory `/ufs/dinther/BUILD/clients/odbc'
Making clean in setup
make[3]: Entering directory `/ufs/dinther/BUILD/clients/odbc/setup'
test -z "libMonetODBCs.la" || rm -f libMonetODBCs.la
rm -f ./so_locations
rm -rf .libs _libs
rm -f *.o
test -z "" || rm -f
rm -f *.lo
make[3]: Leaving directory `/ufs/dinther/BUILD/clients/odbc/setup'
Making clean in driver
make[3]: Entering directory `/ufs/dinther/BUILD/clients/odbc/driver'
test -z "libMonetODBC.la" || rm -f libMonetODBC.la
rm -f ./so_locations
rm -rf .libs _libs
rm -f *.o
test -z "ODBC.def" || rm -f ODBC.def
rm -f *.lo
make[3]: Leaving directory `/ufs/dinther/BUILD/clients/odbc/driver'
Making clean in samples
make[3]: Entering directory `/ufs/dinther/BUILD/clients/odbc/samples'
rm -f arraytest odbcsample1 testgetinfo
rm -rf .libs _libs
rm -f
rm -f *.o
test -z "" || rm -f
rm -f *.lo
make[3]: Leaving directory `/ufs/dinther/BUILD/clients/odbc/samples'
make[3]: Entering directory `/ufs/dinther/BUILD/clients/odbc'
rm -rf .libs _libs
test -z "" || rm -f
rm -f *.lo
make[3]: Leaving directory `/ufs/dinther/BUILD/clients/odbc'
make[2]: Leaving directory `/ufs/dinther/BUILD/clients/odbc'
Making clean in perl
make[2]: Entering directory `/ufs/dinther/BUILD/clients/perl'
Making clean in MonetDB-CLI-MapiPP
make[3]: Entering directory `/ufs/dinther/BUILD/clients/perl/MonetDB-CLI-MapiPP'
Making clean in MonetDB
make[4]: Entering directory `/ufs/dinther/BUILD/clients/perl/MonetDB-CLI-MapiPP/MonetDB'
Making clean in CLI
make[5]: Entering directory `/ufs/dinther/BUILD/clients/perl/MonetDB-CLI-MapiPP/MonetDB/CLI'
rm -rf .libs _libs
test -z "" || rm -f
rm -f *.lo
make[5]: Leaving directory `/ufs/dinther/BUILD/clients/perl/MonetDB-CLI-MapiPP/MonetDB/CLI'
make[5]: Entering directory `/ufs/dinther/BUILD/clients/perl/MonetDB-CLI-MapiPP/MonetDB'
rm -rf .libs _libs
test -z "" || rm -f
rm -f *.lo
make[5]: Leaving directory `/ufs/dinther/BUILD/clients/perl/MonetDB-CLI-MapiPP/MonetDB'
make[4]: Leaving directory `/ufs/dinther/BUILD/clients/perl/MonetDB-CLI-MapiPP/MonetDB'
make[4]: Entering directory `/ufs/dinther/BUILD/clients/perl/MonetDB-CLI-MapiPP'
rm -rf .libs _libs
test -z "" || rm -f
rm -f *.lo
make[4]: Leaving directory `/ufs/dinther/BUILD/clients/perl/MonetDB-CLI-MapiPP'
make[3]: Leaving directory `/ufs/dinther/BUILD/clients/perl/MonetDB-CLI-MapiPP'
Making clean in MonetDB-CLI
make[3]: Entering directory `/ufs/dinther/BUILD/clients/perl/MonetDB-CLI'
Making clean in MonetDB
make[4]: Entering directory `/ufs/dinther/BUILD/clients/perl/MonetDB-CLI/MonetDB'
rm -rf .libs _libs
test -z "" || rm -f
rm -f *.lo
make[4]: Leaving directory `/ufs/dinther/BUILD/clients/perl/MonetDB-CLI/MonetDB'
make[4]: Entering directory `/ufs/dinther/BUILD/clients/perl/MonetDB-CLI'
rm -rf .libs _libs
test -z "" || rm -f
rm -f *.lo
make[4]: Leaving directory `/ufs/dinther/BUILD/clients/perl/MonetDB-CLI'
make[3]: Leaving directory `/ufs/dinther/BUILD/clients/perl/MonetDB-CLI'
Making clean in DBD
make[3]: Entering directory `/ufs/dinther/BUILD/clients/perl/DBD'
Making clean in monetdb
make[4]: Entering directory `/ufs/dinther/BUILD/clients/perl/DBD/monetdb'
rm -rf .libs _libs
test -z "" || rm -f
rm -f *.lo
make[4]: Leaving directory `/ufs/dinther/BUILD/clients/perl/DBD/monetdb'
make[4]: Entering directory `/ufs/dinther/BUILD/clients/perl/DBD'
rm -rf .libs _libs
test -z "" || rm -f
rm -f *.lo
make[4]: Leaving directory `/ufs/dinther/BUILD/clients/perl/DBD'
make[3]: Leaving directory `/ufs/dinther/BUILD/clients/perl/DBD'
Making clean in Tests
make[3]: Entering directory `/ufs/dinther/BUILD/clients/perl/Tests'
rm -rf .libs _libs
test -z "malsample.pl sqlsample.pl" || rm -f malsample.pl sqlsample.pl
rm -f *.lo
make[3]: Leaving directory `/ufs/dinther/BUILD/clients/perl/Tests'
make[3]: Entering directory `/ufs/dinther/BUILD/clients/perl'
rm -rf .libs _libs
test -z "" || rm -f
rm -f *.lo
make[3]: Leaving directory `/ufs/dinther/BUILD/clients/perl'
make[2]: Leaving directory `/ufs/dinther/BUILD/clients/perl'
Making clean in php
make[2]: Entering directory `/ufs/dinther/BUILD/clients/php'
Making clean in lib
make[3]: Entering directory `/ufs/dinther/BUILD/clients/php/lib'
rm -rf .libs _libs
test -z "" || rm -f
rm -f *.lo
make[3]: Leaving directory `/ufs/dinther/BUILD/clients/php/lib'
make[3]: Entering directory `/ufs/dinther/BUILD/clients/php'
rm -rf .libs _libs
test -z "" || rm -f
rm -f *.lo
make[3]: Leaving directory `/ufs/dinther/BUILD/clients/php'
make[2]: Leaving directory `/ufs/dinther/BUILD/clients/php'
Making clean in ruby
make[2]: Entering directory `/ufs/dinther/BUILD/clients/ruby'
Making clean in adapter
make[3]: Entering directory `/ufs/dinther/BUILD/clients/ruby/adapter'
rm -rf .libs _libs
test -z "activerecord-monetdb-adapter-0.1.gem" || rm -f activerecord-monetdb-adapter-0.1.gem
rm -f *.lo
[ '/ufs/dinther/dev/dev/clients/ruby/adapter' -ef . ] || rm -f 'lib/active_record/connection_adapters/monetdb_adapter.rb'
[ '/ufs/dinther/dev/dev/clients/ruby/adapter' -ef . ] || rm -f 'activerecord-monetdb-adapter-0.1.gemspec'
[ '/ufs/dinther/dev/dev/clients/ruby/adapter' -ef . ] || rmdir 'lib/active_record/connection_adapters'
rmdir: failed to remove ‘lib/active_record/connection_adapters’: No such file or directory
make[3]: *** [mostlyclean-local-activerecord-monetdb-adapter-0_1_gemspec] Error 1
make[3]: Leaving directory `/ufs/dinther/BUILD/clients/ruby/adapter'
make[2]: *** [clean-recursive] Error 1
make[2]: Leaving directory `/ufs/dinther/BUILD/clients/ruby'
make[1]: *** [clean-recursive] Error 1
make[1]: Leaving directory `/ufs/dinther/BUILD/clients'
make: *** [clean-recursive] Error 1
bash-4.2$
### Expected Results:
no error when make clean is run multiple times
## Comment 20487
Date: 2014-11-28 18:52:59 +0100
From: Martin van Dinther <<martin.van.dinther>>
I have done a run where I started mserver5 with --trace, so using:
./INSTALL/bin/mserver5 --trace
The trace output file is too large to attach (3MB), but you can read it from /ufs/dinther/trace_2014_11_28.out
I used SQuirreL 3.6 (http://www.squirrelsql.org/installation) to issue the following SQL queries:
1) SELECT id, name, schema_id, query, type, system, commit_action, readonly, temporary FROM tables WHERE name like '_tables';
-- this should return 2 rows.
2) select user
-- this should return 1 row with value: monetdb
mclient will not reproduce this behavior.
I also tested it via a dedicated Java test program, but could it also not reproduce.
So please download SQuirreL 3.6 (or copy it from /ufs/dinther/Downloads/squirrel-sql-3.6-standard.jar), install via:
java -jar squirrel-sql-3.6-standard.jar
Next start SQuirreL 3.6 via:
~/squirrel-sql-3.6/./squirrel-sql.sh
setup a Driver and create a Connection Alias, see https://www.monetdb.org/Documentation/Cookbooks/SQLrecipies/Clients/SQuirreL for instructions
## Comment 20489
Date: 2014-11-28 18:54:57 +0100
From: Martin van Dinther <<martin.van.dinther>>
ignore the previous comment 1. it belongs to bug #3631.
## Comment 20561
Date: 2015-01-20 10:53:04 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [1f0944853528](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=1f0944853528) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=1f0944853528](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=1f0944853528)
Changeset description:
Don't try to remove non-existent directories.
This fixes bug #3632.
## Comment 20600
Date: 2015-01-29 14:07:32 +0100
From: @sjoerdmullender
Oct2014-SP2 has been released.
| running make clean twice gives an error in clients/ruby/adapter | https://api.github.com/repos/MonetDB/MonetDB/issues/3632/comments | 0 | 2020-11-30T12:57:57Z | 2024-06-27T12:05:46Z | https://github.com/MonetDB/MonetDB/issues/3632 | 753,457,306 | 3,632 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-28 12:48:56 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: GDK devs <<bugs-common>>
Version: -- development
CC: @mlkersten, @drstmane
Last updated: 2014-12-05 13:16:49 +0100
## Comment 20473
Date: 2014-11-28 12:48:56 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:33.0) Gecko/20100101 Firefox/33.0
Build Identifier:
Normally the query "select * from tables" is executed in 15 to 20 msecs.
However if you create a server using
configure --prefix=$HOME/INSTALL --enable-assert --disable-debug
and started as:
INSTALL/bin/mserver5 -d10 --set embedded_r=true
then the same query is executed the first time in 9 secs and the second time in 49 secs.
the mserver5 console also outputs a lot of:
delay initial 4[4774451407447032] memory 16894527376[13190987776.000000]
delay initial 4[4774451407451105] memory 16894527488[13190987776.000000]
delay initial 4[291] memory 16894552840[13190987776.000000]
delay initial 3[407] memory 16894554920[13190987776.000000]
delay initial 3[429] memory 16894566544[13190987776.000000]
It also looks a lot of CPU time is used (and possibly memory) as the whole computer is barely responding.
Eventually I had to kill the mserver5 process.
Reproducible: Always
### Steps to Reproduce:
1. hg update -r default
2. configure --prefix=$HOME/INSTALL --enable-assert --disable-debug #3. make
4. make install
5. start INSTALL/bin/mserver5 -d10 --set embedded_r=true
6. start mclient
7. execute query: select * from tables;
8. execute same query again: select * from tables;
### Actual Results:
bash-4.2$ ./start_INSTALL_mserver5.sh
builtin opt gdk_dbpath = /ufs/dinther/INSTALL/var/monetdb5/dbfarm/demo
builtin opt gdk_debug = 0
builtin opt gdk_vmtrim = no
builtin opt monet_prompt = >
builtin opt monet_daemon = no
builtin opt mapi_port = 50000
builtin opt mapi_open = false
builtin opt mapi_autosense = false
builtin opt sql_optimizer = default_pipe
builtin opt sql_debug = 0
cmdline opt embedded_r = true
cmdline opt gdk_debug = 10
MonetDB 5 server v11.20.0
This is an unreleased version
Serving database 'demo', using 8 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs and 128bit integers dynamically linked
Found 15.356 GiB available main-memory.
