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: 2014-07-10 17:29:14 +0200
From: Richard Hughes <<richard.monetdb>>
To: SQL devs <<bugs-sql>>
Version: 11.17.17 (Jan2014-SP2)
CC: @njnes
Last updated: 2014-10-31 14:14:34 +0100
## Comment 19897
Date: 2014-07-10 17:29:14 +0200
From: Richard Hughes <<richard.monetdb>>
sql>select timestamp '1-1-1'+tmonth*interval '1' month from (select 2014*12+7) as t(tmonth);
TypeException:user.s27_3[12]:'mtime.timestamp_add_month_interval' undefined in:
_19:any := mtime.timestamp_add_month_interval(_9:timestamp, _18:lng)
program contains errors
sql>explain select timestamp '1-1-1'+tmonth*interval '1' month from (select 2014*12+7) as t(tmonth);
TypeException:user.s28_3[12]:'mtime.timestamp_add_month_interval' undefined in:
_19:any := mtime.timestamp_add_month_interval(_9:timestamp, _18:lng)
+-----------------------------------------------------------------------------+
| mal |
+=============================================================================+
| function user.s28_3{autoCommit=true}(A0:timestamp,A1:int,A2:int,A3:sht,A4:l |
: ng):void; :
| sql.mvc(); |
| A0; |
| calc.timestamp(X_8,7); |
| A1; |
| A2; |
| A3; |
| calc.*(X_12,X_13); |
| calc.lng(X_14); |
| A4; |
| calc.+(X_15,X_16); |
| calc.*(X_11,X_17); |
| mtime.timestamp_add_month_interval(X_9,X_18); |
| sql.exportValue(1,".L","sql_add_single_value","timestamp",7,0,12,X_19," |
: "); :
| end s28_3; |
| querylog.define("explain select timestamp \\'1-1-1\\'+tmonth*interval \\' |
: 1\\' month from (select 2014*12+7) as t(tmonth);","default_pipe") :
+-----------------------------------------------------------------------------+
Notice that the MAL doesn't contain any variable names. This seems...curious...to me.
Expected result:
+----------------------------+
| single_value |
+============================+
| 2014-08-01 [000000000000](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=000000000000) |
+----------------------------+
The code works if I use interval '1' day (or any interval shorter than a month) instead. Likewise '1' year doesn't work. Basically timestamp_add_month_interval() appears broken and timestamp_add_msec_interval() seems fine. In this case. Simpler queries like "select timestamp '1-1-1' + 5 * interval '1' month" work.
The back story (probably irrelevant to finding & fixing the bug):
I'm trying to aggregate a large table containing a timestamp column by the year+month of that column, and ended up with a query something like:
select timestamp '1-1-1'+(tmonth-1)*interval '1' month,count(*) from (select "year"(t)*12+"month"(t) from data) as t(tmonth) group by tmonth;
If you can suggest a more cunning way to implement that query then I'm all ears.
## Comment 19898
Date: 2014-07-11 12:31:43 +0200
From: Richard Hughes <<richard.monetdb>>
Another data point (possibly simpler to debug):
Works:
select interval '1' month*("year"(t)) from data;
Does not work:
select interval '1' month*("year"(t)*1) from data;
TypeException:user.s45_2[26]:'bat.insert' undefined in: _76:any := bat.insert(_68:bat[:oid,:int], _74:oid, _73:lng)
## Comment 20285
Date: 2014-10-11 18:57:06 +0200
From: @njnes
added the missing mal interface (with lng type for months)
## Comment 20379
Date: 2014-10-31 14:14:34 +0100
From: @sjoerdmullender
Oct2014 has been released.
| timestamp + month interval generates bogus MAL? | https://api.github.com/repos/MonetDB/MonetDB/issues/3510/comments | 0 | 2020-11-30T12:52:18Z | 2024-06-27T12:03:41Z | https://github.com/MonetDB/MonetDB/issues/3510 | 753,453,469 | 3,510 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-07-04 17:47:54 +0200
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: -- development
Last updated: 2016-04-11 11:46:25 +0200
## Comment 19884
Date: 2014-07-04 17:47:54 +0200
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:30.0) Gecko/20100101 Firefox/30.0
Build Identifier:
MonetDB currently does not support conversions of numeric or date or timestamp data to a string in a specific layout format, as is possible with to_char(num_ts_val, 'format_mask')
See also:
http://www.postgresql.org/docs/9.3/interactive/functions-formatting.html
and
http://www.techonthenet.com/oracle/functions/to_char.php
Reproducible: Always
tests for to_char() are already defined in
sql/test/pg_regress/sql/in8.sql
sql/test/pg_regress/sql/numeric.sql
sql/test/pg_regress/sql/timestamp.sql
sql/test/pg_regress/sql/timestamptz.sql
but they currently all produce errors as the to_char() is not supported.
| add support for formatted conversion to string using function: to_char(num_ts_val, 'format_mask') | https://api.github.com/repos/MonetDB/MonetDB/issues/3509/comments | 0 | 2020-11-30T12:52:16Z | 2024-06-28T13:40:19Z | https://github.com/MonetDB/MonetDB/issues/3509 | 753,453,448 | 3,509 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-07-04 16:47:13 +0200
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.17.17 (Jan2014-SP2)
CC: @njnes
Last updated: 2014-10-31 14:14:02 +0100
## Comment 19882
Date: 2014-07-04 16:47:13 +0200
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:30.0) Gecko/20100101 Firefox/30.0
Build Identifier:
Implicit conversion of strings '0 ' or ' 1234 ' or '123456 ' to type smallint or integer or bigint fails, but ' -1234' does not fail.
Reproducible: Always
### Steps to Reproduce:
CREATE TABLE INT2_TBL(f1 smallint);
INSERT INTO INT2_TBL(f1) VALUES ('0 ');
INSERT INTO INT2_TBL(f1) VALUES (' 1234 ');
CREATE TABLE INT4_TBL(f1 integer);
INSERT INTO INT4_TBL(f1) VALUES (' 0 ');
INSERT INTO INT4_TBL(f1) VALUES ('123456 ');
### Actual Results:
QUERY = INSERT INTO INT2_TBL(f1) VALUES ('0 ');
ERROR = !conversion of string '0 ' to type sht failed.
QUERY = INSERT INTO INT2_TBL(f1) VALUES (' 1234 ');
ERROR = !conversion of string ' 1234 ' to type sht failed.
QUERY = INSERT INTO INT4_TBL(f1) VALUES (' 0 ');
ERROR = !conversion of string ' 0 ' to type int failed.
QUERY = INSERT INTO INT4_TBL(f1) VALUES ('123456 ');
ERROR = !conversion of string '123456 ' to type int failed.
### Expected Results:
These string with trailing spaces should be accepted in the implicit conversion from string to integer.
For instance the following inserts are accepted:
INSERT INTO INT2_TBL(f1) VALUES (' -1234');
INSERT INTO INT4_TBL(f1) VALUES (' -123456');
See sql/test/pg_regress/Tests/int2.sql
and sql/test/pg_regress/Tests/int4.sql
## Comment 19883
Date: 2014-07-04 16:59:59 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [1715d0918489](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=1715d0918489) 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=1715d0918489](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=1715d0918489)
Changeset description:
Converted castings of int2, int4, int8 to cast() syntax
Some tests fail in MonetDB which should not fail. See bug #3508
Updated stable.out and stable.err for int2, int4 and int8
Added substitution rules to the postgress2sql99.sh script
## Comment 19885
Date: 2014-07-04 18:47:30 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [c47ea152d459](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=c47ea152d459) 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=c47ea152d459](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=c47ea152d459)
Changeset description:
Converted castings of float4 and float8 to cast() syntax
Some tests fail in MonetDB which should not fail. See bug #3508.
Updated stable.out and stable.err for float4 and float8
Added substitution rules to postgress2sql99.sh utility script
## Comment 19886
Date: 2014-07-09 10:45:38 +0200
From: @njnes
current code indeed allows for leading spaces and not for trailing spaces.
Removing the code for allowing leading spaces is easy and will result in all example inserts to fail.
## Comment 19942
Date: 2014-08-02 19:03:29 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [9d7469c47ad8](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9d7469c47ad8) 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=9d7469c47ad8](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=9d7469c47ad8)
Changeset description:
fix for bug #3508, ie handle trailing spaces
## Comment 20352
Date: 2014-10-31 14:14:02 +0100
From: @sjoerdmullender
Oct2014 has been released.
| conversion of string '0 ' to type smallint or integer fails | https://api.github.com/repos/MonetDB/MonetDB/issues/3508/comments | 0 | 2020-11-30T12:52:14Z | 2024-06-27T12:03:39Z | https://github.com/MonetDB/MonetDB/issues/3508 | 753,453,427 | 3,508 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-07-03 13:48:31 +0200
From: mike <<mike_1975>>
To: SQL devs <<bugs-sql>>
Version: 11.17.17 (Jan2014-SP2)
CC: mike_1975
Last updated: 2014-07-04 16:16:08 +0200
## Comment 19865
Date: 2014-07-03 13:48:31 +0200
From: mike <<mike_1975>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Firefox/24.0
Build Identifier:
i have a sample dataset created which contains a number and a varchar.
here is the schema and other details on how to reproduce it reliably every time.
drop table test;
create table test(id bigint, msg varchar(300));
COPY INTO test from '/tmp/go_test.txt' USING DELIMITERS ',';
SELECT id,msg from test where msg like '%g%' order by id limit 8000;
the above query runs perfectly fine. however the below query crashes it everytime.
SELECT id,msg from test where msg like '%go%' order by id limit 8000;
also, i have seen it crashing it with other queries ( but i failed to note them down)
LIMIT or no LIMIT sometimes make a difference for eg if u put a suitable limit, it doesn't crash, but as soon as you increase limit, it starts crashing again . However, if you don't select the varchar column, the crash never occurs.
Reproducible: Always
### Steps to Reproduce:
1. mentioned in details above
2.
3.
### Actual Results:
mserver crashes with
Bus error (core dumped)
### Expected Results:
query should run without crashing
if i run with gdb attached, it gives following information,
Program received signal SIGBUS, Bus error.