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2014 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Listening for connection requests on mapi:monetdb://127.0.0.1:50000/
MonetDB/GIS module loaded
MonetDB/SQL module loaded
MonetDB/R module loaded
> SQL catalog created, loading sql scripts once
loading sql script: 09_like.sql
loading sql script: 10_math.sql
loading sql script: 11_times.sql
loading sql script: 12_url.sql
loading sql script: 13_date.sql
loading sql script: 14_inet.sql
loading sql script: 15_querylog.sql
loading sql script: 16_tracelog.sql
loading sql script: 19_cluster.sql
loading sql script: 20_vacuum.sql
loading sql script: 21_dependency_functions.sql
loading sql script: 22_clients.sql
loading sql script: 23_skyserver.sql
loading sql script: 24_zorder.sql
loading sql script: 25_debug.sql
loading sql script: 26_sysmon.sql
loading sql script: 39_analytics.sql
loading sql script: 39_analytics_hge.sql
loading sql script: 40_geom.sql
loading sql script: 40_json.sql
loading sql script: 40_json_hge.sql
loading sql script: 41_jsonstore.sql
loading sql script: 45_uuid.sql
loading sql script: 46_gsl.sql
loading sql script: 75_storagemodel.sql
loading sql script: 80_statistics.sql
loading sql script: 80_udf.sql
loading sql script: 80_udf_hge.sql
loading sql script: 89_generator_hge.sql
loading sql script: 90_generator.sql
loading sql script: 99_system.sql
delay initial 4[4774451407447032] memory 16894527376[13190987776.000000]
delay initial 4[4774451407451105] memory 16894527488[13190987776.000000]
delay initial 4[291] memory 16894552840[13190987776.000000]
delay initial 3[407] memory 16894554920[13190987776.000000]
delay initial 3[429] memory 16894566544[13190987776.000000]
delay initial 2[431] memory 16894568048[13190987776.000000]
delay initial 1[464] memory 16894568672[13190987776.000000]
delay initial 1[472] memory 16894568672[13190987776.000000]
delay initial 1[479] memory 16894569008[13190987776.000000]
delay initial 3[633] memory 16894570880[13190987776.000000]
delay initial 3[633] memory 16894570880[13190987776.000000]
delay initial 2[633] memory 16894570880[13190987776.000000]
delay initial 1[633] memory 16894570880[13190987776.000000]
delay initial 1[658] memory 16894581544[13190987776.000000]
delay initial 1[775] memory 16894581544[13190987776.000000]
delay initial 2[777] memory 16894581544[13190987776.000000]
delay initial 3[837] memory 16894581544[13190987776.000000]
delay initial 2[870] memory 16894582536[13190987776.000000]
delay initial 1[870] memory 16894582536[13190987776.000000]
delay initial 1[870] memory 16894582536[13190987776.000000]
delay initial 1[1003] memory 16894609920[13190987776.000000]
delay initial 1[1036] memory 16894610328[13190987776.000000]
delay initial 1[1040] memory 16894620584[13190987776.000000]
delay initial 2[1055] memory 16894621080[13190987776.000000]
delay initial 3[1498] memory 16894622648[13190987776.000000]
delay initial 3[1550] memory 16894622648[13190987776.000000]
delay initial 3[2416] memory 16894622648[13190987776.000000]
delay initial 3[2500] memory 16894621656[13190987776.000000]
delay initial 3[2500] memory 16894623192[13190987776.000000]
delay initial 1[2500] memory 16894623192[13190987776.000000]
delay initial 1[2500] memory 16894624696[13190987776.000000]
delay initial 2[2500] memory 16894623192[13190987776.000000]
delay initial 3[3031] memory 16894611912[13190987776.000000]
delay initial 3[3032] memory 16894610920[13190987776.000000]
delay initial 3[3107] memory 16894609624[13190987776.000000]
delay initial 2[3107] memory 16894609624[13190987776.000000]
delay initial 1[3124] memory 16894619864[13190987776.000000]
delay initial 2[3735] memory 16894618904[13190987776.000000]
delay initial 3[3817] memory 16894617912[13190987776.000000]
delay initial 3[3851] memory 16894617416[13190987776.000000]
delay initial 2[3865] memory 16894617384[13190987776.000000]
delay initial 4[5544] memory 16894581768[13190987776.000000]
delay initial 4[9607] memory 16894572048[13190987776.000000]
delay initial 4[13671] memory 16894572096[13190987776.000000]
delay initial 4[17734] memory 16894572144[13190987776.000000]
delay initial 4[21797] memory 16894572200[13190987776.000000]
delay initial 4[25860] memory 16894572248[13190987776.000000]
delay initial 4[29924] memory 16894572296[13190987776.000000]
delay initial 4[33986] memory 16894572344[13190987776.000000]
delay initial 4[38051] memory 16894572400[13190987776.000000]
delay initial 4[42116] memory 16894572456[13190987776.000000]
delay initial 4[46182] memory 16894572512[13190987776.000000]
delay initial 4[50280] memory 16894572512[13190987776.000000]
delay initial 4[54596] memory 16894530792[13190987776.000000]
delay initial 3[76] memory 22515002896[13190987776.000000]
delay initial 3[76] memory 22515002896[13190987776.000000]
delay initial 4[24] memory 22514990936[13190987776.000000]
delay initial 2[123] memory 22515003952[13190987776.000000]
delay initial 1[123] memory 22515004752[13190987776.000000]
delay initial 1[125] memory 22515006320[13190987776.000000]
delay initial 1[125] memory 22515017480[13190987776.000000]
delay initial 1[127] memory 22515019048[13190987776.000000]
delay initial 1[130] memory 22515020904[13190987776.000000]
delay initial 3[161] memory 22515033064[13190987776.000000]
delay initial 2[161] memory 22515034920[13190987776.000000]
delay initial 2[161] memory 22515034920[13190987776.000000]
delay initial 1[161] memory 22515035768[13190987776.000000]
delay initial 1[161] memory 22515048480[13190987776.000000]
delay initial 1[175] memory 22515049880[13190987776.000000]
delay initial 3[215] memory 22515061704[13190987776.000000]
delay initial 3[232] memory 22515062008[13190987776.000000]
delay initial 3[245] memory 22515062008[13190987776.000000]
delay initial 3[347] memory 22515062008[13190987776.000000]
delay initial 3[395] memory 22515061016[13190987776.000000]
delay initial 3[395] memory 22515062552[13190987776.000000]
delay initial 2[395] memory 22515062552[13190987776.000000]
delay initial 1[395] memory 22515062552[13190987776.000000]
delay initial 1[395] memory 22515063544[13190987776.000000]
delay initial 1[479] memory 22515043264[13190987776.000000]
delay initial 2[480] memory 22515041536[13190987776.000000]
delay initial 1[480] memory 22515042336[13190987776.000000]
delay initial 2[482] memory 22515040848[13190987776.000000]
delay initial 1[482] memory 22515040848[13190987776.000000]
delay initial 1[581] memory 22515040352[13190987776.000000]
delay initial 1[581] memory 22515040352[13190987776.000000]
delay initial 1[582] memory 22515040352[13190987776.000000]
delay initial 1[582] memory 22515040352[13190987776.000000]
delay initial 1[634] memory 22515050096[13190987776.000000]
delay initial 1[634] memory 22515049600[13190987776.000000]
delay initial 1[729] memory 22515048000[13190987776.000000]
delay initial 1[730] memory 22515048000[13190987776.000000]
delay initial 2[802] memory 22515047008[13190987776.000000]
delay initial 4[1343] memory 22515020984[13190987776.000000]
delay initial 4[2436] memory 22515011264[13190987776.000000]
delay initial 4[4419] memory 22515011312[13190987776.000000]
delay initial 4[8014] memory 22515011360[13190987776.000000]
delay initial 4[12086] memory 22515011416[13190987776.000000]
delay initial 4[16158] memory 22515011464[13190987776.000000]
delay initial 4[20223] memory 22515011512[13190987776.000000]
delay initial 4[24289] memory 22515011560[13190987776.000000]
delay initial 4[28355] memory 22515011616[13190987776.000000]
delay initial 4[32422] memory 22515011544[13190987776.000000]
delay initial 4[36489] memory 22515011600[13190987776.000000]
delay initial 4[40576] memory 22515011600[13190987776.000000]
delay initial 4[44709] memory 22514969880[13190987776.000000]
bash-4.2$ mclient
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.20.0 (unreleased), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>select * from tables;
+------+------------------+------+------------------+------+-------+------+-------+------+
| id | name | sche | query | type | syste | comm | reado | temp |
: : : ma_i : : : m : it_a : nly : orar :
: : : d : : : : ctio : : y :
: : : : : : : n : : :
+======+==================+======+==================+======+=======+======+=======+======+
| 2001 | schemas | 2000 | null | 0 | true | 0 | false | 0 |
| 2007 | types | 2000 | null | 0 | true | 0 | false | 0 |
| 2016 | functions | 2000 | null | 0 | true | 0 | false | 0 |
| 2027 | args | 2000 | null | 0 | true | 0 | false | 0 |
| 2036 | sequences | 2000 | null | 0 | true | 0 | false | 0 |
| 2046 | dependencies | 2000 | null | 0 | true | 0 | false | 0 |
| 2050 | connections | 2000 | null | 0 | true | 0 | false | 0 |
| 2059 | _tables | 2000 | null | 0 | true | 0 | false | 0 |
| 2068 | _columns | 2000 | null | 0 | true | 0 | false | 0 |
| 2079 | keys | 2000 | null | 0 | true | 0 | false | 0 |
| 2086 | idxs | 2000 | null | 0 | true | 0 | false | 0 |
| 2091 | triggers | 2000 | null | 0 | true | 0 | false | 0 |
| 2102 | objects | 2000 | null | 0 | true | 0 | false | 0 |
| 2107 | _tables | 2106 | null | 0 | true | 2 | false | 0 |
| 2116 | _columns | 2106 | null | 0 | true | 2 | false | 0 |
| 2127 | keys | 2106 | null | 0 | true | 2 | false | 0 |
| 2134 | idxs | 2106 | null | 0 | true | 2 | false | 0 |
| 2139 | triggers | 2106 | null | 0 | true | 2 | false | 0 |
| 2150 | objects | 2106 | null | 0 | true | 2 | false | 0 |
| 5636 | tables | 2000 | SELECT * FROM (S | 1 | true | 0 | false | 0 |
: : : : ELECT p.*, 0 AS : : : : : :
: : : : "temporary" FROM : : : : : :
: : : : "sys"."_table...> : : : : :
| 5646 | columns | 2000 | SELECT * FROM (S | 1 | true | 0 | false | 0 |
: : : : ELECT p.* FROM " : : : : : :
: : : : sys"."_columns" : : : : : :
: : : : AS p UNION ALL...> : : : : :
| 5662 | db_user_info | 2000 | null | 0 | true | 0 | false | 0 |
| 5668 | users | 2000 | SELECT u."name" | 1 | true | 0 | false | 0 |
: : : : AS "name", ui."f : : : : : :
: : : : ullname", ui."de : : : : : :
: : : : fault_schema" ...> : : : : :
| 5672 | user_role | 2000 | null | 0 | true | 0 | false | 0 |
| 5675 | auths | 2000 | null | 0 | true | 0 | false | 0 |
| 5679 | privileges | 2000 | null | 0 | true | 0 | false | 0 |
| 5860 | querylog_catalog | 2000 | -- create table | 1 | true | 0 | false | 0 |
: : : : views for conven : : : : : :
: : : : ience : : : : : :
: : : : create view sy...> : : : : :
| 5872 | querylog_calls | 2000 | create view sys. | 1 | true | 0 | false | 0 |
: : : : querylog_calls a : : : : : :
: : : : s select * from : : : : : :
: : : : sys.querylog_c...> : : : : :
| 5890 | querylog_history | 2000 | create view sys. | 1 | true | 0 | false | 0 |
: : : : querylog_history : : : : : :
: : : : as : : : : : :
: : : : select qd.*, q...> : : : : :
| 5927 | tracelog | 2000 | create view sys. | 1 | true | 0 | false | 0 |
: : : : tracelog as sele : : : : : :
: : : : ct * from sys.tr : : : : : :
: : : : acelog(); : : : : : :
| 6052 | sessions | 2000 | create view sys. | 1 | true | 0 | false | 0 |
: : : : sessions as sele : : : : : :
: : : : ct * from sys.se : : : : : :
: : : : ssions(); : : : : : :
| 6132 | optimizers | 2000 | create view sys. | 1 | true | 0 | false | 0 |
: : : : optimizers as se : : : : : :
: : : : lect * from sys. : : : : : :
: : : : optimizers(); : : : : : :
| 6140 | environment | 2000 | create view sys. | 1 | true | 0 | false | 0 |
: : : : environment as s : : : : : :
: : : : elect * from sys : : : : : :
: : : : .environment(); : : : : : :
| 6178 | queue | 2000 | create view sys. | 1 | true | 0 | false | 0 |
: : : : queue as select : : : : : :
: : : : * from sys.queue : : : : : :
: : : : (); : : : : : :
| 6869 | storage | 2000 | create view sys. | 1 | true | 0 | false | 0 |
: : : : storage as selec : : : : : :
: : : : t * from sys.sto : : : : : :
: : : : rage(); : : : : : :
| 6881 | storagemodelinpu | 2000 | null | 0 | true | 0 | false | 0 |
: : t : : : : : : : :
| 6929 | storagemodel | 2000 | create view sys. | 1 | true | 0 | false | 0 |
: : : : storagemodel as : : : : : :
: : : : select * from sy : : : : : :
: : : : s.storagemodel...> : : : : :
| 6939 | tablestoragemode | 2000 | -- A summary of | 1 | true | 0 | false | 0 |
: : l : : the table storag : : : : : :
: : : : e requirement is : : : : : :
: : : : is available ...> : : : : :
| 6954 | statistics | 2000 | null | 0 | true | 0 | false | 0 |
| 7092 | systemfunctions | 2000 | null | 0 | true | 0 | false | 0 |
+------+------------------+------+------------------+------+-------+------+-------+------+
40 tuples (9s) !8 fields truncated!
note: to disable dropping columns and/or truncating fields use \w-1
sql>select * from tables;
+------+------------------+------+------------------+------+-------+------+-------+------+
| id | name | sche | query | type | syste | comm | reado | temp |
: : : ma_i : : : m : it_a : nly : orar :
: : : d : : : : ctio : : y :
: : : : : : : n : : :
+======+==================+======+==================+======+=======+======+=======+======+
| 2001 | schemas | 2000 | null | 0 | true | 0 | false | 0 |
| 2007 | types | 2000 | null | 0 | true | 0 | false | 0 |
| 2016 | functions | 2000 | null | 0 | true | 0 | false | 0 |
| 2027 | args | 2000 | null | 0 | true | 0 | false | 0 |
| 2036 | sequences | 2000 | null | 0 | true | 0 | false | 0 |
| 2046 | dependencies | 2000 | null | 0 | true | 0 | false | 0 |
| 2050 | connections | 2000 | null | 0 | true | 0 | false | 0 |
| 2059 | _tables | 2000 | null | 0 | true | 0 | false | 0 |
| 2068 | _columns | 2000 | null | 0 | true | 0 | false | 0 |
| 2079 | keys | 2000 | null | 0 | true | 0 | false | 0 |
| 2086 | idxs | 2000 | null | 0 | true | 0 | false | 0 |
| 2091 | triggers | 2000 | null | 0 | true | 0 | false | 0 |
| 2102 | objects | 2000 | null | 0 | true | 0 | false | 0 |
| 2107 | _tables | 2106 | null | 0 | true | 2 | false | 0 |
| 2116 | _columns | 2106 | null | 0 | true | 2 | false | 0 |
| 2127 | keys | 2106 | null | 0 | true | 2 | false | 0 |
| 2134 | idxs | 2106 | null | 0 | true | 2 | false | 0 |
| 2139 | triggers | 2106 | null | 0 | true | 2 | false | 0 |
| 2150 | objects | 2106 | null | 0 | true | 2 | false | 0 |
| 5636 | tables | 2000 | SELECT * FROM (S | 1 | true | 0 | false | 0 |
: : : : ELECT p.*, 0 AS : : : : : :
: : : : "temporary" FROM : : : : : :
: : : : "sys"."_table...> : : : : :
| 5646 | columns | 2000 | SELECT * FROM (S | 1 | true | 0 | false | 0 |
: : : : ELECT p.* FROM " : : : : : :
: : : : sys"."_columns" : : : : : :
: : : : AS p UNION ALL...> : : : : :
| 5662 | db_user_info | 2000 | null | 0 | true | 0 | false | 0 |
| 5668 | users | 2000 | SELECT u."name" | 1 | true | 0 | false | 0 |
: : : : AS "name", ui."f : : : : : :
: : : : ullname", ui."de : : : : : :
: : : : fault_schema" ...> : : : : :
| 5672 | user_role | 2000 | null | 0 | true | 0 | false | 0 |
| 5675 | auths | 2000 | null | 0 | true | 0 | false | 0 |
| 5679 | privileges | 2000 | null | 0 | true | 0 | false | 0 |
| 5860 | querylog_catalog | 2000 | -- create table | 1 | true | 0 | false | 0 |
: : : : views for conven : : : : : :
: : : : ience : : : : : :
: : : : create view sy...> : : : : :
| 5872 | querylog_calls | 2000 | create view sys. | 1 | true | 0 | false | 0 |
: : : : querylog_calls a : : : : : :
: : : : s select * from : : : : : :
: : : : sys.querylog_c...> : : : : :
| 5890 | querylog_history | 2000 | create view sys. | 1 | true | 0 | false | 0 |
: : : : querylog_history : : : : : :
: : : : as : : : : : :
: : : : select qd.*, q...> : : : : :
| 5927 | tracelog | 2000 | create view sys. | 1 | true | 0 | false | 0 |
: : : : tracelog as sele : : : : : :
: : : : ct * from sys.tr : : : : : :
: : : : acelog(); : : : : : :
| 6052 | sessions | 2000 | create view sys. | 1 | true | 0 | false | 0 |
: : : : sessions as sele : : : : : :
: : : : ct * from sys.se : : : : : :
: : : : ssions(); : : : : : :
| 6132 | optimizers | 2000 | create view sys. | 1 | true | 0 | false | 0 |
: : : : optimizers as se : : : : : :
: : : : lect * from sys. : : : : : :
: : : : optimizers(); : : : : : :
| 6140 | environment | 2000 | create view sys. | 1 | true | 0 | false | 0 |
: : : : environment as s : : : : : :
: : : : elect * from sys : : : : : :
: : : : .environment(); : : : : : :
| 6178 | queue | 2000 | create view sys. | 1 | true | 0 | false | 0 |
: : : : queue as select : : : : : :
: : : : * from sys.queue : : : : : :
: : : : (); : : : : : :
| 6869 | storage | 2000 | create view sys. | 1 | true | 0 | false | 0 |
: : : : storage as selec : : : : : :
: : : : t * from sys.sto : : : : : :
: : : : rage(); : : : : : :
| 6881 | storagemodelinpu | 2000 | null | 0 | true | 0 | false | 0 |
: : t : : : : : : : :
| 6929 | storagemodel | 2000 | create view sys. | 1 | true | 0 | false | 0 |
: : : : storagemodel as : : : : : :
: : : : select * from sy : : : : : :
: : : : s.storagemodel...> : : : : :
| 6939 | tablestoragemode | 2000 | -- A summary of | 1 | true | 0 | false | 0 |
: : l : : the table storag : : : : : :
: : : : e requirement is : : : : : :
: : : : is available ...> : : : : :
| 6954 | statistics | 2000 | null | 0 | true | 0 | false | 0 |
| 7092 | systemfunctions | 2000 | null | 0 | true | 0 | false | 0 |
+------+------------------+------+------------------+------+-------+------+-------+------+
40 tuples (49.3s) !8 fields truncated!