[Switching to Thread 0x7fffee2e3700 (LWP 12180)]
0x00000034e853fd10 in __memcpy_ssse3 () from /lib64/libc.so.6
(gdb) bt
0 0x00000034e853fd10 in __memcpy_ssse3 () from /lib64/libc.so.6
1 0x00007ffff743f68d in insert_string_bat () from /tmp/lib/libbat.so.10
2 0x00007ffff7446bb3 in BATappend () from /tmp/lib/libbat.so.10
3 0x00007ffff7c37145 in MATpackIncrement () from /tmp/lib/libmonetdb5.so.16
4 0x00007ffff7b73180 in runMALsequence () from /tmp/lib/libmonetdb5.so.16
5 0x00007ffff7b74f3e in DFLOWworker () from /tmp/lib/libmonetdb5.so.16
6 0x00000034e8c07de3 in start_thread () from /lib64/libpthread.so.0
7 0x00000034e84f616d in clone () from /lib64/libc.so.6
(gdb)
## Comment 19866
Date: 2014-07-03 14:16:25 +0200
From: mike <<mike_1975>>
you can download the dataset from here for the next 20 days.admins can u attach the dataset to this bug??
http://www.fileconvoy.com/dfl.php?id=ga8001b9ac7331ae2999524854afe2808cc87baa8e
( the dataset is around 4 MB while bugzilla allowed only 1 MB )
## Comment 19867
Date: 2014-07-03 15:26:59 +0200
From: @sjoerdmullender
A Bus Error is almost always due to a full disk. (I have never seen any other cause, but can't categorically rule out any other cause.)
MonetDB uses memory maps for larger chunks of data. When a new BAT is created, a file is allocated and resized to whatever it needs to be. This file is then mapped (mmap) and the mapped memory is written to. Linux doesn't allocate the space for the file until it is needed. When space is needed because the memory map is written to disk, but the disk is too full, Linux will kill the process with a Bus Error.
In other words, please check your disk space.
## Comment 19870
Date: 2014-07-03 17:36:35 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [45acd0d7a20a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=45acd0d7a20a) 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=45acd0d7a20a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=45acd0d7a20a)
Changeset description:
Use posix_fallocate if available to extend files.
Using this function will ensure that disk blocks are allocated, so
when the file is the memory mapped and the memory then written to, we
should never get a Bus Error because the disk is full.
This hopefully fixes bug #64, bug #3202, and bug #3507, and possibly also
bug #3502.
## Comment 19872
Date: 2014-07-03 17:58:11 +0200
From: mike <<mike_1975>>
My test machine has atleast 1000 GB disk free space, moreover i am able to import the entire test dataset (size 1 GB) without any issues not only in 1 table but more than 10 tables, so space is most probably not an issue. the one provided to you is just a cut out part to recreate the issue.
Moreover as pointed out above
This query runs fine
a) SELECT id,msg from test where msg like '%g%' order by id limit 8000;
this query crashes
b) SELECT id,msg from test where msg like '%go%' order by id limit 8000;
Now, if you see the the queries in detail ( query a is superset of query b).
query *a* results will contain query *b* results + some other results also.
so, actually by inference, query a should also crash ( if space is issue). However, this doesn't happen, so problem is probably somewhere else.
Also, as i mentioned, i am not only seeing this with %go% but also with some other words, such as %co% ( not in the dataset attached , but in other dataset i have created for testing purposes, let me know, if you need that also)
also, note that this crash only happens when varchar field is selected, so issue might be some other place and might need a deep look.
Also, i will try out your fix also.. and update ASAP
## Comment 19873
Date: 2014-07-03 18:13:12 +0200
From: mike <<mike_1975>>
The following changeset doesn't fix the issue/bug. the database still crashes.
Changeset [45acd0d7a20a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=45acd0d7a20a) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
Also, can you attach the dataset file from comment 1 to this bug, so as the dataset doesn't get lost after 20 days.
i will be conductiing more tests today and update this bug if i have more information
## Comment 19874
Date: 2014-07-03 18:31:24 +0200
From: mike <<mike_1975>>
Also, i have compiled monetdb with debug options --enable-assert --enable-debug -disable-optimize --enable-developer
and then again ran the test mentioned above, please note that your fix mentioned above was also applied.
I then obtained the backtrace from gdb.
However, the stack show clearly that the space is not the problem. the problem occurs probably in strput.
(gdb) bt
0 0x00000034e8489f3b in __memset_sse2 () from /lib64/libc.so.6
1 0x00007ffff7548831 in strPut (h=0x7fffd80048a0, dst=0x7fffedfc0610,
v=0x7fffee72c268 "goldlasso.com") at gdk_atoms.c:1120
2 0x00007ffff7751d1d in project_any (bn=0x7fffd8003870, l=0x7fffd8003af0, r=0x7fffe40030f0,
nilcheck=0) at gdk_join.c:2885
3 0x00007ffff77539d5 in BATproject (l=0x7fffd8003af0, r=0x7fffe40030f0) at gdk_join.c:3039
4 0x00007fffefb206d5 in DELTAproject (result=0x7fffe00b8890, sub=0x7fffe00b8850,
col=0x7fffe00b8750, uid=0x7fffe00b8770, uval=0x7fffe00b8780, ins=0x0) at sql.c:2131
5 0x00007fffefb1f932 in DELTAproject2 (result=0x7fffe00b8890, sub=0x7fffe00b8850,
col=0x7fffe00b8750, uid=0x7fffe00b8770, uval=0x7fffe00b8780) at sql.c:1917
6 0x00007ffff7af2cf5 in malCommandCall (stk=0x7fffe00b82e0, pci=0x7fffe00c8a70)
at mal_interpreter.c:103
---Type <return> to continue, or q <return> to quit---
7 0x00007ffff7af566b in runMALsequence (cntxt=0x6462a8, mb=0x7fffe00bd2d0, startpc=32,
stoppc=33, stk=0x7fffe00b82e0, env=0x0, pcicaller=0x0) at mal_interpreter.c:651
8 0x00007ffff7af9a6d in DFLOWworker (T=0x7ffff7f261d8 <workers+56>) at mal_dataflow.c:358
9 0x00000034e8c07de3 in start_thread () from /lib64/libpthread.so.0
10 0x00000034e84f616d in clone () from /lib64/libc.so.6
(gdb)
Can you run the test on your end and confirm the issue.
## Comment 19875
Date: 2014-07-03 22:55:40 +0200
From: @sjoerdmullender
(In reply to comment 1)
> you can download the dataset from here for the next 20 days.admins can u
> attach the dataset to this bug??
> http://www.fileconvoy.com/dfl.
> php?id=ga8001b9ac7331ae2999524854afe2808cc87baa8e
>
> ( the dataset is around 4 MB while bugzilla allowed only 1 MB )
I copied the dataset to my system, so I can try to reproduce the problem.
## Comment 19876
Date: 2014-07-04 11:37:27 +0200
From: @sjoerdmullender
I tried to reproduce the crash but failed.
I see that the crash in your case happened inside MATpackIncrement which suggests the number of cores on your system might be relevant.
So what are the specs of your system? In particular, how many threads does mserver5 report it is using?
## Comment 19877
Date: 2014-07-04 11:44:32 +0200
From: @sjoerdmullender
(In reply to comment 6)
> (gdb) bt
> 0 0x00000034e8489f3b in __memset_sse2 () from /lib64/libc.so.6
> 1 0x00007ffff7548831 in strPut (h=0x7fffd80048a0, dst=0x7fffedfc0610,
> v=0x7fffee72c268 "goldlasso.com") at gdk_atoms.c:1120
What is the result of
print *h
at this point?
> 2 0x00007ffff7751d1d in project_any (bn=0x7fffd8003870, l=0x7fffd8003af0,
> r=0x7fffe40030f0,
> nilcheck=0) at gdk_join.c:2885
> 3 0x00007ffff77539d5 in BATproject (l=0x7fffd8003af0, r=0x7fffe40030f0) at
> gdk_join.c:3039
> 4 0x00007fffefb206d5 in DELTAproject (result=0x7fffe00b8890,
> sub=0x7fffe00b8850,
> col=0x7fffe00b8750, uid=0x7fffe00b8770, uval=0x7fffe00b8780, ins=0x0) at
> sql.c:2131
> 5 0x00007fffefb1f932 in DELTAproject2 (result=0x7fffe00b8890,
> sub=0x7fffe00b8850,
> col=0x7fffe00b8750, uid=0x7fffe00b8770, uval=0x7fffe00b8780) at
> sql.c:1917
> 6 0x00007ffff7af2cf5 in malCommandCall (stk=0x7fffe00b82e0,
> pci=0x7fffe00c8a70)
> at mal_interpreter.c:103
> ---Type <return> to continue, or q <return> to quit---
> 7 0x00007ffff7af566b in runMALsequence (cntxt=0x6462a8, mb=0x7fffe00bd2d0,
> startpc=32,
> stoppc=33, stk=0x7fffe00b82e0, env=0x0, pcicaller=0x0) at
> mal_interpreter.c:651
> 8 0x00007ffff7af9a6d in DFLOWworker (T=0x7ffff7f261d8 <workers+56>) at
> mal_dataflow.c:358
> 9 0x00000034e8c07de3 in start_thread () from /lib64/libpthread.so.0
> 10 0x00000034e84f616d in clone () from /lib64/libc.so.6
> (gdb)
## Comment 19878
Date: 2014-07-04 13:16:26 +0200
From: mike <<mike_1975>>
Testing is being done in a Virtualbox VM (however, the crash even occurred on a real machine, having 16GB RAM and AMD FX-6300).
The VM has 4 GB RAM and 2 cores and mserver reports the following at startup
Serving database 'test_db', using 2 threads
Found 3.830 GiB available main-memory.
The VM is running Oracle linux 7 beta.
Also,your request for print *h is below.
(gdb) print *h
Cannot access memory at address 0x0;
(gdb)
## Comment 19879
Date: 2014-07-04 13:52:50 +0200
From: @sjoerdmullender
(In reply to comment 10)
> Testing is being done in a Virtualbox VM (however, the crash even occurred
> on a real machine, having 16GB RAM and AMD FX-6300).
> The VM has 4 GB RAM and 2 cores and mserver reports the following at startup
> Serving database 'test_db', using 2 threads
> Found 3.830 GiB available main-memory.
Pity, I still can't reproduce, even with 2 threads.
> The VM is running Oracle linux 7 beta.
>
> Also,your request for print *h is below.
> (gdb) print *h
> Cannot access memory at address 0x0;
> (gdb)
This suggests h == 0x0, but in the stack trace it distinctly is not. And there are no assignments to h in the strPut function.
## Comment 19880
Date: 2014-07-04 15:36:24 +0200
From: mike <<mike_1975>>
I have figured out the issue and culprit has been a mistake at my end.
Sorry, but this was due to a failed/partial merged copy operation resulting in faulty GDKload function code/file which somehow resulted in faulty bats in case of varchars only.
thanks for the help and providing me pointers to track it down further.
I am marking it closed since the bug was invalid.
## Comment 19881
Date: 2014-07-04 16:16:08 +0200
From: @sjoerdmullender
I'm glad the issue is resolved. It did trigger the change referred to in comment 3, which is a good thing.
| mserver/mclient crashes if a string column is selected in output column | https://api.github.com/repos/MonetDB/MonetDB/issues/3507/comments | 0 | 2020-11-30T12:52:11Z | 2024-06-28T13:16:55Z | https://github.com/MonetDB/MonetDB/issues/3507 | 753,453,391 | 3,507 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-30 16:36:12 +0200
From: @bartscheers
To: SQL devs <<bugs-sql>>
Version: 11.17.17 (Jan2014-SP2)
CC: @njnes
Last updated: 2014-10-31 14:13:58 +0100
## Comment 19863
Date: 2014-06-30 16:36:12 +0200
From: @bartscheers
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:30.0) Gecko/20100101 Firefox/30.0
Build Identifier:
Second of two insert statements terminates mserver5. Type conversion from number to varchar goes wrong.
Reproducible: Always
### Steps to Reproduce:
1.monetdb stop ins_bug; monetdb destroy -f ins_bug; monetdb create ins_bug; monetdb release ins_bug; monetdb start ins_bug #2.CREATE TABLE config
(id SERIAL
,dataset INT NOT NULL
,section VARCHAR(100)
,key VARCHAR(100)
,value VARCHAR(500)
,type VARCHAR(5)
,UNIQUE (dataset, section, key)
)
;
3. INSERT INTO config (dataset, section, key, value, type) VALUES (6, 'quality_lofar', 'low_bound', 1.9, 'float');
INSERT INTO config (dataset, section, key, value, type) VALUES (6, 'quality_lofar', 'high_bound', 80, 'int');
### Actual Results:
stopping database 'ins_bug'... done
destroyed database: ins_bug
created database in maintenance mode: ins_bug
taken database out of maintenance mode: ins_bug
starting database 'ins_bug'... done
operation successful
Connection terminated
### Expected Results:
another succesful operation...
## Comment 19888
Date: 2014-07-09 10:47:49 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [17cd8c14c3b4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=17cd8c14c3b4) 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=17cd8c14c3b4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=17cd8c14c3b4)
Changeset description:
don't assert but return error on atom_cast (fix bug #3506)
## Comment 19933
Date: 2014-07-28 18:58:40 +0200
From: @njnes
Added test varchar_conversion_crash.Bug-3506.sql
## Comment 19934
Date: 2014-07-28 19:36:19 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [d89feba26343](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d89feba26343) 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=d89feba26343](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=d89feba26343)
Changeset description:
added test for bug #3506
## Comment 20347
Date: 2014-10-31 14:13:58 +0100
From: @sjoerdmullender
Oct2014 has been released.
| conversion to varchar terminates mserver | https://api.github.com/repos/MonetDB/MonetDB/issues/3506/comments | 0 | 2020-11-30T12:52:08Z | 2024-06-27T12:03:37Z | https://github.com/MonetDB/MonetDB/issues/3506 | 753,453,362 | 3,506 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-27 17:22:18 +0200
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.17.17 (Jan2014-SP2)
CC: @njnes
Last updated: 2014-10-31 14:15:01 +0100
## Comment 19858
Date: 2014-06-27 17:22:18 +0200
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:30.0) Gecko/20100101 Firefox/30.0
Build Identifier:
SQL statements with syntax
fail to execute
SELECT '' AS "Not False" WHERE f1 = NOT FALSE;
SELECT '' AS "Not True" WHERE f1 = NOT TRUE;
Reproducible: Always
### Steps to Reproduce:
CREATE TABLE BOOLTBL1 (f1 bool);
INSERT INTO BOOLTBL1 (f1) VALUES (cast('true' AS boolean));
INSERT INTO BOOLTBL1 (f1) VALUES ('true');
INSERT INTO BOOLTBL1 (f1) VALUES ('false');
-- next two queries fail with a syntax error:
SELECT f1 FROM BOOLTBL1 WHERE f1 = NOT FALSE;
SELECT f1 FROM BOOLTBL1 WHERE f1 = NOT TRUE;
-- next similar queries run fine:
SELECT f1 FROM BOOLTBL1 WHERE f1 = (NOT FALSE);
SELECT f1 FROM BOOLTBL1 WHERE NOT FALSE = f1;
SELECT f1 FROM BOOLTBL1 WHERE NOT f1 = FALSE;
SELECT f1 FROM BOOLTBL1 WHERE f1 = (NOT TRUE);
SELECT f1 FROM BOOLTBL1 WHERE NOT TRUE = f1;
SELECT f1 FROM BOOLTBL1 WHERE NOT f1 = TRUE;
something is wrong with the order of execution of an = NOT expression
### Actual Results:
Error: syntax error, unexpected NOT in: "select f1 from booltbl1 where f1 = not"
### Expected Results:
no syntax error and query result output
These tests are already included in sql/test/pg_regress/Test/boolean.sql
together with corrected stable.out and stable.err files.
So once this bug is resolved that test should report OK instead of current MAJOR
## Comment 19887
Date: 2014-07-09 10:47:46 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [be17bc388a09](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=be17bc388a09) 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=be17bc388a09](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=be17bc388a09)
Changeset description:
add not true/not false parsing (solving bug #3505)
## Comment 19982
Date: 2014-08-06 19:06:53 +0200
From: @njnes
fixed handling of bool with not.
## Comment 19987
Date: 2014-08-06 19:12:37 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [516d270cbe32](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=516d270cbe32) 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=516d270cbe32](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=516d270cbe32)
Changeset description:
fixed bugs 3505 and 3511
## Comment 20402
Date: 2014-10-31 14:15:01 +0100
From: @sjoerdmullender
Oct2014 has been released.
| expression with <boolean> = NOT <boolean> returns a syntax error but NOT <boolean> = <boolean> not | https://api.github.com/repos/MonetDB/MonetDB/issues/3505/comments | 0 | 2020-11-30T12:52:05Z | 2024-06-27T12:03:36Z | https://github.com/MonetDB/MonetDB/issues/3505 | 753,453,328 | 3,505 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-25 16:16:56 +0200
From: @hannesmuehleisen
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2014-10-31 14:14:51 +0100
## Comment 19856
Date: 2014-06-25 16:16:56 +0200
From: @hannesmuehleisen
Often, we know that a CSV file has a header, but we do not want to specify the number of rows to be imported. However, the COPY INTO command currently does only allow OFFSET 2 (to skip the header) to be specified when n RECORDS is also present. This makes little sense and should be fixed.
So,
COPY [ int_val [ OFFSET int_val ] RECORDS ] INTO table_name
FROM ( file_name ',' ... | STDIN )
[ [USING] DELIMITERS field_separator [',' record_separator [ ',' string_quote ]]] [ NULL AS null_string ] [LOCKED]
should be
COPY [ int_val RECORDS ] [ OFFSET int_val ] INTO table_name
FROM ( file_name ',' ... | STDIN )
[ [USING] DELIMITERS field_separator [',' record_separator [ ',' string_quote ]]] [ NULL AS null_string ] [LOCKED]
## Comment 19893
Date: 2014-07-09 15:46:27 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [08f2048691fb](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=08f2048691fb) 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=08f2048691fb](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=08f2048691fb)
Changeset description:
added limit without records in copy into. Beware this can only work
with file based input (not with streaming!) Fixes bug #3504
## Comment 19983
Date: 2014-08-06 19:08:28 +0200
From: @njnes
fixed
## Comment 20393
Date: 2014-10-31 14:14:51 +0100
From: @sjoerdmullender
Oct2014 has been released.
| COPY INTO does not allow OFFSET without specifying amount of records | https://api.github.com/repos/MonetDB/MonetDB/issues/3504/comments | 0 | 2020-11-30T12:52:03Z | 2024-06-27T12:03:35Z | https://github.com/MonetDB/MonetDB/issues/3504 | 753,453,294 | 3,504 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-25 05:20:10 +0200
From: Clint Cummins <<ccummins>>
To: clients devs <<bugs-clients>>
Version: 11.17.17 (Jan2014-SP2)
CC: @hannesmuehleisen
Duplicates: #3471
Last updated: 2014-06-30 23:55:02 +0200
## Comment 19851
Date: 2014-06-25 05:20:10 +0200
From: Clint Cummins <<ccummins>>
User-Agent: Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.1; WOW64; Trident/4.0; SLCC2; .NET CLR 2.0.50727; .NET CLR 3.5.30729; .NET CLR 3.0.30729; Media Center PC 6.0; InfoPath.3; .NET4.0C; .NET4.0E)
Build Identifier:
A column of type TINYINT displays incorrectly in SQuirreL SQL Client -
the value shown is too high by 48.
Examples: true value 0, displayed value 48; true value 4, displayed value 52 .
Other column types, such as SMALLINT display perfectly.
Occurs with both monetdb-jdbc-2.11.jar and in 2.10 .
I do not know if it is a bug in the MonetDB JDBC interface, or a bug in Squirrel.
Reproducible: Always
### Steps to Reproduce:
1. Create a table TINYBUG where one column C1 is type TINYINT.
2. Put a few values in the column C1, like 0 1 2 3 4 .
3. Connect with SQuirreL and SELECT * FROM TINYBUG;
### Actual Results:
It displays 48 49 50 51 52
### Expected Results:
It should display 0 1 2 3 4
64-bit
libpcre: 8.13 2011-08-16
openssl: OpenSSL 1.0.1b 26 Apr 2012
libxml2: 2.7.8
## Comment 19852
Date: 2014-06-25 05:21:50 +0200
From: Clint Cummins <<ccummins>>
Created attachment 288
screen shots from mclient and Squirrel showing bug
> Attached file: [flightdelays_mclient_squirrel_bug.gif](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3503_flightdelays_mclient_squirrel_bug.gif_288) (image/gif, 58322 bytes)
> Description: screen shots from mclient and Squirrel showing bug
## Comment 19853
Date: 2014-06-25 09:20:57 +0200
From: @hannesmuehleisen
On it.
## Comment 19854
Date: 2014-06-25 09:44:19 +0200
From: @hannesmuehleisen
After updating the JDBC driver to the latest version from Mercurial, the problem disappeared. Could you test with the driver in the attachment?
## Comment 19855
Date: 2014-06-25 09:44:40 +0200
From: @hannesmuehleisen
Created attachment 289
JDBC driver JAR
> Attached file: [monetdb-jdbc-2.11.jar](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3503_monetdb-jdbc-2.11.jar_289) (application/x-java-archive, 113781 bytes)
> Description: JDBC driver JAR
## Comment 19857
Date: 2014-06-27 00:27:18 +0200
From: Clint Cummins <<ccummins>>
(In reply to comment 3)
> After updating the JDBC driver to the latest version from Mercurial, the
> problem disappeared. Could you test with the driver in the attachment?
Yes - the problem is fixed!
Thank you for your extremely fast and helpful work.
## Comment 19861
Date: 2014-06-30 14:20:08 +0200
From: @sjoerdmullender
*** This bug has been marked as a duplicate of bug #3471 ***
## Comment 19864
Date: 2014-06-30 23:55:02 +0200
From: Clint Cummins <<ccummins>>
Thank you.