sql>
### Expected Results:
normal response times of 15 msecs
When configuring without disabling debug, so
configure --prefix=$HOME/INSTALL --enable-assert
the problem is gone.
## Comment 20476
Date: 2014-11-28 13:19:50 +0100
From: @drstmane
While it might be interesting to analyze in detail where the time is spent, assertions might (and may) in general be arbitrarily complex and thus arbitrarily expensive. Thus, I do not really consider this a *bug*, rather a *feature*. Released binary production versions of MonetDB are alway compile with assertions disabled. The recommended configuration for performance tests is --disable-debug --disable-assert --enable-optimize .
Having said that, the combination of debug mask 10 (-d10 / --debug=10), i.e., property checking, and --enable-assert does do (as requested by these to options ;-)) intensive property checking on any intermediate BAT, hence, this might already explain the observed and reported performance differences ...
NB.
the configure defaults for HG sources are
--disable-optimize --enable-debug --enable-assert
the configure defaults for source tarballs are
--disable-optimize --disable-debug --disable-assert
## Comment 20477
Date: 2014-11-28 13:30:05 +0100
From: @drstmane
I cannot reproduce this with neither Oct2014 nor default, neither with not without -d10 ...
Could you please trace your query execution to see where time goes?
## Comment 20478
Date: 2014-11-28 13:32:38 +0100
From: @mlkersten
I can not reproduce it using the default.
## Comment 20479
Date: 2014-11-28 13:36:04 +0100
From: Martin van Dinther <<martin.van.dinther>>
the problem also occurs when compiling without assertions enabled, so using
configure --prefix=$HOME/INSTALL
and start server with -d10 and without embedded_r, so
INSTALL/bin/mserver5
The problem does not exist with the Oct2014 release
The problem did not exist 10 days ago with the default branch
So something is broken.
What do all these
delay initial 2[5241] memory 16894686848[13190987776.000000]
delay initial 3[5255] memory 16894685856[13190987776.000000]
delay initial 4[8323] memory 16894650528[13190987776.000000]
delay initial 4[12413] memory 16894640808[13190987776.000000]
delay initial 4[16532] memory 16894640856[13190987776.000000]
delay initial 4[20601] memory 16894640904[13190987776.000000]
delay initial 4[24671] memory 16894640960[13190987776.000000]
delay initial 4[28736] memory 16894641008[13190987776.000000]
mean???
## Comment 20480
Date: 2014-11-28 14:11:46 +0100
From: @mlkersten
The debugging output indicates that your command line triggered the thread debug flag PARdebug being set.
## Comment 20481
Date: 2014-11-28 14:30:18 +0100
From: Martin van Dinther <<martin.van.dinther>>
It appears mserver5 is consuming all inetrnal memory and the system is starting to do swapping:
KiB Mem: 16102280 total, 15926188 used, 176092 free, 2152 buffers
KiB Swap: 8388604 total, 8162496 used, 226108 free, 359820 cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
7012 dinther 20 0 22.103g 0.014t 0 S 9.3 91.1 1:01.28 mserver5
1272 dinther 20 0 695676 15364 5988 S 1.0 0.1 0:48.81 gnome-terminal-
6292 root 20 0 254912 13652 3808 S 0.7 0.1 3:11.64 Xorg
7 root 20 0 0 0 0 S 0.3 0.0 5:21.19 rcu_sched
1133 dinther 20 0 1373396 56008 948 S 0.3 0.3 3:13.28 firefox
7193 dinther 20 0 2809940 148556 0 S 0.3 0.9 0:06.43 java
1 root 20 0 50768 2764 1032 S 0.0 0.0 0:16.82 systemd
2 root 20 0 0 0 0 S 0.0 0.0 0:00.48 kthreadd
I can show it on my computer.
## Comment 20482
Date: 2014-11-28 14:32:26 +0100
From: Martin van Dinther <<martin.van.dinther>>
The problem can be reproduced better by using SQuirreL instead of mclient.
## Comment 20483
Date: 2014-11-28 15:24:39 +0100
From: Martin van Dinther <<martin.van.dinther>>
Please use SQuirreL and JDBC driver to reproduce the problem.
After you have connected to the server and run the query the second time you will see the issue.
## Comment 20484
Date: 2014-11-28 16:28:45 +0100
From: @drstmane
what does `hg sum` say in your case?
## Comment 20485
Date: 2014-11-28 16:33:14 +0100
From: @drstmane
NB: the code that issues the "delay initial" messages is identical between Oct2014 & default, and the statement that issues the very message(s) reported here (line 217 in monetdb5/mal/mal_resource.c) is not guarded by conditional PARDEBUG; cf.,
$ hg locate | xargs grep -n --color -9 'delay initial'
monetdb5/mal/mal_resource.c-197- rss = GDKmem_cursize();
monetdb5/mal/mal_resource.c-198- /* ample of memory available*/
monetdb5/mal/mal_resource.c-199- if ( rss < MEMORY_THRESHOLD * monet_memory && usec <= TIMESLICE)
monetdb5/mal/mal_resource.c-200- return;
monetdb5/mal/mal_resource.c-201-
monetdb5/mal/mal_resource.c-202- /* worker reporting time spent in usec! */
monetdb5/mal/mal_resource.c-203- clk = usec / 1000;
monetdb5/mal/mal_resource.c-204-
monetdb5/mal/mal_resource.c-205- if ( clk > DELAYUNIT ) {
monetdb5/mal/mal_resource.c:206: PARDEBUG mnstr_printf(GDKstdout, "delay initial "LLFMT"n", clk);
monetdb5/mal/mal_resource.c-207- (void) ATOMIC_DEC(running, runningLock, "MALresourceFairness");
monetdb5/mal/mal_resource.c-208- /* always keep one running to avoid all waiting */
monetdb5/mal/mal_resource.c-209- while (clk > 0 && running >= 2 && delayed < MAX_DELAYS) {
monetdb5/mal/mal_resource.c-210- /* speed up wake up when we have memory */
monetdb5/mal/mal_resource.c-211- if (rss < MEMORY_THRESHOLD * monet_memory)
monetdb5/mal/mal_resource.c-212- break;
monetdb5/mal/mal_resource.c-213- threads = GDKnr_threads > 0 ? GDKnr_threads : 1;
monetdb5/mal/mal_resource.c-214- delay = (unsigned int) ( ((double)DELAYUNIT * running) / threads);
monetdb5/mal/mal_resource.c-215- if (delay) {
monetdb5/mal/mal_resource.c-216- if ( delayed++ == 0){
monetdb5/mal/mal_resource.c:217: mnstr_printf(GDKstdout, "delay initial %u["LLFMT"] memory "SZFMT"[%f]\n", delay, clk, rss, MEMORY_THRESHOLD * monet_memory);
monetdb5/mal/mal_resource.c-218- mnstr_flush(GDKstdout);
monetdb5/mal/mal_resource.c-219- }
monetdb5/mal/mal_resource.c-220- MT_sleep_ms(delay);
monetdb5/mal/mal_resource.c-221- rss = GDKmem_cursize();
monetdb5/mal/mal_resource.c-222- } else break;
monetdb5/mal/mal_resource.c-223- clk -= DELAYUNIT;
monetdb5/mal/mal_resource.c-224- }
monetdb5/mal/mal_resource.c-225- (void) ATOMIC_INC(running, runningLock, "MALresourceFairness");
monetdb5/mal/mal_resource.c-226- }
## Comment 20486
Date: 2014-11-28 18:08:44 +0100
From: Martin van Dinther <<martin.van.dinther>>
bash-4.2$ hg sum
parent: 53982:b58221ff559c
Approved changed stable output
branch: default
commit: 19 unknown (clean)
update: (current)
bash-4.2$
## Comment 20488
Date: 2014-11-28 18:54:01 +0100
From: Martin van Dinther <<martin.van.dinther>>
I have done a run where I started mserver5 with --trace, so using:
./INSTALL/bin/mserver5 --trace
The trace output file is too large to attach (3MB), but you can read it from /ufs/dinther/trace_2014_11_28.out
I used SQuirreL 3.6 (http://www.squirrelsql.org/installation) to issue the following SQL queries:
1) SELECT id, name, schema_id, query, type, system, commit_action, readonly, temporary FROM tables WHERE name like '_tables';
-- this should return 2 rows.
2) select user
-- this should return 1 row with value: monetdb
mclient will not reproduce this behavior.
I also tested it via a dedicated Java test program, but could it also not reproduce.
So please download SQuirreL 3.6 (or copy it from /ufs/dinther/Downloads/squirrel-sql-3.6-standard.jar), install via:
java -jar squirrel-sql-3.6-standard.jar
Next start SQuirreL 3.6 via:
~/squirrel-sql-3.6/./squirrel-sql.sh
setup a Driver and create a Connection Alias, see https://www.monetdb.org/Documentation/Cookbooks/SQLrecipies/Clients/SQuirreL for instructions
## Comment 20501
Date: 2014-12-01 11:02:44 +0100
From: @drstmane
Hi,
In fact, I meant to trace by prefixing your query with keyword TRACE --- but may be that's not possible via SQuirreL (although MonetDB correctly returns two valid result sets in that case ...)?
I have no idea what SQuirreL does, but when I start mserver5 with --trace and run your two queries via mclient, the server produces a trace of a mere 42 kB (not 3 MB) --- with the original "select * from tables;" the trace is a mere 36 kB.
Anyway, to get any grip on this, we'd need a trace for only one query, or better, two traces for the same query, once "fast" when MonetDB is configured with --enable-assert and once "slow" when MonetDB is configured with --disable-assert. Then we can compare the two ...
Best,
Stefan
## Comment 20516
Date: 2014-12-05 13:16:49 +0100
From: Martin van Dinther <<martin.van.dinther>>
A reboot of my computer seems to have resolved the issue (luckily).
| huge performance issue when sw is compiled with --enable-assert --disable-debug | https://api.github.com/repos/MonetDB/MonetDB/issues/3631/comments | 0 | 2020-11-30T12:57:52Z | 2024-06-28T13:15:30Z | https://github.com/MonetDB/MonetDB/issues/3631 | 753,457,258 | 3,631 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-28 01:53:18 +0100
From: @skinkie
To: SQL devs <<bugs-sql>>
Version: 11.19.3 (Oct2014)
CC: @njnes
Last updated: 2015-01-29 14:07:19 +0100
## Comment 20463
Date: 2014-11-28 01:53:18 +0100
From: @skinkie
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/40.0.2214.6 Safari/537.36
Build Identifier:
CREATE TABLE "sys"."sirivm" (
"vehiclenumber" INTEGER,
"recordedattime" TIMESTAMP,
"linkdistance" INTEGER,
"percentage" DECIMAL(5,2),
"lineref" CHARACTER LARGE OBJECT,
"dataframeref" CHARACTER LARGE OBJECT,
"datedvehiclejourneyref" CHARACTER LARGE OBJECT,
"monitored" BOOLEAN,
"mon_error" CHARACTER LARGE OBJECT,
"blockref" CHARACTER LARGE OBJECT,
"delay" INTEGER,
"lon" DOUBLE,
"lat" DOUBLE,
"laststopref" CHARACTER LARGE OBJECT,
"laststoporder" INTEGER,
"lastaimedarr" TIMESTAMP,
"lastaimeddep" TIMESTAMP,
"monstopref" CHARACTER LARGE OBJECT,
"monstoporder" INTEGER,
"monaimedarr" TIMESTAMP,
"monaimeddep" TIMESTAMP,
"monatstop" CHARACTER LARGE OBJECT,
"monexpectedarr" TIMESTAMP,
"monexpecteddep" TIMESTAMP,
"nextstopref" CHARACTER LARGE OBJECT,
"nextstoporder" INTEGER,
"nextaimedarr" TIMESTAMP,
"nextaimeddep" TIMESTAMP,
"nextexpectedarr" TIMESTAMP,
"nextexpecteddep" TIMESTAMP,
"journeypattern" CHARACTER LARGE OBJECT
);
select a.recordedattime, a.monexpectedarr, a.monexpecteddep, a.monaimedarr, b.monaimeddep, a.delay, b.delay, (b.recordedattime - a.monaimedarr)/1000 as ourdelay, b.recordedattime, b.monexpectedarr, b.monexpecteddep, monstopref from sirivm as a join sirivm as b using (vehiclenumber, lineref, datedvehiclejourneyref, monstopref) where a.monatstop = 'false' and b.monatstop = 'true' order by abs(ourdelay) desc limit 10;
0x00007ffff2c96cf3 in rel_order_by_column_exp (sql=0x7fffe4005430, R=0x7ffff0215a10, column_r=0x7fffe4108c60, f=4) at rel_select.c:4509
4509 if (is_processed(r) && e) {
(gdb) bt
0 0x00007ffff2c96cf3 in rel_order_by_column_exp (sql=0x7fffe4005430, R=0x7ffff0215a10, column_r=0x7fffe4108c60, f=4) at rel_select.c:4509
1 0x00007ffff2c972cc in rel_order_by (sql=0x7fffe4005430, R=0x7ffff0215ab0, orderby=0x7fffe4108d40, f=4) at rel_select.c:4604
2 0x00007ffff2c999cb in rel_select_exp (sql=0x7fffe4005430, rel=0x7fffe413c6f0, sn=0x7fffe4108a80, ek=...) at rel_select.c:5208
3 0x00007ffff2c9a1c2 in rel_query (sql=0x7fffe4005430, rel=0x0, sq=0x7fffe4108a80, toplevel=1, ek=..., apply=8) at rel_select.c:5335
4 0x00007ffff2c9bb3d in rel_subquery (sql=0x7fffe4005430, rel=0x0, sq=0x7fffe4108a80, ek=..., apply=8) at rel_select.c:5714
5 0x00007ffff2c9bc15 in rel_selects (sql=0x7fffe4005430, s=0x7fffe4108a80) at rel_select.c:5731
6 0x00007ffff2c83eb9 in rel_semantic (sql=0x7fffe4005430, s=0x7fffe4108a80) at rel_semantic.c:264
7 0x00007ffff2bc80c8 in sql_symbol2relation (c=0x7fffe4005430, sym=0x7fffe4108a80) at sql.c:125
8 0x00007ffff2bf0f42 in SQLparser (c=0x7097c0) at sql_scenario.c:2308
9 0x00007ffff78f60d6 in runPhase (c=0x7097c0, phase=1) at mal_scenario.c:526
10 0x00007ffff78f6278 in runScenarioBody (c=0x7097c0) at mal_scenario.c:561
11 0x00007ffff78f6432 in runScenario (c=0x7097c0) at mal_scenario.c:590
12 0x00007ffff78f7dd8 in MSserveClient (dummy=0x7097c0) at mal_session.c:459
13 0x00007ffff7384920 in thread_starter (arg=0x7fffec000a20) at gdk_system.c:505
14 0x00007ffff4de9f3a in start_thread () from /lib64/libpthread.so.0
15 0x00007ffff4b23dad in clone () from /lib64/libc.so.6
Removing the abs from the query solves the issue.
Reproducible: Always
MonetDB 5 server v11.19.4 (64-bit, 64-bit oids)
This is an unreleased version
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2014 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 31.4GiB available memory, 8 available cpu cores
Libraries:
libpcre: 8.33 2013-05-28 (compiled with 8.33)
openssl: OpenSSL 1.0.1h 5 Jun 2014 (compiled with OpenSSL 1.0.1h 5 Jun 2014)
libxml2: 2.9.1 (compiled with 2.9.1)
Compiled by: skinkie@bigdata.openebs.nl (x86_64-unknown-linux-gnu)
Compilation: gcc -g -Werror -Wall -Wextra -W -Werror-implicit-function-declaration -Wpointer-arith -Wdeclaration-after-statement -Wundef -Wformat=2 -Wno-format-nonliteral -Winit-self -Winvalid-pch -Wmissing-declarations -Wmissing-format-attribute -Wmissing-prototypes -Wold-style-definition -Wpacked -Wunknown-pragmas -Wvariadic-macros -fstack-protector-all -Wstack-protector -Wpacked-bitfield-compat -Wsync-nand -Wjump-misses-init -Wmissing-include-dirs -Wlogical-op -Wunreachable-code
Linking : /usr/x86_64-pc-linux-gnu/bin/ld -m elf_x86_64
## Comment 20490
Date: 2014-11-30 15:35:59 +0100
From: @njnes
improved error message (and fixed crash)
## Comment 20491
Date: 2014-11-30 15:36:22 +0100
From: @njnes
test added order_by_incorrect_error.Bug-3630.sql
## Comment 20492
Date: 2014-11-30 15:52:54 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [44aa50bb3051](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=44aa50bb3051) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=44aa50bb3051](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=44aa50bb3051)
Changeset description:
fixed bug #3630 and added a test
## Comment 20596
Date: 2015-01-29 14:07:19 +0100
From: @sjoerdmullender
Oct2014-SP2 has been released.
| segv on rel_order_by_column_exp | https://api.github.com/repos/MonetDB/MonetDB/issues/3630/comments | 0 | 2020-11-30T12:57:49Z | 2024-06-27T12:05:43Z | https://github.com/MonetDB/MonetDB/issues/3630 | 753,457,231 | 3,630 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-27 19:58:26 +0100
From: Duncan Garmonsway <<nacnudus>>
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
CC: @njnes
Last updated: 2015-01-29 14:07:09 +0100
## Comment 20462
Date: 2014-11-27 19:58:26 +0100
From: Duncan Garmonsway <<nacnudus>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/39.0.2171.42 Safari/537.36
Build Identifier:
Reproducible: Always
### Steps to Reproduce:
1. monetdbd start monetdb
nacnudus@nactop:~% monetdb create test
created database in maintenance mode: test
nacnudus@nactop:~% monetdb release test
taken database out of maintenance mode: test
nacnudus@nactop:~% mclient test
user(nacnudus):monetdb
password:
Welcome to mclient, the MonetDB/SQL interactive terminal (Oct2014)
Database: MonetDB v11.19.3 (Oct2014), 'mapi:monetdb://nactop:50000/test'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>CREATE FUNCTION test(b1 BOOLEAN)
more>RETURNS INTEGER
more>BEGIN
more> IF b1 THEN RETURN 1;
more> ELSEIF NOT(b1) THEN RETURN 0;
more> ELSE RETURN NULL;
more> END IF;
more>END;
operation successful (31.235ms)
sql>SELECT test(true), test(false), test(null);
### Actual Results:
+------+------+-------------------+
| L1 | L2 | test_single_value |
+======+======+===================+
| 1 | 1 | 1 |
+------+------+-------------------+
1 tuple (1.470ms)
### Expected Results:
+------+------+-------------------+
| L1 | L2 | test_single_value |
+======+======+===================+
| 1 | 0 | null |
+------+------+-------------------+
1 tuple (1.470ms)
## Comment 20493
Date: 2014-11-30 18:33:48 +0100
From: @njnes
properly generate the else parts and implement 3 value logic in else cases
Added test ifthenelse.Bug-3629
## Comment 20494
Date: 2014-11-30 18:34:54 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [091b184e5199](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=091b184e5199) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=091b184e5199](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=091b184e5199)
Changeset description:
added test for bug #3629
## Comment 20495
Date: 2014-11-30 18:36:15 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [a7b88bea38e6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=a7b88bea38e6) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=a7b88bea38e6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=a7b88bea38e6)
Changeset description:
fixes for bug #3629
properly handle list of else parts
added 3 value logic in if then etc
## Comment 20593
Date: 2015-01-29 14:07:09 +0100
From: @sjoerdmullender
Oct2014-SP2 has been released.
| IF THEN ELSEIF always evaluates the first test as true | https://api.github.com/repos/MonetDB/MonetDB/issues/3629/comments | 0 | 2020-11-30T12:57:47Z | 2024-06-27T12:05:42Z | https://github.com/MonetDB/MonetDB/issues/3629 | 753,457,199 | 3,629 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-27 19:34:15 +0100
From: Jerry Evans <<g4>>
To: SQL devs <<bugs-sql>>
Version: 11.19.7 (Oct2014-SP1)
Last updated: 2015-01-29 14:07:41 +0100
## Comment 20461
Date: 2014-11-27 19:34:15 +0100
From: Jerry Evans <<g4>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/39.0.2171.65 Safari/537.36
Build Identifier:
Script that demonstrates problem
-----------------------------snip-----------------------
Create a tiny table to demonstrate
CREATE TABLE "tm_bug" (
"uid" integer NOT NULL,
"weight" double NOT NULL default 1.0,
"filter" integer NOT NULL default -1
);
insert values.
INSERT INTO "tm_bug" VALUES (1,1.0,1);
INSERT INTO "tm_bug" VALUES (2,1.0,1);
INSERT INTO "tm_bug" VALUES (3,1.0,2);
these queries return expectd data
select count(uid) as cnt1,stddev_pop(weight) as f1 from tm_bug where filter = 1;
select count(uid) as cnt2,stddev_pop(weight) as f2 from tm_bug where filter = 2;
this will return type mismatch
select count(uid) as cnt3,stddev_pop(weight) as f3 from tm_bug where filter = 3;
-----------------------------snip-----------------------
Reproducible: Always
### Steps to Reproduce:
The script above contains the repeatable error but the select that fails is here:
select count(uid) as cnt3,stddev_pop(weight) as f3 from tm_bug where filter = 3;
### Actual Results:
Type mismatch
### Expected Results:
One of 2 possibilities:
returned 2 columns, with cnt3 = 0 and f3 either null or 0.
or raised a divide by zero error as this is most likely the root cause of the problem given that stddev_pop would in this case have a 0 valued denominator
it would be good to have this behaviour configurable. As things stand it is a problem when used in more complex multi-term queries like this (which might well be made on the grounds of efficiency)
select count(uid), sum(weight), avg(weight), stddev_pop(weight) from tm_bug where filter = 3;
returns 'type mismatch'
select count(uid), sum(weight), avg(weight) from tm_bug where filter = 3;
returns 0,null,null
which somehow makes more sense intuitively ...?
## Comment 20464
Date: 2014-11-28 09:59:36 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [c11095d21be5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=c11095d21be5) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=c11095d21be5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=c11095d21be5)
Changeset description:
Distinguish type mismatch from empty bat in stddev/variance.
Also add test.
This fixes bug #3628.
## Comment 20465
Date: 2014-11-28 10:02:23 +0100
From: @sjoerdmullender
Not a division by 0 but just a lack of distinguising bad input (wrong type for calculating standard deviation) from no input.
I think the result for an "empty" standard deviation and variance should be null, so that's what it is. But if the SQL standard says differently, we can change it.
## Comment 20474
Date: 2014-11-28 12:49:00 +0100
From: Jerry Evans <<g4>>
Thanks for the attention to this. I agree. It makes much more sense to return null - this also greatly simplifies client code.
## Comment 20608
Date: 2015-01-29 14:07:41 +0100
From: @sjoerdmullender
Oct2014-SP2 has been released.
| mclient and ODBC driver report 'type mismatch' when stddev_pop used in a select which returns 0 rows | https://api.github.com/repos/MonetDB/MonetDB/issues/3628/comments | 0 | 2020-11-30T12:57:44Z | 2024-06-27T12:05:42Z | https://github.com/MonetDB/MonetDB/issues/3628 | 753,457,174 | 3,628 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-24 11:45:46 +0100
From: @bartscheers
To: SQL devs <<bugs-sql>>
Version: 11.19.3 (Oct2014)
CC: @njnes
Last updated: 2015-08-28 13:41:44 +0200
## Comment 20455
Date: 2014-11-24 11:45:46 +0100
From: @bartscheers
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:33.0) Gecko/20100101 Firefox/33.0
Build Identifier:
SQRT in CASE statement is executed when it shouldn't.
The case statement precalculates the values, and when negative values are encountered the sqrt gives a Math exception: Numerical argument out of domain message.
A workaround exists (for this example, but maybe not for all), by pulling the SQRT out of the CASE statement.
It gives the same error for the CASE statement in the SELECT projection and WHERE clause
Reproducible: Always
### Steps to Reproduce:
1. run run.sh
2.
3.
### Actual Results:
Math exception: Numerical argument out of domain
### Expected Results:
Result set
## Comment 20456
Date: 2014-11-24 11:49:08 +0100
From: @bartscheers
Created attachment 311
zip file of 3 files: run.sh, t1.sql and queries.sql
Use the run.sh to run it all
> Attached file: [case_sqrt.zip](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3627_case_sqrt.zip_311) (application/zip, 1024 bytes)
> Description: zip file of 3 files: run.sh, t1.sql and queries.sql
## Comment 20457
Date: 2014-11-24 15:41:06 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [682d22ac9255](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=682d22ac9255) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=682d22ac9255](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=682d22ac9255)
Changeset description:
Added test for bug #3627.
## Comment 20706
Date: 2015-03-11 17:01:06 +0100
From: @njnes
fixed by the relational optimizer
## Comment 21184
Date: 2015-08-28 13:41:44 +0200
From: @sjoerdmullender
Jul2015 has been released.
| SQRT in CASE does not work as of Oct2014 | https://api.github.com/repos/MonetDB/MonetDB/issues/3627/comments | 0 | 2020-11-30T12:57:42Z | 2024-06-27T12:05:40Z | https://github.com/MonetDB/MonetDB/issues/3627 | 753,457,137 | 3,627 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-23 13:34:51 +0100
From: @skinkie
To: SQL devs <<bugs-sql>>
Version: 11.19.3 (Oct2014)
CC: @njnes
Last updated: 2015-01-29 14:07:36 +0100
## Comment 20454
Date: 2014-11-23 13:34:51 +0100
From: @skinkie
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/40.0.2214.6 Safari/537.36
Build Identifier:
Casting a column to a different type without aliasing it to the same name results in program contains errors;
create table test2 (i int, b bool);
insert into test2 values (1, true);
select cast(b as smallint), (i > 0 and b) from test2;
TypeException:user.s33_1[22]:'calc.and' undefined in: _69:any := calc.and(_65:bit, _67:sht)
select cast(b as smallint) as b, (i > 0 and b) from test2;
+------+---------+
| b | and_>_i |
+======+=========+
| 1 | true |
+------+---------+
1 tuple (1.751ms)
my original problem was, but it is difficult to reproduce;
TypeException:user.s32_1[928]:'bat.insert' undefined in: _2267:any := bat.insert(_2257:bat[:oid,:bit], _2265:oid, _2264:sht)
Reproducible: Always
MonetDB 5 server v11.19.4 (64-bit, 64-bit oids)
This is an unreleased version
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2014 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 31.4GiB available memory, 8 available cpu cores
Libraries:
libpcre: 8.33 2013-05-28 (compiled with 8.33)
openssl: OpenSSL 1.0.1h 5 Jun 2014 (compiled with OpenSSL 1.0.1h 5 Jun 2014)
libxml2: 2.9.1 (compiled with 2.9.1)
Compiled by: skinkie@bigdata.openebs.nl (x86_64-unknown-linux-gnu)
Compilation: gcc -g -Werror -Wall -Wextra -W -Werror-implicit-function-declaration -Wpointer-arith -Wdeclaration-after-statement -Wundef -Wformat=2 -Wno-format-nonliteral -Winit-self -Winvalid-pch -Wmissing-declarations -Wmissing-format-attribute -Wmissing-prototypes -Wold-style-definition -Wpacked -Wunknown-pragmas -Wvariadic-macros -fstack-protector-all -Wstack-protector -Wpacked-bitfield-compat -Wsync-nand -Wjump-misses-init -Wmissing-include-dirs -Wlogical-op -Wunreachable-code
Linking : /usr/x86_64-pc-linux-gnu/bin/ld -m elf_x86_64
## Comment 20496
Date: 2014-12-01 09:17:35 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [d54e3de83771](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d54e3de83771) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=d54e3de83771](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=d54e3de83771)
Changeset description:
fixed bug #3626 by adding an extra alias to cast expressions
## Comment 20497
Date: 2014-12-01 09:20:21 +0100
From: @njnes
fixed by adding an extra label/alias
## Comment 20603
Date: 2015-01-29 14:07:36 +0100
From: @sjoerdmullender
Oct2014-SP2 has been released.
| casting a type without alias results in program contains errors | https://api.github.com/repos/MonetDB/MonetDB/issues/3626/comments | 0 | 2020-11-30T12:57:39Z | 2024-06-27T12:05:39Z | https://github.com/MonetDB/MonetDB/issues/3626 | 753,457,101 | 3,626 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-21 19:15:31 +0100
From: Richard Hughes <<richard.monetdb>>
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.19.3 (Oct2014)
CC: @njnes
Last updated: 2015-08-28 13:42:18 +0200
## Comment 20453
Date: 2014-11-21 19:15:31 +0100
From: Richard Hughes <<richard.monetdb>>
Created attachment 310
proposed fix
I reproducibly crashed mserver5 using the statement "create table foo as select distinct * from bar with data;" where bar has 80 columns. The exact location of the crashes varied considerably, but they were always within OPTmergetableImplementation() or one of its children.