It appears my scan of recent bug reports for JDBC
failed because it takes about 10 seconds for the Description section
of the bug report to appear in my browser. Now I know to wait for it!
| TINYINT displays incorrectly in SQuirreL - is too high by 48 | https://api.github.com/repos/MonetDB/MonetDB/issues/3503/comments | 0 | 2020-11-30T12:52:00Z | 2024-06-28T13:16:54Z | https://github.com/MonetDB/MonetDB/issues/3503 | 753,453,270 | 3,503 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-18 16:58:45 +0200
From: andrei <<andrei.ciobotar>>
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.17.17 (Jan2014-SP2)
CC: @hannesmuehleisen, @mlkersten
Last updated: 2014-10-31 14:13:48 +0100
## Comment 19848
Date: 2014-06-18 16:58:45 +0200
From: andrei <<andrei.ciobotar>>
sql>\d table
CREATE TABLE "sys"."table" (
"bigint1" BIGINT DEFAULT '0',
"date" DATE DEFAULT '1970-01-01',
"bigint2" BIGINT DEFAULT '0',
"bigint3" BIGINT DEFAULT '0',
"bigint4" BIGINT DEFAULT '1'
);
Table has 1,556,041,536 entries.
We're attempting to aggregate on a day by day basis over the bigint4 field and are running the following query (example for a single day):
'INSERT INTO table_aggregation select "bigint1", "date", "bigint2", "bigint3", sum("bigint4") from table WHERE "date" = CURRENT_DATE() - INTERVAL '6' DAY group by "bigint1", "date", "bigint2", "bigint3"'
Every 2-3 queries, the database where the table resides is killed by SIGBUS:
2014-06-18 09:17:25 MSG merovingian[15877]: database 'table_database' (9656) was killed by signal SIGBUS
After the kill signal, the first database connection attempt is refused:
2014-06-18 09:33:35 ERR merovingian[15877]: client error: cannot connect: Connection refused
We need to attempt the connection a second time for it to be successful. Any info on the possible causes of SIGBUS and how to avoid this kind of signal is appreciated.
Kind regards,
## Comment 19849
Date: 2014-06-18 17:00:02 +0200
From: @hannesmuehleisen
Are you sure there is enough space on your disk?
## Comment 19850
Date: 2014-06-18 17:04:30 +0200
From: andrei <<andrei.ciobotar>>
Yea, we're sitting at around 75-80% usage when the error hits (with around 20GB available). It's currently running on a test machine with 24GB RAM and is swapping heavily.
## Comment 19871
Date: 2014-07-03 17:36:37 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [45acd0d7a20a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=45acd0d7a20a) 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=45acd0d7a20a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=45acd0d7a20a)
Changeset description:
Use posix_fallocate if available to extend files.
Using this function will ensure that disk blocks are allocated, so
when the file is the memory mapped and the memory then written to, we
should never get a Bus Error because the disk is full.
This hopefully fixes bug #64, bug #3202, and bug #3507, and possibly also
bug #3502.
## Comment 20250
Date: 2014-10-04 23:30:59 +0200
From: @mlkersten
Let's close this one until further information
## Comment 20338
Date: 2014-10-31 14:13:48 +0100
From: @sjoerdmullender
Oct2014 has been released.
| Database was killed by signal SIGBUS | https://api.github.com/repos/MonetDB/MonetDB/issues/3502/comments | 0 | 2020-11-30T12:51:57Z | 2024-06-27T12:03:33Z | https://github.com/MonetDB/MonetDB/issues/3502 | 753,453,239 | 3,502 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-18 14:08:45 +0200
From: sandy.moens
To: Merovingian devs <<bugs-merovingian>>
Version: 11.21.19 (Jul2015-SP4)
CC: @hannesmuehleisen, renalkoclok, sandy.moens
Last updated: 2020-06-03 09:17:18 +0200
## Comment 19847
Date: 2014-06-18 14:08:45 +0200
From: sandy.moens
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_3) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/35.0.1916.153 Safari/537.36
Build Identifier:
When connecting to a server through ssh, the environment locale is set to LC_CTYPE=UTF-8.
Then after creating a database, when starting the database a crash occurs.
Changing the locale beforehand to LC_CTYPE=en_US.UTF-8 solves this issue.
Reproducible: Always
### Steps to Reproduce:
1.Connect through ssh on a Mac
2.monetdbd start mdb/test
3.monetdbd set port=50050 mdb/test
4.monetdbd start mdb/test
5.monetdb create t
6.monetdb start t
### Actual Results:
starting database 't'... FAILED
start: starting 't' failed: database 't' has crashed after starting, manual intervention needed, check monetdbd's logfile for details
### Expected Results:
starting database 't'... done
## Comment 20007
Date: 2014-08-07 14:40:58 +0200
From: @hannesmuehleisen
What does the logfile actually say?
## Comment 21120
Date: 2015-08-12 09:01:49 +0200
From: @sjoerdmullender
When I try to start the server on Linux (Fedora 22) with LC_CTYPE=UTF-8 in the environment, I get a message:
!FATAL: cannot set locale
(Look in the merovingian.log file for the error message.)
When I do the same on a mac, the server starts up fine.
If this is the same for you, this is hardly a bug. We cannot work in a "bad" environment. It is then better to exit early than to limp along and mess things up.
## Comment 27772
Date: 2020-06-03 09:17:18 +0200
From: @sjoerdmullender
The content of attachment 673 has been deleted for the following reason:
spam
| Starting database fails because of locale settings | https://api.github.com/repos/MonetDB/MonetDB/issues/3501/comments | 0 | 2020-11-30T12:51:55Z | 2024-06-28T13:16:54Z | https://github.com/MonetDB/MonetDB/issues/3501 | 753,453,221 | 3,501 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-18 11:31:43 +0200
From: László Sándor <<sandorl>>
To: clients devs <<bugs-clients>>
Version: 11.17.17 (Jan2014-SP2)
Last updated: 2014-07-31 09:50:27 +0200
## Comment 19846
Date: 2014-06-18 11:31:43 +0200
From: László Sándor <<sandorl>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_3) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/37.0.2055.2 Safari/537.36
Build Identifier:
Stata fixed how they will talk to the MonetDB driver in the future, but they say the MonetDB driver is less standard, so perhaps we should be aware of this:
We [Stata] were passing a NULL string to the ODBC
api call SQLTables and SQLColumns where the MonetDB driver wants an empty string
(""). Some other drivers need a NULL string passed in to work (Oracle being
one) so I believe this is a driver bug, but I have patched the problem in Stata.
I have added code in -odbc- to check for the MonetDB driver and pass an empty
string in that case.
Reproducible: Always
### Steps to Reproduce:
As StataCorp investigated this, see above, I skip these steps.
## Comment 19859
Date: 2014-06-30 11:49:17 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [20d392e6f109](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=20d392e6f109) 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=20d392e6f109](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=20d392e6f109)
Changeset description:
Accept NULL pointer in wide character arguments.
This fixes bug #3500.
## Comment 19860
Date: 2014-06-30 11:50:08 +0200
From: @sjoerdmullender
It turns out, the problem only occurs in the wide character interfaces. This has now been fixed.
| MonetDB driver wants an empty string for SQLTables and SQLColumns API calls, where other drivers expect NULL | https://api.github.com/repos/MonetDB/MonetDB/issues/3500/comments | 0 | 2020-11-30T12:40:31Z | 2024-06-27T12:03:31Z | https://github.com/MonetDB/MonetDB/issues/3500 | 753,445,919 | 3,500 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-18 10:52:53 +0200
From: Pete Hollobon <<pete>>
To: Merovingian devs <<bugs-merovingian>>
Version: 11.21.19 (Jul2015-SP4)
Last updated: 2016-04-11 11:24:37 +0200
## Comment 19845
Date: 2014-06-18 10:52:53 +0200
From: Pete Hollobon <<pete>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/35.0.1916.153 Safari/537.36
Build Identifier:
MonetDB becomes unresponsive and upon restarting merovingian.log contains many "unknown state: 4" entries. This is happening once or twice a day on a ~3GB database.
We're running Jan2014-SP2 (official binaries) on Ubuntu server 14.04.
We had the same database on Jan2014-SP1 / Ubuntu 12.10 for several months and didn't see this issue.
Example:
2014-06-15 23:10:56 MSG example[36117]: MonetDB 5 server v11.17.17 "Jan2014-SP2"
2014-06-15 23:10:56 MSG example[36117]: Serving database 'example', using 16 threads
2014-06-15 23:10:56 MSG example[36117]: Compiled for x86_64-pc-linux-gnu/64bit with 64bit OIDs dynamically linked
2014-06-15 23:10:56 MSG example[36117]: Found 22.987 GiB available main-memory.
2014-06-15 23:10:56 MSG example[36117]: Copyright (c) 1993-July 2008 CWI.