The problem was that the 'mat' array was being overrun and was randomly overwriting (usually) malloced InstrPtr instances. mat is initialized on line 1453 to mb->vtop elements, however there are numerous newTmpVariable() calls throughout that optimizer which can grow vtop considerably after mat has been created, and hence apparently grow the storage requirement for mat correspondingly.
It's possible that I did this to myself because the patch from bug #3548 can produce more complicated programs than would otherwise be the case, but I see no reason why a sufficiently arcane program couldn't be created even without that patch, and the statement in question only reads from one table so the multi-mitosis code shouldn't even be doing anything.
The attached patch simply mechanically changes mat_t to matlist_t {mat_t *v;int top;int size;}, i.e. a standard resizeable array implementation. This makes my test case work. You may be able to think of a smarter way to do it (or may completely disagree with my analysis).
> Attached file: [opt-mergetable-mat-overrun.patch](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3625_opt-mergetable-mat-overrun.patch_310) (text/plain, 30112 bytes)
> Description: proposed fix
## Comment 20993
Date: 2015-07-11 11:45:08 +0200
From: @njnes
tooks some time, but applied patch
## Comment 21209
Date: 2015-08-28 13:42:18 +0200
From: @sjoerdmullender
Jul2015 has been released.
| SIGSEGV because mat array can overrun in opt_mergetable.c | https://api.github.com/repos/MonetDB/MonetDB/issues/3625/comments | 0 | 2020-11-30T12:57:36Z | 2024-06-27T12:05:38Z | https://github.com/MonetDB/MonetDB/issues/3625 | 753,457,062 | 3,625 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-21 17:01:03 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.17.21 (Jan2014-SP3)
CC: @njnes
Last updated: 2015-01-29 14:07:38 +0100
## Comment 20452
Date: 2014-11-21 17:01:03 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:33.0) Gecko/20100101 Firefox/33.0
Build Identifier:
insert of incomplete or invalid ip address values in inet column is silently accepted but the values are not stored (they become/show nil)
Reproducible: Always
### Steps to Reproduce:
1. start mserver5 (MonetDB 5 server v11.19.3 "Oct2014")
2. start mclient
3. run the SQL queries:
CREATE TABLE INET_TBL (ip inet);
INSERT INTO INET_TBL VALUES ('192.168.1');
SELECT * FROM INET_TBL;
INSERT INTO INET_TBL VALUES ('192.168.2.');
SELECT * FROM INET_TBL;
INSERT INTO INET_TBL VALUES ('193');
INSERT INTO INET_TBL VALUES ('193.169');
INSERT INTO INET_TBL VALUES ('193.169.13');
INSERT INTO INET_TBL VALUES ('193.169.13.4');
INSERT INTO INET_TBL VALUES ('193.169.13.4/80');
SELECT * FROM INET_TBL;
INSERT INTO INET_TBL VALUES ('193.169.13.4/32');
SELECT * FROM INET_TBL;
### Actual Results:
bash-4.2$ mclient -p 41000
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.19.3 (Oct2014), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>CREATE TABLE INET_TBL (ip inet);
operation successful (38.052ms)
sql>SELECT * FROM INET_TBL;
+----+
| ip |
+====+
+----+
0 tuples (1.059ms)
sql>NSERT INTO INET_TBL VALUES ('192.168.1');
syntax error, unexpected IDENT in: "nsert"
sql>INSERT INTO INET_TBL VALUES ('192.168.1');
1 affected row (2.850ms)
sql>SELECT * FROM INET_TBL;
+-------------------------------------------------------------------------------------------------+
| ip |
+=================================================================================================+
| (nil) |
+-------------------------------------------------------------------------------------------------+
1 tuple (2.138ms)
sql>INSERT INTO INET_TBL VALUES ('192.168.2.');
1 affected row (13.271ms)
sql>SELECT * FROM INET_TBL;
+-------------------------------------------------------------------------------------------------+
| ip |
+=================================================================================================+
| (nil) |
| 192.168.2.0 |
+-------------------------------------------------------------------------------------------------+
2 tuples (0.802ms)
sql>INSERT INTO INET_TBL VALUES ('193');
1 affected row (3.815ms)
sql>INSERT INTO INET_TBL VALUES ('193.169');
1 affected row (2.872ms)
sql>INSERT INTO INET_TBL VALUES ('193.169.13');
1 affected row (2.020ms)
sql>INSERT INTO INET_TBL VALUES ('193.169.13.4');
1 affected row (3.601ms)
sql>INSERT INTO INET_TBL VALUES ('193.169.13.4/80');
1 affected row (2.466ms)
sql>SELECT * FROM INET_TBL;
+-------------------------------------------------------------------------------------------------+
| ip |
+=================================================================================================+
| (nil) |
| 192.168.2.0 |
| (nil) |
| (nil) |
| (nil) |
| 193.169.13.4 |
| (nil) |
+-------------------------------------------------------------------------------------------------+
7 tuples (0.662ms)
sql>INSERT INTO INET_TBL VALUES ('193.169.13.4/32');
1 affected row (11.396ms)
sql>SELECT * FROM INET_TBL;
+-------------------------------------------------------------------------------------------------+
| ip |
+=================================================================================================+
| (nil) |
| 192.168.2.0 |
| (nil) |
| (nil) |
| (nil) |
| 193.169.13.4 |
| (nil) |
| 193.169.13.4 |
+-------------------------------------------------------------------------------------------------+
8 tuples (0.568ms)
sql>
### Expected Results:
sql>INSERT INTO INET_TBL VALUES ('192.168.1');
Error: invalid inet address: 192.168.1
sql>SELECT * FROM INET_TBL;
0 tuple (2.138ms)
sql>INSERT INTO INET_TBL VALUES ('192.168.2.');
1 affected row (13.271ms)
sql>SELECT * FROM INET_TBL;
+-------------------------------------------------------------------------------------------------+
| ip |
+=================================================================================================+
192.168.2.0 |
+-------------------------------------------------------------------------------------------------+
1 tuples (0.802ms)
sql>INSERT INTO INET_TBL VALUES ('193');
Error: invalid inet address: 193
sql>INSERT INTO INET_TBL VALUES ('193.169');
Error: invalid inet address: 193.169
sql>INSERT INTO INET_TBL VALUES ('193.169.13');
Error: invalid inet address: 193.169.13
sql>INSERT INTO INET_TBL VALUES ('193.169.13.4');
1 affected row (3.601ms)
sql>INSERT INTO INET_TBL VALUES ('193.169.13.4/80');
1 affected row (2.466ms)
sql>SELECT * FROM INET_TBL;
ip
192.168.2.0 193.169.13.4
193.169.13.4/80
3 tuples (0.662ms)
## Comment 20499
Date: 2014-12-01 10:01:30 +0100
From: @njnes
added parsing errors
## Comment 20504
Date: 2014-12-01 19:25:27 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [7ad1ee2a65a7](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=7ad1ee2a65a7) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=7ad1ee2a65a7](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=7ad1ee2a65a7)
Changeset description:
added test for inet parsing bug #3624
## Comment 20517
Date: 2014-12-05 13:21:21 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [0a110f07a888](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=0a110f07a888) made by Martin van Dinther <martin.van.dinther@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=0a110f07a888](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=0a110f07a888)
Changeset description:
Reapproved test output after fix for bug #3624
## Comment 20605
Date: 2015-01-29 14:07:38 +0100
From: @sjoerdmullender
Oct2014-SP2 has been released.
| insert of incomplete or invalid ip address values in inet column is silently accepted but the values are not stored (they become/show nil) | https://api.github.com/repos/MonetDB/MonetDB/issues/3624/comments | 0 | 2020-11-30T12:57:33Z | 2024-06-27T12:05:37Z | https://github.com/MonetDB/MonetDB/issues/3624 | 753,457,036 | 3,624 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-20 16:06:13 +0100
From: zephyr_dodolee
To: clients devs <<bugs-clients>>
Version: 11.19.3 (Oct2014)
Last updated: 2014-11-21 08:00:41 +0100
## Comment 20449
Date: 2014-11-20 16:06:13 +0100
From: zephyr_dodolee
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.10; rv:33.0) Gecko/20100101 Firefox/33.0
Build Identifier:
command \d[Stvsfn]+ gives no result, other \ command works as expected.
Reproducible: Always
### Steps to Reproduce:
1. mclient -u monetdb -d database
2. enter password
3. sql>\d
### Actual Results:
shows a new sql> prompt at the next line
### Expected Results:
TABLE ...
TABLE ...
...
sql>
the command works on 11.15.19
## Comment 20450
Date: 2014-11-20 16:29:41 +0100
From: @sjoerdmullender
\d only lists tables and views in the current schema that are not system tables/views. If you want to see all tables and views, use
\d *.*
## Comment 20451
Date: 2014-11-21 07:59:33 +0100
From: zephyr_dodolee
Thanks！
\d *.* works just as expected.
| command \d[Stvsfn]+ gives no result | https://api.github.com/repos/MonetDB/MonetDB/issues/3623/comments | 0 | 2020-11-30T12:57:31Z | 2024-06-28T13:15:29Z | https://github.com/MonetDB/MonetDB/issues/3623 | 753,457,010 | 3,623 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-20 07:27:30 +0100
From: @mlkersten
To: SQL devs <<bugs-sql>>
Version: 11.17.13 (Jan2014-SP1)
CC: @njnes
Last updated: 2015-04-22 22:37:57 +0200
## Comment 20447
Date: 2014-11-20 07:27:30 +0100
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.10; rv:33.0) Gecko/20100101 Firefox/33.0
Build Identifier:
An erroneous type downcast is triggered in the query below on the columns in the range join.
File added to the test suite.
A bypass is to properly cast the generate_series arguments to integer.
Reproducible: Always
### Steps to Reproduce:
CREATE FUNCTION seq_char(val integer, seq string, pos integer, cigar string)
RETURNS INTEGER
BEGIN
RETURN 1;
END;
CREATE TABLE "bam"."alignments_2" (
"virtual_offset" BIGINT NOT NULL,
"qname" CHARACTER LARGE OBJECT NOT NULL,
"flag" SMALLINT NOT NULL,
"rname" CHARACTER LARGE OBJECT NOT NULL,
"pos" INTEGER NOT NULL,
"epos" INTEGER NOT NULL,
"mapq" SMALLINT NOT NULL,
"cigar" CHARACTER LARGE OBJECT NOT NULL,
"rnext" CHARACTER LARGE OBJECT NOT NULL,
"pnext" INTEGER NOT NULL,
"tlen" INTEGER NOT NULL,
"seq" CHARACTER LARGE OBJECT NOT NULL,
"qual" CHARACTER LARGE OBJECT NOT NULL,
CONSTRAINT "alignments_2_pkey_virtual_offset" PRIMARY KEY ("virtual_offset")
);
ALTER TABLE bam.alignments_2 SET READ ONLY;
EXPLAIN SELECT s.value AS refpos, COUNT(*) AS cnt
FROM
generate_series(128, 18960) AS s
JOIN (
SELECT epos, pos, seq, cigar
FROM bam.alignments_2
WHERE pos > 0
) AS al
ON (
s.value >= al.pos AND s.value <= al."epos"
AND seq_char(s.value, al.seq, al.pos, al.cigar) IS NOT NULL
)
GROUP BY refpos
ORDER BY cnt DESC
LIMIT 10;
DROP TABLE bam.alignments_2;
DROP FUNCTION seq_char;
## Comment 20448
Date: 2014-11-20 07:31:18 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [f80d7e960167](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=f80d7e960167) made by Martin Kersten <mk@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=f80d7e960167](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=f80d7e960167)
Changeset description:
Added a type-cast error bug #3622
## Comment 20509
Date: 2014-12-03 13:33:58 +0100
From: @njnes
fixed by the correction done for bug #3633
## Comment 20510
Date: 2014-12-03 13:41:21 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [1837da979e2f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=1837da979e2f) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=1837da979e2f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=1837da979e2f)
Changeset description:
added test for bug #3622
## Comment 20606
Date: 2015-01-29 14:07:39 +0100
From: @sjoerdmullender
Oct2014-SP2 has been released.
## Comment 20833
Date: 2015-04-22 22:37:57 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [5bb9e125f99c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=5bb9e125f99c) made by Stefan Manegold <Stefan.Manegold@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=5bb9e125f99c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=5bb9e125f99c)
Changeset description:
make test typecast_error.Bug-3622 work also without SAMTOOLS / BAM support,
i.e., without schema "bam" existing
| Type resolution error | https://api.github.com/repos/MonetDB/MonetDB/issues/3622/comments | 0 | 2020-11-30T12:57:28Z | 2024-06-27T12:05:35Z | https://github.com/MonetDB/MonetDB/issues/3622 | 753,456,980 | 3,622 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-11 12:16:58 +0100
From: Frédéric Jolliton <<frederic.jolliton+monetdb>>
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @drstmane
Last updated: 2014-11-28 10:03:20 +0100
## Comment 20436
Date: 2014-11-11 12:16:58 +0100
From: Frédéric Jolliton <<frederic.jolliton+monetdb>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:33.0) Gecko/20100101 Firefox/33.0
Build Identifier:
Not sure if this is a bug. Shouldn't the hexadecimal literal be treated the same as decimal one?
I was trying to insert hexadecimal constants for hugeint, but while no errors was reported, I discovered that it was silently truncated to 64 bits.
sql>select 18446744073709551616; -- 0x10000000000000000 in hex
+----------------------+
| single_value |
+======================+
| 18446744073709551616 |
+----------------------+
1 tuple (0.381ms)
sql>select 0x10000000000000000;
+--------------+
| single_value |
+==============+
| 0 |
+--------------+
1 tuple (0.362ms)
Reproducible: Always
## Comment 20437
Date: 2014-11-12 10:53:21 +0100
From: @sjoerdmullender
Are you sure you're using the Jan2014-SP3 release for this?
The value 18446744073709551616 does not fit in a 64-bit integer, so it is not accepted by the server:
sql>select 18446744073709551616;
integer value too large or not a number (18446744073709551616) in: "select 18446744073709551616"
This is also true for the Oct2014 release.
In the default branch, which you would have to compile for yourself, there is initial support for 128-bit integers, so there it works:
sql>select 18446744073709551616;
+----------------------+
| single_value |
+======================+
| 18446744073709551616 |
+----------------------+
1 tuple (0.632ms)
But it looks like there, the support for 128-bit hexadecimal values isn't there yet:
sql>select 0x10000000000000000;
+--------------+
| single_value |
+==============+
| 0 |
+--------------+
1 tuple (0.629ms)
## Comment 20438
Date: 2014-11-12 10:58:26 +0100
From: Frédéric Jolliton <<frederic.jolliton+monetdb>>
Seems like I chosen the wrong version when creating this entry. It is for the development branch actually. (We are using hugeint to store IPv6.)
## Comment 20440
Date: 2014-11-14 09:50:50 +0100
From: @drstmane
Hi,
thanks for reporting!
Indeed, parsing / converting >64 bit hexadecimal literals in SQL is broken in the default branch.
I have a fix pending, but need to discuss with Niels & Sjoerd before committing,
as I see more potential inconsistencies with parsing/converting hexadecimal literals in SQL also in the Oct2014 (and earlier) branch(es).
Stefan
## Comment 20441
Date: 2014-11-14 19:01:49 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [d9e7bfe93ede](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d9e7bfe93ede) made by Stefan Manegold <Stefan.Manegold@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=d9e7bfe93ede](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=d9e7bfe93ede)
Changeset description:
fixed handling of hexadeximal literals in SQL:
- type "long (int)" is EVIL --- we need to use type "lng" instead !
- determine smallest suitable type based on actualy value,
not on number of hexadecimal digits
(i.e., in particular ignore/skip leading zeros)
- use tinyint as smallest type (not only smallint)
- avoid treating valid hex values as NULL:
hex was is
0x8000 NULL (smallint) 32768 (int)
0x80000000 NULL (int) 2147483648 (bigint)
0x8000000000000000 NULL (bigint) ERROR (value too large)
- avoid negative values:
hex was is
0x8001 -32767 (smallint) 32769 (int)
0xffff -1 (smallint) 65535 (int)
0x80000001 -2147483647 (int) 2147483649 (bigint)
0xffffffff -1 (int) 4294967295 (bigint)
0x8000000000000001 -9223372036854775807 (bigint) ERROR (value too large)
0xffffffffffffffff -1 (bigint) ERROR (value too large)
(leading sign (-0x...) is handled separatelyu elsewhere)
(These fixed have been triggered by bug #3621.)
## Comment 20444
Date: 2014-11-17 17:35:02 +0100
From: @drstmane
This (and more) should be fixed in both the Oct2014 branch (up to 64-bit) and the default branch (also incl. 128-bit) as by above mentioned checking (and it's merge into the default branch).
The up to 64-bit fix will be released with the next Oct2014 bug-fix release; the 128-bit fix will be released with the next feature release.
## Comment 20471
Date: 2014-11-28 10:03:20 +0100
From: @sjoerdmullender
Oct2014-SP1 has been released.
| Hexadecimal literal vs decimal literal | https://api.github.com/repos/MonetDB/MonetDB/issues/3621/comments | 0 | 2020-11-30T12:57:25Z | 2024-06-27T12:05:34Z | https://github.com/MonetDB/MonetDB/issues/3621 | 753,456,952 | 3,621 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-10 15:33:54 +0100
From: Frédéric Jolliton <<frederic.jolliton+monetdb>>
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2014-11-28 10:03:16 +0100
## Comment 20435
Date: 2014-11-10 15:33:54 +0100
From: Frédéric Jolliton <<frederic.jolliton+monetdb>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:33.0) Gecko/20100101 Firefox/33.0
Build Identifier:
Sadly, we encountered another bug related to UNION ALL. When trying to sort the result of an UNION ALL, the ORDER BY clause doesn't seem to be applied correctly.
Reproducible: Always
### Steps to Reproduce:
SELECT val FROM (SELECT 1 AS val UNION ALL SELECT 2) AS foo ORDER BY -val ASC;
### Actual Results:
1
2
### Expected Results:
2
1
The explain for the query (the "-val" computation appears nowhere):
_4:bat[:oid,:bte] = bat.new(nil:oid, nil:bte)
_7:bat[:oid,:bte] = sql.single(A0:bte)
_8:bat[:oid,:bte] = bat.append(_4:bat[:oid,:bte], _7:bat[:oid,:bte], true:bit)
_9:bat[:oid,:bte] = bat.append(_8:bat[:oid,:bte], A1:bte, true:bit)
_10:bat[:oid,:bte], r1_13:bat[:oid,:oid], r2_13:bat[:oid,:oid] = algebra.subsort(_9:bat[:oid,:bte], false:bit, false:bit)
_14:bat[:oid,:bte] = algebra.leftfetchjoin(r1_13:bat[:oid,:oid], _9:bat[:oid,:bte])
_15:int = sql.resultSet(1:int, 1:int, _14:bat[:oid,:bte])
_17:void = sql.rsColumn(_15:int, ".foo":str, "val":str, "tinyint":str, 8:int, 0:int, _14:bat[:oid,:bte])
_23:streams = io.stdout()
_24:void = sql.exportResult(_23:streams, _15:int)
Wrapping the subselect makes the sort to work:
SELECT val FROM (SELECT * FROM (SELECT 1 AS val UNION ALL SELECT 2) AS foo) AS bar ORDER BY -val ASC;
It produces, as expected:
2
1
$ bin/mserver5 --version
MonetDB 5 server v11.20.0 (64-bit, 64-bit oids, 128-bit integers)
## Comment 20439
Date: 2014-11-12 11:00:05 +0100
From: Frédéric Jolliton <<frederic.jolliton+monetdb>>
Version changed to development. (Is that a bug in the tracker? Both my tickets were created with "development" as version, but got stored with Jan2014-SP3..)
## Comment 20442
Date: 2014-11-16 18:49:26 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [504294a644fd](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=504294a644fd) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=504294a644fd](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=504294a644fd)
Changeset description:
fixed bug in order with column expression ie bug #3620
## Comment 20443
Date: 2014-11-16 18:50:01 +0100
From: @njnes
Fixed properly handle the extra (order by) expression and extra projection
## Comment 20445
Date: 2014-11-18 10:30:34 +0100
From: Frédéric Jolliton <<frederic.jolliton+monetdb>>
Works great. Thanks for the quick fix!
While we found some bugs that could worry us sometimes, we are also impressed by how quickly you fix them. This is greatly appreciated and reassuring.
## Comment 20466
Date: 2014-11-28 10:03:16 +0100
From: @sjoerdmullender
Oct2014-SP1 has been released.
| ORDER BY broken when using UNION ALL | https://api.github.com/repos/MonetDB/MonetDB/issues/3620/comments | 0 | 2020-11-30T12:57:23Z | 2024-06-27T12:05:33Z | https://github.com/MonetDB/MonetDB/issues/3620 | 753,456,925 | 3,620 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-05 11:33:37 +0100
From: @hannesmuehleisen
To: SQL devs <<bugs-sql>>
Version: 11.19.3 (Oct2014)
Last updated: 2015-01-29 14:07:17 +0100
## Comment 20426
Date: 2014-11-05 11:33:37 +0100
From: @hannesmuehleisen
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/38.0.2125.111 Safari/537.36
Build Identifier:
When starting MonetDB on Windows 7 64 bits (64 bits MonetDB version), we get
"The program can't start because MSVCR80.dll is missing from your computer. Try reinstalling the program to fix this problem."
Reproducible: Always
### Actual Results:
No dice
### Expected Results:
MonetDB starting
## Comment 20427
Date: 2014-11-05 11:36:04 +0100
From: @hannesmuehleisen
Box also WAMP, XAMPP and MySQL installed
## Comment 20428
Date: 2014-11-05 11:39:17 +0100
From: @hannesmuehleisen
Also happens in the Oct2014 release. Also, the MonetDB Server batch file breaks if there happens to be a folder name that contains an ampersand (&) character in the %PATH% variable.
## Comment 20429
Date: 2014-11-06 11:29:53 +0100
From: @sjoerdmullender
I cannot reproduce this on a clean system. All updates applied, and firefox installed, but nothing else. Certainly no old versions of MonetDB installed.
Can you try with the Microsoft dependency walker (depends.exe from http://www.dependencywalker.com) which exe or dll in the installation requires the missing dll?
## Comment 20520
Date: 2014-12-09 09:36:21 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [e764764bccab](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e764764bccab) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=e764764bccab](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=e764764bccab)
Changeset description:
Put DLLs into bin directory together with EXE.
This should fix bug #3619.
## Comment 20521
Date: 2014-12-09 09:39:36 +0100
From: @sjoerdmullender
It turned out the failing system had an unrelated version of iconv.dll in its system folder. In addition, it turned out that Windows looks in the system folder before looking into the folders mentioned in the Path variable to locate a DLL. In other words, the wrong (i.e., not our own) iconv.dll was loaded.
Since it also turned out that Windows looks in the folder in which the EXE was found before it looks in the system folder, we should be able to force the use of our own DLLs by mnoving them into the same folder as the EXEs.
## Comment 20522
Date: 2014-12-09 10:39:48 +0100
From: @hannesmuehleisen
Thanks, Sjoerd!
## Comment 20525
Date: 2014-12-11 13:44:32 +0100
From: @sjoerdmullender
Now that Hannes has tested the new installer and reported that it works, we can close the bug report.
## Comment 20595
Date: 2015-01-29 14:07:17 +0100
From: @sjoerdmullender
Oct2014-SP2 has been released.
| Missing dll on MonetDB Start | https://api.github.com/repos/MonetDB/MonetDB/issues/3619/comments | 0 | 2020-11-30T12:57:20Z | 2024-06-27T12:05:32Z | https://github.com/MonetDB/MonetDB/issues/3619 | 753,456,899 | 3,619 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-05 04:52:46 +0100
From: Julien Marie <<jm>>
To: Merovingian devs <<bugs-merovingian>>
Version: 11.21.19 (Jul2015-SP4)
CC: jm
Last updated: 2016-04-11 11:24:42 +0200
## Comment 20424
Date: 2014-11-05 04:52:46 +0100
From: Julien Marie <<jm>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_8_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/38.0.2125.111 Safari/537.36
Build Identifier:
When trying to release the database, outputs release: no such database: mochio. The monetdbd outputs illegal instruction.
Reproducible: Always
### Steps to Reproduce:
1. create the farm
2. create the db
3. try to release the db
### Actual Results:
Didnt work. Crashed.
### Expected Results:
Releasing the db
Compiled from source on Power8 / Ubuntu 14.04 on RunAbove cloud system ( http://runabove.com/ )
| Illegal instruction thrown when releasing db | https://api.github.com/repos/MonetDB/MonetDB/issues/3618/comments | 0 | 2020-11-30T12:57:18Z | 2024-06-28T13:15:29Z | https://github.com/MonetDB/MonetDB/issues/3618 | 753,456,865 | 3,618 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-03 13:49:37 +0100
From: Christian Braun <<hcb>>
To: SQL devs <<bugs-sql>>
Version: 11.19.3 (Oct2014)
CC: @mlkersten
Last updated: 2014-11-03 14:40:06 +0100
## Comment 20420
Date: 2014-11-03 13:49:37 +0100
From: Christian Braun <<hcb>>
User-Agent: Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:33.0) Gecko/20100101 Firefox/33.0
Build Identifier:
$ mclient db
Welcome to mclient, the MonetDB/SQL interactive terminal (Oct2014)
Database: MonetDB v11.19.3 (Oct2014), 'mapi:monetdb://makalu:50000/db'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>select * from sys.queue;
SELECT: no such operator 'queue'
sql>select * from sys.queue();
SELECT: no such operator 'queue'
sql>
Reproducible: Always
## Comment 20421
Date: 2014-11-03 14:22:19 +0100
From: @mlkersten
The problem is not confirmed.
Did you use an old database? what was the previous MonetDB version deployed>
Oct2014> mclient -d Octdemo
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.20.0 (unreleased), 'mapi:monetdb://vienna.ins.cwi.nl:50000/Octdemo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>
sql>select * from sys.queue();
+------+---------+----------------------------+----------------------------+----------+---------+-----------+----------------------------+
| qtag | user | started | estimate | progress | status | tag | query |
+======+=========+============================+============================+==========+=========+===========+============================+
| 312 | monetdb | 2014-11-03 13:20:50.000000 | null | null | running | 1005145@0 | select * from sys.queue(); |
+------+---------+----------------------------+----------------------------+----------+---------+-----------+----------------------------+
1 tuple (1.201ms)
## Comment 20422
Date: 2014-11-03 14:40:00 +0100
From: Christian Braun <<hcb>>
Created attachment 309
merovingian log file
> Attached file: [merovingian.log](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3617_merovingian.log_309) (text/x-log, 29480 bytes)
> Description: merovingian log file
## Comment 20423
Date: 2014-11-03 14:40:06 +0100
From: Christian Braun <<hcb>>
I was running MonetDB 5 server v11.17.21 "Jan2014-SP3".
I have attached the merovingian log file
| system monitor (sys.queue) not working after update | https://api.github.com/repos/MonetDB/MonetDB/issues/3617/comments | 0 | 2020-11-30T12:57:15Z | 2024-06-28T13:15:28Z | https://github.com/MonetDB/MonetDB/issues/3617 | 753,456,837 | 3,617 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-01 19:49:52 +0100
From: @mlkersten
To: SQL devs <<bugs-sql>>
Version: 11.17.13 (Jan2014-SP1)
CC: @hannesmuehleisen
Last updated: 2014-11-28 10:03:18 +0100
## Comment 20418
Date: 2014-11-01 19:49:52 +0100
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:33.0) Gecko/20100101 Firefox/33.0
Build Identifier:
Coalesce is incompatible with postgresql.