2014-06-15 23:10:56 MSG example[36117]: Copyright (c) August 2008-2014 MonetDB B.V., all rights reserved
2014-06-15 23:10:56 MSG example[36117]: Visit http://www.monetdb.org/ for further information
2014-06-15 23:10:56 MSG example[36117]: Listening for UNIX domain connection requests on mapi:monetdb:///var/monetdb5/dbfarm/prod/example/.mapi.s
ock
2014-06-15 23:10:56 MSG example[36117]: MonetDB/SQL module loaded
2014-06-16 15:11:10 MSG merovingian[36104]: caught SIGTERM, starting shutdown sequence
2014-06-16 15:11:11 MSG control[36104]: control channel closed
2014-06-16 15:11:12 MSG merovingian[36104]: sending process 36117 (database 'example') the TERM signal
2014-06-16 15:11:13 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:13 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:14 ERR merov
ingian[36104]: unknown state: 42014-06-16 15:11:14 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:15 ERR merovingian[36104]: unknown state:
42014-06-16 15:11:16 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:16 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:17 ERR mer
ovingian[36104]: unknown state: 42014-06-16 15:11:17 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:18 ERR merovingian[36104]: unknown stat
e: 42014-06-16 15:11:18 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:19 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:19 ERR m
erovingian[36104]: unknown state: 42014-06-16 15:11:20 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:20 ERR merovingian[36104]: unknown st
ate: 42014-06-16 15:11:21 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:21 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:22 ERR
merovingian[36104]: unknown state: 42014-06-16 15:11:22 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:23 ERR merovingian[36104]: unknown
state: 42014-06-16 15:11:23 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:24 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:24 E
RR merovingian[36104]: unknown state: 42014-06-16 15:11:25 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:25 ERR merovingian[36104]: unknow
n state: 42014-06-16 15:11:26 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:26 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:27
ERR merovingian[36104]: unknown state: 42014-06-16 15:11:27 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:28 ERR merovingian[36104]: unkn
own state: 42014-06-16 15:11:28 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:29 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:
29 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:30 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:30 ERR merovingian[36104]: un
known state: 42014-06-16 15:11:31 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:31 ERR merovingian[36104]: unknown state: 42014-06-16 15:1
1:32 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:32 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:33 ERR merovingian[36104]:
unknown state: 42014-06-16 15:11:33 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:34 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:34 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:35 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:35 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:36 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:36 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:37 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:37 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:38 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:38 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:39 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:39 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:40 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:40 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:41 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:41 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:42 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:42 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:43 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:43 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:44 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:44 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:45 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:45 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:46 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:46 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:47 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:48 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:48 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:49 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:49 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:50 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:50 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:51 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:51 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:52 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:52 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:53 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:53 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:54 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:54 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:55 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:55 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:56 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:56 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:57 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:57 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:58 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:58 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:59 ERR merovingian[36104]: unknown state: 42014-06-16 15:11:59 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:00 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:00 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:01 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:01 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:02 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:02 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:03 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:03 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:04 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:04 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:05 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:05 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:06 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:06 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:07 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:07 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:08 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:08 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:09 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:09 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:10 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:10 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:11 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:11 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:12 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:12 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:13 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:13 ERR merovingian[36104]: unknown state: 42014-06-16 15:12:13 ERR merovingian[36104]: timeout of 60 seconds expired, sending process 36117 (database 'example') the KILL signal
2014-06-16 15:12:13 MSG merovingian[36104]: Merovingian 1.7 stopped
2014-06-16 15:12:13 MSG merovingian[36104]: database 'example' (36117) was killed by signal SIGKILL
Reproducible: Sometimes
| Unresponsive mserver ("unknown state: 4" in logs on restart) | https://api.github.com/repos/MonetDB/MonetDB/issues/3499/comments | 0 | 2020-11-30T12:40:29Z | 2024-06-28T13:16:53Z | https://github.com/MonetDB/MonetDB/issues/3499 | 753,445,895 | 3,499 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-15 02:22:17 +0200
From: George <<georgy.ramonov>>
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.17.17 (Jan2014-SP2)
Last updated: 2014-10-31 14:14:16 +0100
## Comment 19843
Date: 2014-06-15 02:22:17 +0200
From: George <<georgy.ramonov>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/34.0.1847.116 Safari/537.36
Build Identifier:
Suppose we have a table
CREATE TABLE test.simple_table("id" INTEGER);
INSERT INTO test.simple_table VALUES (1),(2),(3),(4),(5);
Let's say I want to run a simple query:
SELECT COUNT(*) AS val
FROM test.simple_table
ORDER BY val DESC;
+-------+
| v |
+=======+
| 5 |
+-------+
Runs fine.
However, if add a LIMIT clause:
SELECT COUNT(*) AS val
FROM test.simple_table
ORDER BY val DESC
LIMIT 5;
I get the following
TypeException:user.s37_7[17]:'algebra.markT' undefined in: _30:any := algebra.markT(_27:oid, _29:oid)
program contains errors
However, if I add a primitive HAVING clause
SELECT COUNT(*) AS val
FROM test.simple_table
HAVING 1 > 0
ORDER BY val DESC
LIMIT 5;
+-------+
| v |
+=======+
| 5 |
+-------+
It runs fine again. What's going on?
------ query execution plans attached below ------
-- without LIMIT
sql>explain select count(*) as v from simple_table order by v;
+------------------------------------------------------------------------------+
| mal |
+==============================================================================+
| function user.s3_7{autoCommit=true}():void; |
| barrier X_36 := language.dataflow(); |
| X_2 := sql.mvc(); |
| X_3:bat[:oid,:oid] := sql.tid(X_2,"medical","simple_table"); |
| X_6 := sql.bind(X_2,"medical","simple_table","id",0); |
| (X_9,r1_9) := sql.bind(X_2,"medical","simple_table","id",2); |
| X_12 := sql.bind(X_2,"medical","simple_table","id",1); |
| X_14 := sql.delta(X_6,X_9,r1_9,X_12); |
| X_15 := algebra.leftfetchjoin(X_3,X_14); |
| X_16 := aggr.count(X_15); |
| exit X_36; |
| sql.exportValue(1,"medical.L1","v","wrd",64,0,6,X_16,""); |
| end s3_7; |
| querylog.define("explain select count(*) as v from simple_table order by v |
: ;","default_pipe")
====================
-- with LIMIT
explain select count(*) as v from simple_table order by v limit 5;
TypeException:user.s4_7[17]:'algebra.markT' undefined in: _30:any := algebra.markT(_27:oid, _29:oid)
+------------------------------------------------------------------------------+
| mal |
+==============================================================================+
| function user.s4_7{autoCommit=true}():void; |
| sql.mvc(); |
| X_3:bat[:oid,:oid] := sql.tid(X_2,"medical","simple_table"); |
| sql.bind(X_2,"medical","simple_table","id",0); |
| (X_9,r1_9) := sql.bind(X_2,"medical","simple_table","id",2); |
| sql.bind(X_2,"medical","simple_table","id",1); |
| sql.delta(X_6,X_9,r1_9,X_12); |
| algebra.leftfetchjoin(X_3,X_14); |
| aggr.count(X_15); |
| calc.wrd(0:wrd); |
| calc.wrd(5:wrd); |
| X_21:bat[:oid,:wrd] := bat.new(nil:oid,nil:wrd); |
| bat.append(X_21,X_16); |
| calc.+(X_17,X_19); |
| pqueue.topn_max(X_21,X_25); |
| algebra.find(X_26,0@0); |
| calc.oid(0@0); |
| algebra.markT(X_27,X_29); |
| bat.reverse(X_30); |
| sql.single(X_16); |
| algebra.leftfetchjoin(X_31,X_32); |
| calc.wrd(0:wrd); |
| calc.wrd(5:wrd); |
| calc.+(X_34,X_35); |
| calc.-(X_36,1); |
| algebra.subslice(X_33,X_34,X_37); |
| algebra.leftfetchjoin(X_38,X_33); |
| sql.resultSet(1,1,X_39); |
| sql.rsColumn(X_40,".","single_value","wrd",64,0,X_39); |
| io.stdout(); |
| sql.exportResult(X_47,X_40); |
| end s4_7; |
| querylog.define("explain select count(*) as v from simple_table order by v |
: limit 5;","default_pipe")
Reproducible: Always
### Steps to Reproduce:
Create a table in a test schema.
CREATE TABLE test.simple_table("id" INTEGER);
INSERT INTO test.simple_table VALUES (1),(2),(3),(4),(5);
Run aggregation with limit
SELECT COUNT(*) AS val
FROM test.simple_table
ORDER BY val DESC
LIMIT 5;
### Actual Results:
TypeException:user.s9_7[17]:'algebra.markT' undefined in: _31:any := algebra.markT(_28:oid, _30:oid)
### Expected Results:
+-------+
| v |
+=======+
| 5 |
+-------+
mserver5 --version
MonetDB 5 server v11.17.17 "Jan2014-SP2" (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 11.6GiB available memory, 6 available cpu cores
Libraries:
libpcre: 8.35 2014-04-04 (compiled with 8.35)
openssl: OpenSSL 1.0.1g 7 Apr 2014 (compiled with OpenSSL 1.0.1g 7 Apr 2014)
libxml2: 2.9.1 (compiled with 2.9.1)
Compiled by: george@arch (x86_64-unknown-linux-gnu)
Compilation: gcc -O3 -fomit-frame-pointer -pipe -D_FORTIFY_SOURCE=2
Linking : /usr/bin/ld -m elf_x86_64
## Comment 19895
Date: 2014-07-09 15:46:36 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [d3a1eddca24b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d3a1eddca24b) 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=d3a1eddca24b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=d3a1eddca24b)
Changeset description:
fixed bug #3498, make sure we have a table (not a single row)
before we do the topn.
fixed bug #3489, distinct with order by, use the proper extend directly for the possible order by columns (which are not part of the output)
## Comment 20151
Date: 2014-09-12 16:09:11 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [1eafbe957e34](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=1eafbe957e34) 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=1eafbe957e34](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=1eafbe957e34)
Changeset description:
Added test for bug #3498.
## Comment 20362
Date: 2014-10-31 14:14:16 +0100
From: @sjoerdmullender
Oct2014 has been released.
| SQL throws TypeException if aggregations and limit statements are both present | https://api.github.com/repos/MonetDB/MonetDB/issues/3498/comments | 0 | 2020-11-30T12:40:26Z | 2024-06-27T12:03:29Z | https://github.com/MonetDB/MonetDB/issues/3498 | 753,445,864 | 3,498 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-13 15:13:42 +0200
From: @bartscheers
To: Merovingian devs <<bugs-merovingian>>
Version: 11.17.17 (Jan2014-SP2)
CC: @hannesmuehleisen
Last updated: 2014-10-31 14:14:10 +0100
## Comment 19842
Date: 2014-06-13 15:13:42 +0200
From: @bartscheers
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:29.0) Gecko/20100101 Firefox/29.0
Build Identifier:
creating, releasing and starting a new database prints a merovingian message in in the merovingian.log. Also for stopping destroying, creating, releasing and starting an existing database.
Everything seems to work fine, but the message assumes it isn't.
Reproducible: Always
### Steps to Reproduce:
1. monetdb create mydb; monetdb release mydb; monetdb start mydb
2.
3.
### Actual Results:
MSG merovingian[26144]: database 'mydb' has crashed after start on 1970-01-01 00:59:59, attempting restart, up min/avg/max: 0s/0s/0s, crash average: 0.00 0.00 0.00 (0-0=0)
### Expected Results:
No crash
Same results for the python monetdb.control module
## Comment 20008
Date: 2014-08-07 14:43:40 +0200
From: @hannesmuehleisen
Confirmed, seeing these messages too.
## Comment 20009
Date: 2014-08-07 15:25:04 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [077f26b22108](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=077f26b22108) 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=077f26b22108](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=077f26b22108)
Changeset description:
Remove writing "bla:" to .gdk_lock file.
It doesn't seem to affect anything, despite the comment.
This fixes bug #3497.
## Comment 20359
Date: 2014-10-31 14:14:10 +0100
From: @sjoerdmullender
Oct2014 has been released.
| monetdb start reports crash in merovingian.log | https://api.github.com/repos/MonetDB/MonetDB/issues/3497/comments | 0 | 2020-11-30T12:40:21Z | 2024-06-27T12:03:28Z | https://github.com/MonetDB/MonetDB/issues/3497 | 753,445,823 | 3,497 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-12 11:42:07 +0200
From: @bartscheers
To: clients devs <<bugs-clients>>
Version: 11.17.17 (Jan2014-SP2)
CC: @skinkie
Last updated: 2015-08-28 13:42:03 +0200
## Comment 19841
Date: 2014-06-12 11:42:07 +0200
From: @bartscheers
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:29.0) Gecko/20100101 Firefox/29.0
Build Identifier:
When two table names start with the same substring, tabbing causes autocompletion starting to substitute schema names for every tab.