Added a file to the BugTracker-2014
Reproducible: Always
## Comment 20419
Date: 2014-11-01 20:17:30 +0100
From: @mlkersten
Pg-regress/Tests/case.sql shows three errors
when the statements are pasted one by one into a mclient session.
Calling Mtest did not show it
1) coalesce query produces wrong results:
sql>SELECT COALESCE(a.f, b.i, b.j)
more> FROM CASE_TBL a, CASE2_TBL b;
+--------------------------+
| isnull_not_isnull_f |
+==========================+
| 10.1 |
| 10.1 |
| 10.1 |
| 10.1 |
| 10.1 |
| 10.1 |
2) corrupting the interface with messages:
sql>SELECT '' AS Five, NULLIF(a.i,b.i) AS "NULLIF(a.i,b.i)",
more> NULLIF(b.i, 4) AS "NULLIF(b.i,4)"
more> FROM CASE_TBL a, CASE2_TBL b;
...
invalid tuple received from server, got 3 columns, expected 5, ignoring
invalid tuple received from server, got 3 columns, expected 5, ignoring
invalid tuple received from server, got 3 columns, expected 5, ignoring
3) unexpected syntax error
>UPDATE CASE_TBL
more> SET i = CASE WHEN b.i >= 2 THEN (2 * j)
more> ELSE (3 * j) END
more> FROM CASE2_TBL b
more> WHERE j = -CASE_TBL.i;
syntax error, unexpected FROM, expecting SCOLON in: "update case_tbl
set i = case when b.i >= 2 then (2 * j)
else ("
## Comment 20431
Date: 2014-11-07 10:06:13 +0100
From: @hannesmuehleisen
> 2) corrupting the interface with messages:
> sql>SELECT '' AS Five, NULLIF(a.i,b.i) AS "NULLIF(a.i,b.i)",
> more> NULLIF(b.i, 4) AS "NULLIF(b.i,4)"
> more> FROM CASE_TBL a, CASE2_TBL b;
> ...
> invalid tuple received from server, got 3 columns, expected 5, ignoring
*cough*
This problem is due to the column names in the MAPI table response header being not quoted and separated with ,. Thus, if the column names contains a comma, this breaks parsing on the client side.
Example:
sql>select 1 as "a,b", 2 as c;
% .L, .L table_name
% a,b, c name
% tinyint, tinyint type
% 1, 1 length
[ 1, 2 ]
## Comment 20458
Date: 2014-11-24 17:11:05 +0100
From: @sjoerdmullender
The bug has nothing to do with coalesce.
## Comment 20459
Date: 2014-11-24 17:16:09 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [007692e0b975](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=007692e0b975) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=007692e0b975](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=007692e0b975)
Changeset description:
Fixed bug #3616 + added test.
## Comment 20460
Date: 2014-11-24 17:17:39 +0100
From: @sjoerdmullender
Fixed in changeset [007692e0b975](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=007692e0b975)
## Comment 20469
Date: 2014-11-28 10:03:18 +0100
From: @sjoerdmullender
Oct2014-SP1 has been released.
| , in column name screws up mclient | https://api.github.com/repos/MonetDB/MonetDB/issues/3616/comments | 0 | 2020-11-30T12:57:13Z | 2024-06-27T12:05:29Z | https://github.com/MonetDB/MonetDB/issues/3616 | 753,456,805 | 3,616 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-01 18:07:22 +0100
From: @mlkersten
To: SQL devs <<bugs-sql>>
Version: 11.17.13 (Jan2014-SP1)
CC: martin.van.dinther, @njnes
Duplicates: #3600
Last updated: 2014-11-07 16:59:28 +0100
## Comment 20417
Date: 2014-11-01 18:07:22 +0100
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:33.0) Gecko/20100101 Firefox/33.0
Build Identifier:
In the context of wisconsin database do the following:
sql>CREATE TEMPORARY TABLE random_tbl( random double);
operation successful (25.927ms)
sql>INSERT INTO random_tbl SELECT count(*) AS random
more> FROM onek WHERE rand() < 1.0/10;
1 affected row (16.465ms)
sql>select * from random_tbl;
+--------+
| random |
+========+
+--------+
0 tuples (0.960ms)
It misses either the type match error of count() and destination column, or ....
part of the pg_regress/Tests/random.sql test
Reproducible: Always
## Comment 20425
Date: 2014-11-05 10:52:16 +0100
From: @njnes
temporary without the proper commit action (preserve rows) will clean out the table on commit. Try without autocommit.
## Comment 20432
Date: 2014-11-07 16:47:44 +0100
From: Martin van Dinther <<martin.van.dinther>>
Duplicate of bug #3600
## Comment 20433
Date: 2014-11-07 16:59:28 +0100
From: @sjoerdmullender
*** This bug has been marked as a duplicate of bug #3600 ***
| Insert in temporary table fails | https://api.github.com/repos/MonetDB/MonetDB/issues/3615/comments | 0 | 2020-11-30T12:57:10Z | 2024-06-28T13:15:27Z | https://github.com/MonetDB/MonetDB/issues/3615 | 753,456,785 | 3,615 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-11-01 00:41:47 +0100
From: Dann Corbit <<dcorbit>>
To: clients devs <<bugs-clients>>
Version: 11.19.3 (Oct2014)
Last updated: 2015-08-28 23:40:53 +0200
## Comment 20411
Date: 2014-11-01 00:41:47 +0100
From: Dann Corbit <<dcorbit>>
User-Agent: Mozilla/5.0 (Windows NT 6.3; Win64; x64; rv:25.0) Gecko/20141021 Firefox/24.9 PaleMoon/25.0.2
Build Identifier:
The error does not occur 100% of the time.
Sometimes I get a partial result set and then the error.
Sometimes I get a complete result set of 8115 rows and no error.
Reproducible: Sometimes
### Steps to Reproduce:
1. Create a 64 bit ODBC data source using the voc sample database
2. Import the database into CONNX
3. Query with InfoNaut
Sometimes, it works, sometimes it does not. It may fail three times in a row or it may succeed three times in a row.
### Actual Results:
Partial result set and an error message on some queries
### Expected Results:
Full result set and no errors for all well formed queries.
I have some bitmap screen shots of the problem that could prove helpful.
All other ODBC drivers tested do not demonstrate this problem (dozens of them).
## Comment 20412
Date: 2014-11-01 00:43:19 +0100
From: Dann Corbit <<dcorbit>>
Created attachment 306
Incomplete result set when the error occurs
> Attached file: [incomplete-when-error-occurs.png](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3614_incomplete-when-error-occurs.png_306) (image/png, 31593 bytes)
> Description: Incomplete result set when the error occurs
## Comment 20413
Date: 2014-11-01 00:43:48 +0100
From: Dann Corbit <<dcorbit>>
Created attachment 307
Error popup seen when the error does occur
> Attached file: [error-pop-up.png](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3614_error-pop-up.png_307) (image/png, 19544 bytes)
> Description: Error popup seen when the error does occur
## Comment 20414
Date: 2014-11-01 00:44:12 +0100
From: Dann Corbit <<dcorbit>>
Created attachment 308
The correct result set when the error does not occur.
> Attached file: [worked.png](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3614_worked.png_308) (image/png, 39215 bytes)
> Description: The correct result set when the error does not occur.
## Comment 20415
Date: 2014-11-01 00:46:04 +0100
From: Dann Corbit <<dcorbit>>
the number of rows returned on error is random (might be one or some larger number like 12 rows).
## Comment 20416
Date: 2014-11-01 01:12:35 +0100
From: Dann Corbit <<dcorbit>>
VOC creation script obtained from here:
https://www.monetdb.org/Documentation/UserGuide/Tutorial/Windows
## Comment 21235
Date: 2015-08-28 23:02:11 +0200
From: Dann Corbit <<dcorbit>>
This bug is no longer reproducible. Some change has corrected the problem, so you can close it now.
The issue was still present in the version released in June, but this release in August 2015 corrects the problem.
This is a very important change because ODBC is now reliable.
## Comment 21236
Date: 2015-08-28 23:05:26 +0200
From: Dann Corbit <<dcorbit>>
I should like to mention that I get 37000 rows per second query speed on :
SELECT * FROM voyages
which is fantastic performance.
## Comment 21237
Date: 2015-08-28 23:40:53 +0200
From: @sjoerdmullender
Thanks for the feedback.
I don't know which change fixed this issue, but I'm glad it is fixed.
| SELECT using ODBC for demo table voyages gives error "Row handle referred to a deleted row or a row marked for deletion | https://api.github.com/repos/MonetDB/MonetDB/issues/3614/comments | 0 | 2020-11-30T12:57:07Z | 2024-06-27T12:05:26Z | https://github.com/MonetDB/MonetDB/issues/3614 | 753,456,751 | 3,614 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-10-31 18:22:43 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.19.3 (Oct2014)
CC: @njnes
Last updated: 2015-05-07 12:37:50 +0200
## Comment 20410
Date: 2014-10-31 18:22:43 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:33.0) Gecko/20100101 Firefox/33.0
Build Identifier:
The following keywords exist as column name in the sys.* (and t,p.*) tables/views:
action, default, schema, start, statement, user.
This causes an undesired error when using the column name in a query such as:
SELECT tbl.action FROM sys.keys tbl;
SELECT tbl.default FROM sys.columns tbl;
SELECT tbl.schema FROM sys.statistics tbl;
SELECT tbl.schema FROM sys.storage tbl;
SELECT tbl.start FROM sys.querylog_calls tbl;
SELECT tbl.start FROM sys.querylog_history tbl;
SELECT tbl.start FROM sys.sequences tbl;
SELECT tbl.statement FROM sys.triggers tbl;
SELECT tbl.user FROM sys.connections tbl;
SELECT tbl.user FROM sys.queue tbl;
SELECT tbl.user FROM sys.sessions tbl;
SELECT tbl.user FROM sys.tracelog tbl;
SQuirreL generates such SQL queries when you view the meta data objects. Due to the error returned by MonetDB, no data is returned and thus no meta data in visible in SQuirreL, which is confusing (as the count(*) does show a positive number). As these SQL query syntax is generated, it is not possible to correct it in SQuirreL.
A solution is
a) to avoid these names (which are reserved keywords) in the data dictionary, this implies however that the current names are changed
b) enhance the SQL parser to not give an error as it found a reserved keyword but first try match a column with that name and if successful use it
Reproducible: Always
### Steps to Reproduce:
1. start mserver5 (MonetDB 5 server v11.19.3 "Oct2014")
2. start SQuirreL
3. run the SQL queries:
SELECT tbl.action FROM sys.keys tbl;
SELECT tbl.action FROM tmp.keys tbl;
SELECT tbl.default FROM sys._columns tbl;
SELECT tbl.default FROM sys.columns tbl;
SELECT tbl.default FROM tmp._columns tbl;
SELECT tbl.schema FROM sys.statistics tbl;
SELECT tbl.schema FROM sys.storage tbl;
SELECT tbl.schema FROM sys.storagemodel tbl;
SELECT tbl.schema FROM sys.storagemodelinput tbl;
SELECT tbl.schema FROM sys.tablestoragemodel tbl;
SELECT tbl.start FROM sys.querylog_calls tbl;
SELECT tbl.start FROM sys.querylog_history tbl;
SELECT tbl.start FROM sys.sequences tbl;
SELECT tbl.statement FROM sys.triggers tbl;
SELECT tbl.statement FROM tmp.triggers tbl;
SELECT tbl.user FROM sys.connections tbl;
SELECT tbl.user FROM sys.queue tbl;
SELECT tbl.user FROM sys.sessions tbl;
SELECT tbl.user FROM sys.tracelog tbl;
SELECT action FROM sys.keys;
SELECT action FROM tmp.keys;
SELECT default FROM sys._columns;
SELECT default FROM sys.columns;
SELECT default FROM tmp._columns;
SELECT schema FROM sys.statistics;
SELECT schema FROM sys.storage;
SELECT schema FROM sys.storagemodel;
SELECT schema FROM sys.storagemodelinput;
SELECT schema FROM sys.tablestoragemodel;
SELECT start FROM sys.querylog_calls;
SELECT start FROM sys.querylog_history;
SELECT start FROM sys.sequences;
SELECT statement FROM sys.triggers;
SELECT statement FROM tmp.triggers;
4. in SQuirreL, select objects tab. Open the tree on sys, next SYSTEM TABLE. select the table "_columns" or any of the other above identified tables. select in the right view the tab Content. It should now show the records of the table sys._columns but no records are shown.
### Actual Results:
SELECT default FROM sys._columns
-- Error: syntax error, unexpected DEFAULT in: "select default"
SELECT default FROM sys.columns
-- Error: syntax error, unexpected DEFAULT in: "select default"
SELECT default FROM tmp._columns
-- Error: syntax error, unexpected DEFAULT in: "select default"
SELECT schema FROM sys.statistics
-- Error: syntax error, unexpected SCHEMA in: "select schema"
SELECT schema FROM sys.storage
-- Error: syntax error, unexpected SCHEMA in: "select schema"
SELECT schema FROM sys.storagemodel
-- Error: syntax error, unexpected SCHEMA in: "select schema"
SELECT schema FROM sys.storagemodelinput
-- Error: syntax error, unexpected SCHEMA in: "select schema"
SELECT schema FROM sys.tablestoragemodel
-- Error: syntax error, unexpected SCHEMA in: "select schema"
SELECT start FROM sys.querylog_calls
-- Error: syntax error, unexpected START in: "select start"
SELECT start FROM sys.querylog_history
-- Error: syntax error, unexpected START in: "select start"
SELECT start FROM sys.sequences
-- Error: syntax error, unexpected START in: "select start"
etc.
### Expected Results:
No syntax error for these queries
## Comment 20653
Date: 2015-02-11 20:45:02 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [914313189a7e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=914313189a7e) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=914313189a7e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=914313189a7e)
Changeset description:
added some more (non)-reserved keywords in the exception list. This
solves bug #3613.