Reproducible: Always
### Steps to Reproduce:
1. monetdb create tabnames; monetdb release tabnames; mclient -dtabnames
2. create table tab_first (i int, d double); create table tab_second (i2 int, d2 double);
3. select * from tab<and now press tab>
or try
4. select * from t<and now press tab>
### Actual Results:
After three tab key strokes:
select * from "sys".""sys".""sys"."tab
### Expected Results:
A list of possible names
This is also the case in other than "sys" schema names
## Comment 21121
Date: 2015-08-12 10:04:34 +0200
From: @sjoerdmullender
Fixed in changeset [5026f191f4c3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=5026f191f4c3)
## Comment 21200
Date: 2015-08-28 13:42:03 +0200
From: @sjoerdmullender
Jul2015 has been released.
| autocompletion table names does not work correctly | https://api.github.com/repos/MonetDB/MonetDB/issues/3496/comments | 0 | 2020-11-30T12:40:18Z | 2024-06-27T12:03:27Z | https://github.com/MonetDB/MonetDB/issues/3496 | 753,445,789 | 3,496 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-11 10:46:10 +0200
From: @drstmane
To: @mlkersten
Version: -- development
CC: @njnes
Last updated: 2014-10-31 14:13:39 +0100
## Comment 19839
Date: 2014-06-11 10:46:10 +0200
From: @drstmane
For test sql/test/centipede/Tests/olap.sql no correct/expected/intended output has been provided, yet.
cf.,
http://monetdb.cwi.nl/testweb/web/showtestoutput.php?serial=51696:3cc2c0060646&target=GNU-Fedora-x86_64-assert-propcheck&module=sql&test=sql%2Ftest%2Fcentipede%2Folap&which=out
We need to check what the correct/expected/intended is (supposed to be) and approve that accordingly.
## Comment 19989
Date: 2014-08-06 19:15:29 +0200
From: @njnes
disabled test until centipede matures
## Comment 20331
Date: 2014-10-31 14:13:39 +0100
From: @sjoerdmullender
Oct2014 has been released.
| Test sql/test/centipede/Tests/olap.sql lacks correct/expected/intended output | https://api.github.com/repos/MonetDB/MonetDB/issues/3495/comments | 0 | 2020-11-30T12:40:16Z | 2024-06-27T12:03:26Z | https://github.com/MonetDB/MonetDB/issues/3495 | 753,445,763 | 3,495 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-11 10:44:16 +0200
From: @drstmane
To: MonetDB5 devs <<bugs-monetdb5>>
Version: -- development
CC: @mlkersten
Last updated: 2014-10-31 14:14:19 +0100
## Comment 19838
Date: 2014-06-11 10:44:16 +0200
From: @drstmane
For tests monetdb5/modules/mal/Tests/pqueue[23].mal no correct/expected/intended output has been provided, yet.
cf.,
http://monetdb.cwi.nl/testweb/web/showtestoutput.php?serial=51696:3cc2c0060646&target=GNU-Fedora-x86_64-assert-propcheck&module=monetdb5&test=monetdb5%2Fmodules%2Fmal%2Fpqueue2&which=out
http://monetdb.cwi.nl/testweb/web/showtestoutput.php?serial=51696:3cc2c0060646&target=GNU-Fedora-x86_64-assert-propcheck&module=monetdb5&test=monetdb5%2Fmodules%2Fmal%2Fpqueue3&which=out
We need to check what the correct/expected/intended is (supposed to be) and approve that accordingly.
## Comment 19964
Date: 2014-08-06 15:01:02 +0200
From: @mlkersten
pqueue has been fixed by Sjoerd.
## Comment 20365
Date: 2014-10-31 14:14:19 +0100
From: @sjoerdmullender
Oct2014 has been released.
| Tests monetdb5/modules/mal/Tests/pqueue[23].mal lack correct/expected/intended output | https://api.github.com/repos/MonetDB/MonetDB/issues/3494/comments | 0 | 2020-11-30T12:40:13Z | 2024-06-27T12:03:25Z | https://github.com/MonetDB/MonetDB/issues/3494 | 753,445,734 | 3,494 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-11 10:35:54 +0200
From: @drstmane
To: MonetDB5 devs <<bugs-monetdb5>>
Version: -- development
CC: @mlkersten
Last updated: 2014-10-31 14:13:46 +0100
## Comment 19837
Date: 2014-06-11 10:35:54 +0200
From: @drstmane
Test monetdb5/modules/mal/Tests/pqueue.mal fails since (not that) recent checkins; cf.
http://monetdb.cwi.nl/testweb/web/showtestoutput.php?serial=51696:3cc2c0060646&target=GNU-Fedora-x86_64-assert-propcheck&module=monetdb5&test=monetdb5%2Fmodules%2Fmal%2Fpqueue&which=out
http://monetdb.cwi.nl/testweb/web/testchange.php?test=51696:3cc2c0060646/GNU-Fedora-x86_64-assert-propcheck/monetdb5/mTests/monetdb5%2Fmodules%2Fmal%2Fpqueue
We need to check what is the correct/expected/intended output of the test,
and fix either the related code or the stable test output or both accordingly.
## Comment 19965
Date: 2014-08-06 15:01:27 +0200
From: @mlkersten
Fixed by new pqueue implementation
## Comment 20335
Date: 2014-10-31 14:13:46 +0100
From: @sjoerdmullender
Oct2014 has been released.
| Test monetdb5/modules/mal/Tests/pqueue.mal fails since recent checkins | https://api.github.com/repos/MonetDB/MonetDB/issues/3493/comments | 0 | 2020-11-30T12:40:11Z | 2024-06-27T12:03:24Z | https://github.com/MonetDB/MonetDB/issues/3493 | 753,445,709 | 3,493 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-09 14:43:48 +0200
From: Kostis Kyzirakos <<kostis.kyzirakos>>
To: Kostis Kyzirakos <<kostis.kyzirakos>>
Version: 11.17.17 (Jan2014-SP2)
Last updated: 2014-06-09 15:50:58 +0200
## Comment 19834
Date: 2014-06-09 14:43:48 +0200
From: Kostis Kyzirakos <<kostis.kyzirakos>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:29.0) Gecko/20100101 Firefox/29.0
Build Identifier:
Cannot populate an MBR column from an SQL dump
Reproducible: Always
### Steps to Reproduce:
1. Create and populate a table that stores a geometry and an MBR
2. Create an SQL dump of the database
3. Create a new database
4. Load the SQL dump of the first database to the new one
### Actual Results:
1. Create and populate a table that stores a geometry and an MBR
⚡ mclient test
sql>CREATE TABLE g (geom GEOMETRY, box MBR);
operation successful (39.373ms)
sql>INSERT INTO g VALUES ('POINT(0 0)', NULL);
1 affected row (15.671ms)
sql>SELECT * FROM g;
+---------------------------------------------+----------------------------------------------+
| geom | box |
+=============================================+==============================================+
| POINT (0.0000000000000000 [0000000000000000](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=0000000000000000) | null |
: 0) : :
+---------------------------------------------+----------------------------------------------+
1 tuple (3.177ms)
sql>UPDATE g SET box = MBR (geom);
1 affected row (9.352ms)
sql>SELECT * FROM g;
+---------------------------------------------+----------------------------------------------+
| geom | box |
+=============================================+==============================================+
| POINT (0.0000000000000000 [0000000000000000](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=0000000000000000) | BOX (0.000000 0.000000, 0.000000 0.000000) |
: 0) : :
+---------------------------------------------+----------------------------------------------+
1 tuple (2.280ms)
sql>exit
2. Check the SQL dump of the database
⚡ msqldump test
-- msqldump dump database Mon Jun 9 13:18:18 2014
-- MonetDB v11.17.17 (Jan2014-SP2), 'mapi:monetdb://leninovo:50000/test'
START TRANSACTION;
SET SCHEMA "sys";
CREATE TABLE "sys"."g" (
"geom" GEOMETRY,
"box" MBR
);
COPY 1 RECORDS INTO "sys"."g" FROM stdin USING DELIMITERS '\t','\n','"';
POINT (0.0000000000000000 0.0000000000000000) BOX (0.000000 0.000000, 0.000000 0.000000)
COMMIT;
3. Create a new database
⚡ monetdb create newdb && monetdb release newdb && monetdb start newdb
created database in maintenance mode: newdb
taken database out of maintenance mode: newdb
starting database 'newdb'... done
4. Load the SQL dump of the first database to the new one
⚡ msqldump test | mclient newdb
auto commit mode: off
auto commit mode: off
operation successful
value 'BOX (0.000000 0.000000, 0.000000 0.000000)' from line 1 field 2 not inserted, expecting type mbr
failed to import table
COMMIT: failed
### Expected Results:
⚡ msqldump test | mclient newdb
auto commit mode: on
auto commit mode: on
operation successful
1 affected row
auto commit mode: on
## Comment 19835
Date: 2014-06-09 15:48:59 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [58b413bef30a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=58b413bef30a) made by Kostis Kyzirakos <kostis@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=58b413bef30a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=58b413bef30a)
Changeset description:
fix for bug #3492 i.e., create an MBR from its textual representation
| Cannot load MBR column from SQL dump | https://api.github.com/repos/MonetDB/MonetDB/issues/3492/comments | 0 | 2020-11-30T12:40:08Z | 2024-06-27T12:03:23Z | https://github.com/MonetDB/MonetDB/issues/3492 | 753,445,685 | 3,492 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-06 18:52:47 +0200
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.17.17 (Jan2014-SP2)
CC: @njnes
Last updated: 2014-10-31 14:14:27 +0100
## Comment 19833
Date: 2014-06-06 18:52:47 +0200
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:29.0) Gecko/20100101 Firefox/29.0
Build Identifier:
Executing SQL query:
SELECT 1 AS one WHERE 1 IN (SELECT 1);
kills the mserver5 (Segmentation fault)
Reproducible: Always
### Steps to Reproduce:
1. start mserver5
2. start SQL Frontend (I tested via SQuirreL and mclient), connect to running mserver5
3. Execute SQL query:
SELECT 1 AS one WHERE 1 IN (SELECT 1)
### Actual Results:
bash-4.2$ /ufs/dinther/MonetDB201405/bin/mserver5 -d10
builtin opt gdk_dbpath = /ufs/dinther/MonetDB201405/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
MonetDB 5 server v11.17.15 "Jan2014-SP2"
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://127.0.0.1:50000/
MonetDB/GIS module loaded
MonetDB/JAQL module loaded
MonetDB/SQL module loaded
>Segmentation fault
bash-4.2$
### Expected Results:
no crashed server
and
a result of the query
It also crashes with following SQL queries:
SELECT 1 AS zero WHERE 1 NOT IN (SELECT 1);
and
SELECT 1 AS zero WHERE 1 IN (SELECT 2);
However it does NOT crash with SQL query;
SELECT 1 AS one WHERE 1 = 1 AND 1 IN (SELECT 1)
## Comment 19896
Date: 2014-07-09 15:46:39 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [d207b2b7ece5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d207b2b7ece5) 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=d207b2b7ece5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=d207b2b7ece5)
Changeset description:
fixed bug #3491, ie not need for a crossproduct if both left and right are only expressions, not relations.