## Comment 20657
Date: 2015-02-12 18:20:23 +0100
From: Martin van Dinther <<martin.van.dinther>>
There appears to be even more columns in system tables with names which are keywords. They can be retrieved with following query:
SELECT 'SELECT distinct '||C.name||' FROM '||S.name||'.'||T.name||';' as SQL_query
FROM sys.columns C join sys.tables T on C.table_id = T.id join sys.schemas S on T.schema_id = S.id
WHERE lower(C.name) in ('as', 'action', 'default','schema', 'start', 'statement', 'user', 'authorization', 'column', 'maxvalue', 'minvalue', 'sample', 'table', 'null', 'unique', 'cycle', 'distinct', 'increment', 'plan')
ORDER BY C.name, S.name, T.name
It produces following 40 occurences:
SELECT distinct action FROM sys.keys;
SELECT distinct action FROM tmp.keys;
SELECT distinct as FROM bam.sq;
SELECT distinct authorization FROM sys.schemas;
SELECT distinct column FROM sys.statistics;
SELECT distinct column FROM sys.storage;
SELECT distinct column FROM sys.storagemodel;
SELECT distinct column FROM sys.storagemodelinput;
SELECT distinct cycle FROM sys.sequences;
SELECT distinct default FROM sys._columns;
SELECT distinct default FROM sys.columns;
SELECT distinct default FROM tmp._columns;
SELECT distinct distinct FROM sys.storagemodelinput;
SELECT distinct increment FROM sys.sequences;
SELECT distinct maxvalue FROM sys.sequences;
SELECT distinct minvalue FROM sys.sequences;
SELECT distinct null FROM sys._columns;
SELECT distinct null FROM sys.columns;
SELECT distinct null FROM tmp._columns;
SELECT distinct sample FROM sys.statistics;
SELECT distinct schema FROM sys.statistics;
SELECT distinct schema FROM sys.storage;
SELECT distinct schema FROM sys.storagemodel;
SELECT distinct schema FROM sys.storagemodelinput;
SELECT distinct schema FROM sys.tablestoragemodel;
SELECT distinct start FROM sys.querylog_calls;
SELECT distinct start FROM sys.querylog_history;
SELECT distinct start FROM sys.sequences;
SELECT distinct statement FROM sys.triggers;
SELECT distinct statement FROM tmp.triggers;
SELECT distinct table FROM sys.statistics;
SELECT distinct table FROM sys.storage;
SELECT distinct table FROM sys.storagemodel;
SELECT distinct table FROM sys.storagemodelinput;
SELECT distinct table FROM sys.tablestoragemodel;
SELECT distinct unique FROM sys.statistics;
SELECT distinct user FROM sys.connections;
SELECT distinct user FROM sys.queue;
SELECT distinct user FROM sys.sessions;
SELECT distinct user FROM sys.tracelog;
in default branch two new ones are introduced:
SELECT distinct plan FROM sys.querylog_catalog;
SELECT distinct plan FROM sys.querylog_history;
Following patch could be used for sql_parse.y in section non_reserved_word:
/* next reserved words are used as column names in system tables, see issue 3613. */
| ACTION { $$ = sa_strdup(SA, "action"); } /* sloppy: officially reserved */
| AS { $$ = sa_strdup(SA, "as"); } /* sloppy: officially reserved */
| AUTHORIZATION { $$ = sa_strdup(SA, "authorization"); }/* sloppy: officially reserved */
| COLUMN { $$ = sa_strdup(SA, "column"); } /* sloppy: officially reserved */
| CYCLE { $$ = sa_strdup(SA, "cycle"); } /* sloppy: officially reserved */
| DISTINCT { $$ = sa_strdup(SA, "distinct"); } /* sloppy: officially reserved */
| INCREMENT { $$ = sa_strdup(SA, "increment"); } /* sloppy: officially reserved */
| MAXVALUE { $$ = sa_strdup(SA, "maxvalue"); } /* sloppy: officially reserved */
| MINVALUE { $$ = sa_strdup(SA, "minvalue"); } /* sloppy: officially reserved */
| SQL_PLAN { $$ = sa_strdup(SA, "plan"); } /* sloppy: officially reserved */
| SAMPLE { $$ = sa_strdup(SA, "sample"); } /* sloppy: officially reserved */
| SCHEMA { $$ = sa_strdup(SA, "schema"); } /* sloppy: officially reserved */
| START { $$ = sa_strdup(SA, "start"); } /* sloppy: officially reserved */
| STATEMENT { $$ = sa_strdup(SA, "statement"); } /* sloppy: officially reserved */
| TABLE { $$ = sa_strdup(SA, "table"); } /* sloppy: officially reserved */
| USER { $$ = sa_strdup(SA, "user"); } /* sloppy: officially reserved */
/* end of 3613 */
## Comment 20666
Date: 2015-02-19 14:06:22 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [19f0a1f83f67](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=19f0a1f83f67) made by Martin van Dinther <martin.van.dinther@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=19f0a1f83f67](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=19f0a1f83f67)
Changeset description:
Removed sloppy parsing for default and user as it changes the way the statement is processed.
Added sloppy parsing for keywords: 'as', 'authorization', 'column', 'cycle', 'distinct', 'increment', 'maxvalue', 'minvalue', 'plan', 'sample' and 'table'.
Also updated and extended tests for the occurences of these keywords as column name.
Also included a query which synthesizes the test queries. Whenever the SQL data dictionary changes and new column are introduced which are reserved keywords, this query will detect it.
This fix is related to bug #3613
## Comment 20676
Date: 2015-02-25 20:47:04 +0100
From: @njnes
you could also telll squirel to quote (any) identifier.
| SQL data dictionary contains columns names which are also special keywords. This causes unexpected/unneeded SQL query errors | https://api.github.com/repos/MonetDB/MonetDB/issues/3613/comments | 0 | 2020-11-30T12:57:05Z | 2024-06-27T12:05:25Z | https://github.com/MonetDB/MonetDB/issues/3613 | 753,456,731 | 3,613 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-10-31 13:34:28 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.19.3 (Oct2014)
Last updated: 2014-11-28 10:03:17 +0100
## Comment 20328
Date: 2014-10-31 13:34:28 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:33.0) Gecko/20100101 Firefox/33.0
Build Identifier:
mserver5: gdk_bat.c:2950: BATassertHeadProps: Assertion `!b->H->nil || seennil' failed.
It happens with releases: MonetDB 5 server v11.19.3 "Oct2014" and MonetDB 5 server v11.20.0.
It does NOT happen with release: MonetDB 5 server v11.17.21 "Jan2014-SP3"
Reproducible: Always
### Steps to Reproduce:
1. start mserver5 (MonetDB 5 server v11.19.3 "Oct2014")
2. start mclient
3. execute SQL:
CREATE TABLE PKTABLE ( ptest1 int, ptest2 int, ptest3 text, PRIMARY KEY(ptest1, ptest2) );
CREATE TABLE FKTABLE ( ftest1 int, ftest2 int, ftest3 int, CONSTRAINT constrname FOREIGN KEY(ftest1, ftest2) REFERENCES PKTABLE MATCH FULL ON DELETE SET NULL ON UPDATE SET NULL);
-- Insert test data into PKTABLE
INSERT INTO PKTABLE VALUES (1, 2, 'Test1');
INSERT INTO PKTABLE VALUES (1, 3, 'Test1-2');
INSERT INTO PKTABLE VALUES (2, 4, 'Test2');
INSERT INTO PKTABLE VALUES (5, 10, 'Test5');
SELECT * FROM PKTABLE;
-- Insert successful rows into FK TABLE
INSERT INTO FKTABLE VALUES (1, 2, 4);
INSERT INTO FKTABLE VALUES (1, 3, 5);
INSERT INTO FKTABLE VALUES (2, 4, 8);
INSERT INTO FKTABLE VALUES (NULL, NULL, 0);
-- Check FKTABLE
SELECT * FROM FKTABLE;
-- Delete a row from PK TABLE
DELETE FROM PKTABLE WHERE ptest1=1 and ptest2=2;
DELETE FROM PKTABLE WHERE ptest1=5 and ptest2=10;
### Actual Results:
bash-4.2$ Oct2014b/bin/mserver5 -d10 --set mapi_port=41000
builtin opt gdk_dbpath = /ufs/dinther/Oct2014b/var/monetdb5/dbfarm/demo
builtin opt gdk_debug = 0
builtin opt gdk_vmtrim = no
builtin opt monet_prompt = >
builtin opt monet_daemon = no
builtin opt mapi_port = 50000
builtin opt mapi_open = false
builtin opt mapi_autosense = false
builtin opt sql_optimizer = default_pipe
builtin opt sql_debug = 0
cmdline opt mapi_port = 41000
cmdline opt gdk_debug = 10
MonetDB 5 server v11.19.3 "Oct2014"
Serving database 'demo', using 8 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs dynamically linked
Found 15.356 GiB available main-memory.
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2014 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Listening for connection requests on mapi:monetdb://127.0.0.1:41000/
MonetDB/GIS module loaded
MonetDB/SQL module loaded
>mserver5: gdk_bat.c:2950: BATassertHeadProps: Assertion `!b->H->nil || seennil' failed.
30319 Aborted Oct2014b/bin/mserver5 -d10 --set mapi_port=41000
bash-4.2$
bash-4.2$ mclient -p 41000
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.19.3 (Oct2014), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>CREATE TABLE PKTABLE ( ptest1 int, ptest2 int, ptest3 text, PRIMARY KEY(ptest1, ptest2) );
operation successful (28.670ms)
sql>CREATE TABLE FKTABLE ( ftest1 int, ftest2 int, ftest3 int, CONSTRAINT constrname FOREIGN KEY(ftest1, ftest2) REFERENCES PKTABLE MATCH FULL ON DELETE SET NULL ON UPDATE SET NULL);
operation successful (6.080ms)
sql>INSERT INTO PKTABLE VALUES (1, 2, 'Test1');
1 affected row (64.206ms)
sql>INSERT INTO PKTABLE VALUES (1, 3, 'Test1-2');
1 affected row (4.726ms)
sql>INSERT INTO PKTABLE VALUES (2, 4, 'Test2');
1 affected row (34.725ms)
sql>INSERT INTO PKTABLE VALUES (5, 10, 'Test5');
1 affected row (2.567ms)
sql>SELECT * FROM PKTABLE;
+--------+--------+---------+
| ptest1 | ptest2 | ptest3 |
+========+========+=========+
| 1 | 2 | Test1 |
| 1 | 3 | Test1-2 |
| 2 | 4 | Test2 |
| 5 | 10 | Test5 |
+--------+--------+---------+
4 tuples (2.484ms)
sql>INSERT INTO FKTABLE VALUES (1, 2, 4);
1 affected row (22.689ms)
sql>INSERT INTO FKTABLE VALUES (1, 3, 5);
1 affected row (5.887ms)
sql>INSERT INTO FKTABLE VALUES (2, 4, 8);
1 affected row (4.403ms)
sql>INSERT INTO FKTABLE VALUES (NULL, NULL, 0);
1 affected row (59.025ms)
sql>SELECT * FROM FKTABLE;
+--------+--------+--------+
| ftest1 | ftest2 | ftest3 |
+========+========+========+
| 1 | 2 | 4 |
| 1 | 3 | 5 |
| 2 | 4 | 8 |
| null | null | 0 |
+--------+--------+--------+
4 tuples (1.311ms)
sql>DELETE FROM PKTABLE WHERE ptest1=1 and ptest2=2;
sql>
bash-4.2$
### Expected Results:
no assertion failure.
no termination of mserver5 process.
It is NOT happening with MonetDB 5 server v11.17.21 "Jan2014-SP3" release.
So something in the new Oct2014 release is broken.
## Comment 20405
Date: 2014-10-31 14:54:00 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [f633127500eb](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=f633127500eb) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=f633127500eb](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=f633127500eb)
Changeset description:
Fix for bug #3612: nil property was incorrectly set.
In fact, if the input bat has a nil value, then the output bat doesn't
necessarily have a nil value. If the value gets sign extended (in
this bug from int to lng and on to oid), the nil becomes something
totally different.
## Comment 20468
Date: 2014-11-28 10:03:17 +0100
From: @sjoerdmullender
Oct2014-SP1 has been released.
| assertion failure when deleting rows from table to which a FK constraint is defined | https://api.github.com/repos/MonetDB/MonetDB/issues/3612/comments | 0 | 2020-11-30T12:57:02Z | 2024-06-27T12:05:24Z | https://github.com/MonetDB/MonetDB/issues/3612 | 753,456,704 | 3,612 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-10-31 11:14:17 +0100
From: Anthony Damico <<ajdamico>>
To: SQL devs <<bugs-sql>>
Version: 11.17.21 (Jan2014-SP3)
CC: ajdamico, @hannesmuehleisen
Last updated: 2014-11-28 10:03:21 +0100
## Comment 20324
Date: 2014-10-31 11:14:17 +0100
From: Anthony Damico <<ajdamico>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:33.0) Gecko/20100101 Firefox/33.0
Build Identifier:
command breaks where it shouldn't
Reproducible: Always
### Steps to Reproduce:
CREATE TABLE tempa ( one double , two integer ) ;
INSERT INTO tempa VALUES ( 1.0 , 1 ) , ( 2.0 , 2 ) ;
SELECT quantile( one , 0.25 ) , quantile( one , 0.5 ) FROM tempa WHERE two > 2 ;
### Actual Results:
sql>CREATE TABLE tempa ( one double , two integer ) ;
operation successful (78.651ms)
sql>INSERT INTO tempa VALUES ( 1.0 , 1 ) , ( 2.0 , 2 ) ;
2 affected row (15.009ms)
sql>SELECT quantile( one , 0.25 ) , quantile( one , 0.5 ) FROM tempa WHERE two >
2 ;
quantile value of -1797693134862315700000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000000000000000000000000000000000
[00000000000000](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=00000000000000) is not in range [0,1]
sql>
### Expected Results:
two columns with zero records or missing values? not this.
## Comment 20325
Date: 2014-10-31 11:33:53 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [849b925afb3d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=849b925afb3d) made by Hannes Muehleisen <hannes@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=849b925afb3d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=849b925afb3d)
Changeset description:
Test for Bug #3611
## Comment 20326
Date: 2014-10-31 11:34:08 +0100
From: @hannesmuehleisen
confirmed in default
## Comment 20406
Date: 2014-10-31 15:41:01 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [957ab8b84580](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=957ab8b84580) made by Hannes Muehleisen <hannes@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=957ab8b84580](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=957ab8b84580)
Changeset description:
Test for Bug #3611
(grafted from 849b925afb3d3dd3f2f7a0da84cf4115fc43e5ac)
## Comment 20407
Date: 2014-10-31 15:41:05 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [a3da268f796f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=a3da268f796f) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=a3da268f796f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=a3da268f796f)
Changeset description:
Fix for bug #3611.
If the main bat is empty, the quantile bat may also be empty. In that
case, just use 0.5 as quantile value so that we get a result with a
nil value.
## Comment 20472
Date: 2014-11-28 10:03:21 +0100
From: @sjoerdmullender
Oct2014-SP1 has been released.
| quantile() and median() commands crash when used 2x on the same variable on a null table | https://api.github.com/repos/MonetDB/MonetDB/issues/3611/comments | 0 | 2020-11-30T12:57:00Z | 2024-06-27T12:05:23Z | https://github.com/MonetDB/MonetDB/issues/3611 | 753,456,671 | 3,611 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.