## Comment 20010
Date: 2014-08-07 15:35:58 +0200
From: @njnes
was fixed some time ago
## Comment 20011
Date: 2014-08-07 15:37:04 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [4cf5c1650534](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=4cf5c1650534) 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=4cf5c1650534](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=4cf5c1650534)
Changeset description:
added test for bug #3491
## Comment 20371
Date: 2014-10-31 14:14:27 +0100
From: @sjoerdmullender
Oct2014 has been released.
| SQL query kills the mserver5 (Segmentation fault) | https://api.github.com/repos/MonetDB/MonetDB/issues/3491/comments | 0 | 2020-11-30T12:40:06Z | 2024-06-27T12:03:22Z | https://github.com/MonetDB/MonetDB/issues/3491 | 753,445,656 | 3,491 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-06 17:33:30 +0200
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.17.17 (Jan2014-SP2)
CC: @njnes
Last updated: 2014-10-31 14:13:59 +0100
## Comment 19832
Date: 2014-06-06 17:33:30 +0200
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:29.0) Gecko/20100101 Firefox/29.0
Build Identifier:
Executing SQL query:
select * from "dependencies_tables_on_views"
kills the mserver5 (Segmentation fault)
Reproducible: Always
### Steps to Reproduce:
1. start mserver5
2. start SQL Frontend (I tested via SQuirreL and mclient), connect to running mserver5
3. Execute SQL query:
select * from "dependencies_tables_on_views"
### Actual Results:
bash-4.2$ /ufs/dinther/MonetDB201405/bin/mserver5 -d10
builtin opt gdk_dbpath = /ufs/dinther/MonetDB201405/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
MonetDB 5 server v11.17.15 "Jan2014-SP2"
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://127.0.0.1:50000/
MonetDB/GIS module loaded
MonetDB/JAQL module loaded
MonetDB/SQL module loaded
>Bulk operator required for str.stringleft
Bulk operator required for str.stringleft
Segmentation fault
bash-4.2$
### Expected Results:
no crashed server
and
a result of the query
The table "dependencies_tables_on_views" is listed as one of the (internal) system tables:
Execute:
select * from "_tables" where name = 'dependencies_tables_on_views'
this returns 1 row
Execute:
select * from "_columns" where table_id in (select id from "_tables" where name = 'dependencies_tables_on_views')
this returns 3 rows
## Comment 19889
Date: 2014-07-09 13:20:54 +0200
From: @njnes
these tables do not longer exist in the default branch. This type of tables existed for handling table return type, which isn't stored in the _tables table
anymore. So fixed for the to be created feature release branch
## Comment 20348
Date: 2014-10-31 14:13:59 +0100
From: @sjoerdmullender
Oct2014 has been released.
| SQL query kills the mserver5 (Segmentation fault) | https://api.github.com/repos/MonetDB/MonetDB/issues/3490/comments | 0 | 2020-11-30T12:40:03Z | 2024-06-27T12:03:21Z | https://github.com/MonetDB/MonetDB/issues/3490 | 753,445,630 | 3,490 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-06 15:07:44 +0200
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.17.17 (Jan2014-SP2)
CC: @njnes
Last updated: 2014-10-31 14:13:36 +0100
## Comment 19831
Date: 2014-06-06 15:07:44 +0200
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:29.0) Gecko/20100101 Firefox/29.0
Build Identifier:
Execute SQL query:
SELECT DISTINCT type AS "Data type" FROM SYS.ARGS ORDER BY type
The result is NOT sorted on the type name values
Reproducible: Always
### Steps to Reproduce:
1. Start mserver5 (with SQL module loaded)
2. Start SQL frontend program (I used SQuirreL together with the MonetDB JDBC driver)
3. execute SQL query:
SELECT DISTINCT type AS "Data type" FROM SYS.ARGS ORDER BY type
### Actual Results:
Data type
table
polygon
blob
char
linestring
varchar
geometry
mbr
url
decimal
multipolygon
double
month_interval
multilinestring
time
timetz
timestamp
date
real
sec_interval
timestamptz
clob
inet
point
multipoint
bigint
any
oid
int
wrd
tinyint
boolean
smallint
### Expected Results:
Data type
any
bigint
blob
boolean
char
clob
date
decimal
double
geometry
inet
int
linestring
mbr
month_interval
multilinestring
multipoint
multipolygon
oid
point
polygon
real
sec_interval
smallint
table
time
timestamp
timestamptz
timetz
tinyint
url
varchar
wrd
The problem is probably caused the combination of DISTINCT and usage of an alias name.
Following queries execute okay:
SELECT DISTINCT type AS "Data type" FROM SYS.ARGS ORDER BY "Data type"
SELECT DISTINCT type AS "Data type" FROM SYS.ARGS ORDER BY 1
SELECT type AS "Data type" FROM SYS.ARGS ORDER BY type
## Comment 19894
Date: 2014-07-09 15:46:34 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [d3a1eddca24b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d3a1eddca24b) 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=d3a1eddca24b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=d3a1eddca24b)
Changeset description:
fixed bug #3498, make sure we have a table (not a single row)
before we do the topn.
fixed bug #3489, distinct with order by, use the proper extend directly for the possible order by columns (which are not part of the output)
## Comment 19985
Date: 2014-08-06 19:10:06 +0200
From: @njnes
fixed
## Comment 20329
Date: 2014-10-31 14:13:36 +0100
From: @sjoerdmullender
Oct2014 has been released.
| SQL query with ORDER BY does not order its result as requested | https://api.github.com/repos/MonetDB/MonetDB/issues/3489/comments | 0 | 2020-11-30T12:40:01Z | 2024-06-27T12:03:20Z | https://github.com/MonetDB/MonetDB/issues/3489 | 753,445,611 | 3,489 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-06-06 14:38:34 +0200
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.17.17 (Jan2014-SP2)
CC: @njnes
Last updated: 2014-10-31 14:14:57 +0100
## Comment 19830
Date: 2014-06-06 14:38:34 +0200
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:29.0) Gecko/20100101 Firefox/29.0
Build Identifier:
SQL query:
SELECT DISTINCT mod, (SELECT COUNT(*) FROM SYS.FUNCTIONS M WHERE M.mod = F.mod) as count
FROM SYS.FUNCTIONS F
ORDER BY mod
takes a long time (3 secs) to complete.
It appears to execute the scalar subquery for each row in table SYS.FUNCTIONS, so some 1256 times, which probably clarifies the 3 secs needed to complete.
Reproducible: Always
### Steps to Reproduce:
1. Start mserver5 (with SQL module loaded)
2. Start SQL frontend program (I used SQuirreL together with the MonetDB JDBC driver)
3. execute SQL query:
SELECT DISTINCT mod, (SELECT COUNT(*) FROM SYS.FUNCTIONS M WHERE M.mod = F.mod) as count
FROM SYS.FUNCTIONS F
ORDER BY mod
It returns 17 rows but takes about 3 secs te complete.
That is very long.
### Actual Results:
mod count
aggr 97
algebra 2
calc 752
dictionary 1
geom 40
gsl 1
inet 9
mmath 41
mtime 54
pcre 3
sql 86
str 96
txtsim 24
udf 4
url 16
user 27
zorder 3
### Expected Results:
Same result but delivered in 0.04 secs
If the SQL parser would rewrite the correlated scalar subquery into:
SELECT DISTINCT F.mod, M.count
FROM SYS.FUNCTIONS F LEFT OUTER JOIN (SELECT mod, COUNT(*) as count FROM SYS.FUNCTIONS GROUP BY mod) M ON F.mod = M.mod
ORDER BY F.mod
it takes only 0.04 secs
SQL parser should be detect these correlated scalar subqueries (in SELECT-clause and/or WHERE-clause and/or GROUP BY-clause) and rewrite them.
## Comment 19996
Date: 2014-08-07 09:31:47 +0200
From: @njnes
performance issue is fixed
## Comment 20399
Date: 2014-10-31 14:14:57 +0100
From: @sjoerdmullender
Oct2014 has been released.
| Slow SQL execution for correlated scalar subquery | https://api.github.com/repos/MonetDB/MonetDB/issues/3488/comments | 0 | 2020-11-30T12:39:58Z | 2024-06-27T12:03:19Z | https://github.com/MonetDB/MonetDB/issues/3488 | 753,445,587 | 3,488 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-05-25 10:33:50 +0200
From: @drstmane
To: Documentation maintainers <<bugs-docs>>
Version: unspecified
CC: @mlkersten, @yzchang
Last updated: 2014-07-31 09:50:16 +0200
## Comment 19825
Date: 2014-05-25 10:33:50 +0200
From: @drstmane
On the download page of the website (https://www.monetdb.org/Downloads), the "Professional services" link in the "If you need more" of the "Downloads" panel on the left-hand side results point to
https://www.monetdb.org/node/268
instead of
https://www.monetdbsolutions.com/
and thus result in a "Page not found" error.
## Comment 19826
Date: 2014-05-25 10:34:52 +0200
From: @drstmane
The same holds for the "Documentation" section (https://www.monetdb.org/Documentation).
## Comment 19827
Date: 2014-05-25 10:38:27 +0200
From: @drstmane
And there are more dead links, e.g.,
https://www.monetdb.org/Home -> Overview -> Getting started -> Quick start guide -> https://www.monetdb.org/node/276
## Comment 19828
Date: 2014-05-25 10:42:16 +0200
From: @mlkersten
The link validator should be ran: http://validator.w3.org/
## Comment 19829
Date: 2014-05-26 15:59:01 +0200
From: @yzchang
I corrected the two broken links to MonetDB Solutions.
I have no idea where the "Quick start guide" should link to, so it's Martin's.
A quick run of http://validator.w3.org/ doesn't reveal any broken link. I probably overlooked something.
## Comment 19836
Date: 2014-06-11 10:31:33 +0200
From: @drstmane
While the other dead links are fixed, this one is still there:
https://www.monetdb.org/Home -> Overview -> Getting started -> Quick start guide -> https://www.monetdb.org/node/276
Maybe this is supposed to point to
https://www.monetdb.org/Documentation/UserGuide
?
## Comment 19840
Date: 2014-06-11 11:16:30 +0200
From: @mlkersten
Solved, thanks
| dead link to "Professional services" | https://api.github.com/repos/MonetDB/MonetDB/issues/3487/comments | 0 | 2020-11-30T12:39:56Z | 2024-06-27T12:03:18Z | https://github.com/MonetDB/MonetDB/issues/3487 | 753,445,558 | 3,487 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-05-25 00:14:16 +0200
From: @drstmane
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.17.17 (Jan2014-SP2)
Last updated: 2014-05-25 10:28:39 +0200
## Comment 19823
Date: 2014-05-25 00:14:16 +0200
From: @drstmane
After changeset [d980f1006597](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d980f1006597) "Fixing some Coverity issues",
tests monetdb5/optimizer/Tests/qep0[012].mal fail,
i.e., they produce less output than before.
See e.g.,
http://monetdb.cwi.nl/testweb/web/testgrid.php?serial=51531:ed9a718d0661&order=platform,arch,compiler&targets=Cla-Fedora-x86_64-oid32,GNU-Darwin-i386-propcheck,GNU-Darwin-x86_64-oid32,GNU-Fedora-x86_64-dbfarm,GNU-Fedora-x86_64-oid32-assert-propcheck,GNU-Fedora-x86_64-oid32-dbfarm,GNU-FreeBSD-x86_64,GNU-Gentoo-powerpc-assert-dbfarm,GNU-OpenIndiana-x86_64,GNU-Ubuntu-i386-assert-propcheck-dbfarm,Int-Fedora-x86_64-oid32-assert,Int-Fedora-x86_64-oid32-propcheck,Mic-Windows7-i386-installer&module=monetdb5&tstlimit=monetdb5/optimizer
http://monetdb.cwi.nl/testweb/web/showtestoutput.php?serial=51531:ed9a718d0661&target=GNU-Fedora-x86_64-dbfarm&module=monetdb5&test=monetdb5%2Foptimizer%2Fqep00&which=out
http://monetdb.cwi.nl/testweb/web/showtestoutput.php?serial=51531:ed9a718d0661&target=GNU-Fedora-x86_64-dbfarm&module=monetdb5&test=monetdb5%2Foptimizer%2Fqep01&which=out
http://monetdb.cwi.nl/testweb/web/showtestoutput.php?serial=51531:ed9a718d0661&target=GNU-Fedora-x86_64-dbfarm&module=monetdb5&test=monetdb5%2Foptimizer%2Fqep02&which=out
## Comment 19824
Date: 2014-05-25 10:28:39 +0200
From: @drstmane
Fixed by changeset [6f0ebec6449d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=6f0ebec6449d)
Thanks!
| changeset d980f1006597 "Fixing some Coverity issues" makes qep tests fail | https://api.github.com/repos/MonetDB/MonetDB/issues/3486/comments | 0 | 2020-11-30T12:39:54Z | 2024-06-27T12:03:17Z | https://github.com/MonetDB/MonetDB/issues/3486 | 753,445,528 | 3,486 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-05-23 15:43:38 +0200
From: @drstmane
To: Documentation maintainers <<bugs-docs>>
Version: unspecified
Last updated: 2015-07-15 17:06:42 +0200
## Comment 19821
Date: 2014-05-23 15:43:38 +0200
From: @drstmane
Section "Latest binary releases" in left-hand panel under https://www.monetdb.org/Downloads does not yet list the (new) EPEL binaries (yum repository).
## Comment 21007
Date: 2015-07-15 17:06:42 +0200
From: @sjoerdmullender
EPEL packages are mentioned (as RHEL & CentOS).
| Download area does not yet mention new EPEL binaries | https://api.github.com/repos/MonetDB/MonetDB/issues/3485/comments | 0 | 2020-11-30T12:39:51Z | 2024-06-27T12:03:16Z | https://github.com/MonetDB/MonetDB/issues/3485 | 753,445,506 | 3,485 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-05-21 07:53:40 +0200
From: @hannesmuehleisen
To: SQL devs <<bugs-sql>>
Version: 11.17.13 (Jan2014-SP1)
CC: @njnes
Last updated: 2014-10-31 14:14:07 +0100
## Comment 19814
Date: 2014-05-21 07:53:40 +0200
From: @hannesmuehleisen
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_3) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/35.0.1916.114 Safari/537.36
Build Identifier:
I am running the following two commands on the DB:
CREATE TABLE DS (LopNr BIGINT, LopNr_Relationsperson BIGINT);
COPY 1 OFFSET 2 RECORDS INTO DS FROM '[...]/DS.txt' USING DELIMITERS '\t','\r\n','"' NULL AS '' LOCKED;
This works fine on OSX and Linux, but not on Windows. Error given:
"Incomplete record at end of file. Failed to import table."
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE
2. COPY INTO
3. ERROR
### Actual Results:
Error message
### Expected Results:
COPY INTO succeeds
## Comment 19815
Date: 2014-05-21 07:54:13 +0200
From: @hannesmuehleisen
Created attachment 286
File that fails to import
> Attached file: [DS.txt](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3484_DS.txt_286) (text/plain, 41 bytes)
> Description: File that fails to import
## Comment 19816
Date: 2014-05-21 07:55:47 +0200
From: @hannesmuehleisen
File has emtpy line at the end, but the COPY INTO reflects that so it should be ignored.
## Comment 19817
Date: 2014-05-21 13:21:30 +0200
From: @sjoerdmullender
Have you tried using ... USING DELIMITERS '\t','\n','"' ... on Windows?
## Comment 19818
Date: 2014-05-21 17:11:12 +0200
From: @hannesmuehleisen
Sjoerd, good guess, that works! Weird...
## Comment 19819
Date: 2014-05-21 17:21:12 +0200
From: @hannesmuehleisen
Of course, this behavior should be somewhat consistent between platforms.
## Comment 19820
Date: 2014-05-21 17:44:56 +0200
From: @sjoerdmullender
If you're reading a file in TEXT mode on Windows, \r\n in the file are transparently converted to \n as seen by the program. So when you specify \r\n as delimiter, you need an extra \r in the input: one to match the \r in the delimiter, and one to be discarded by the transparent text mode conversion.
In order to make things really consistent between platforms, perhaps we should ignore any \r we encounter, both in the input file and in the delimiter (or perhaps only when the \r occurs just before a \n).
## Comment 19890
Date: 2014-07-09 14:04:04 +0200
From: @njnes
if (on windows) we ignore \r before a \n in de delimiter (when reading
a file in text mode) we should have consistent behaviour.
## Comment 19984
Date: 2014-08-06 19:09:09 +0200
From: @njnes
what is the current status of this bug
## Comment 20150
Date: 2014-09-12 14:56:56 +0200
From: @sjoerdmullender
I'm assuming this is fixed now with changesets [1f4ed128ee53](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=1f4ed128ee53) and [80e6748ab8cc](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=80e6748ab8cc)
## Comment 20355
Date: 2014-10-31 14:14:07 +0100
From: @sjoerdmullender
Oct2014 has been released.
| COPY INTO on a file works fine on Linux/OSX, but not on Windows | https://api.github.com/repos/MonetDB/MonetDB/issues/3484/comments | 0 | 2020-11-30T12:39:48Z | 2024-06-27T12:03:14Z | https://github.com/MonetDB/MonetDB/issues/3484 | 753,445,478 | 3,484 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-05-16 06:31:36 +0200
From: @mlkersten
To: SQL devs <<bugs-sql>>
Version: 11.17.13 (Jan2014-SP1)
Last updated: 2014-05-16 08:38:59 +0200
## Comment 19811
Date: 2014-05-16 06:31:36 +0200
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:29.0) Gecko/20100101 Firefox/29.0
Build Identifier:
A missing ',' between two numbers in the values list lead to a connection failure in mclient.
Reproducible: Always
### Steps to Reproduce:
create table tmp2(i tinyint);
insert into tmp2 values(8),(9),(10),(11)(12);
select * from tmp2;
drop table tmp2;
## Comment 19812
Date: 2014-05-16 08:38:59 +0200
From: @mlkersten
invalid.
| client crash on syntax error in SQL | https://api.github.com/repos/MonetDB/MonetDB/issues/3483/comments | 0 | 2020-11-30T12:39:46Z | 2024-06-28T13:16:52Z | https://github.com/MonetDB/MonetDB/issues/3483 | 753,445,452 | 3,483 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-05-08 07:41:51 +0200
From: @mlkersten
To: SQL devs <<bugs-sql>>
Version: 11.17.13 (Jan2014-SP1)
CC: @njnes
Last updated: 2014-05-22 09:52:26 +0200
## Comment 19805
Date: 2014-05-08 07:41:51 +0200
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:28.0) Gecko/20100101 Firefox/28.0
Build Identifier:
Cross product does not produce the correct result
sql>create table t(i integer);
operation successful (72.464ms)
sql>insert into t values(0),(1);
2 affected rows (9.770ms)
sql>select * from t;
+------+
| i |
+======+
| 0 |
| 1 |
+------+
2 tuples (0.924ms)
sql>select * from t,t;
+------+------+
| i | i |
+======+======+
| 0 | 0 |
| 0 | 0 |
| 1 | 1 |
| 1 | 1 |
+------+------+
4 tuples (1.219ms)
Reproducible: Always
### Steps to Reproduce:
create table t(i integer);
insert into t values(0),(1);
select * from t;
### Actual Results:
+------+------+
| i | i |
+======+======+
| 0 | 0 |
| 0 | 0 |
| 1 | 1 |
| 1 | 1 |
+------+------+
### Expected Results:
+------+------+
| i | i |
+======+======+
| 0 | 0 |
| 0 | 1 |
| 1 | 0 |
| 1 | 1 |
+------+------+
## Comment 19806
Date: 2014-05-09 11:43:12 +0200
From: @mlkersten
This is only caused by a self-product and without using correlation names.
(select * from t as A, t as B) works
## Comment 19809
Date: 2014-05-11 23:14:28 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [ec0a3ea4be34](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ec0a3ea4be34) 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=ec0a3ea4be34](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ec0a3ea4be34)
Changeset description:
fix for bug #3482, ie return an error on abiguous table names
## Comment 19810
Date: 2014-05-11 23:14:41 +0200
From: @njnes
fixed by returning an error
| Crossproduct error | https://api.github.com/repos/MonetDB/MonetDB/issues/3482/comments | 0 | 2020-11-30T12:39:44Z | 2024-06-27T12:03:12Z | https://github.com/MonetDB/MonetDB/issues/3482 | 753,445,426 | 3,482 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-05-01 16:54:58 +0200
From: @hannesmuehleisen
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.17.13 (Jan2014-SP1)
CC: @njnes
Last updated: 2014-05-22 09:52:27 +0200
## Comment 19790
Date: 2014-05-01 16:54:58 +0200
From: @hannesmuehleisen
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/35.0.1916.69 Safari/537.36
Build Identifier:
One of the Java examples (http://dev.monetdb.org/hg/MonetDB/file/fe1fb3ce0c9b/java/example/SQLcopyinto.java) uses COPY INTO FROM stdin over a low-level MAPI socket as an example for bulk load. However, if multiple such COPY INTO statements are used, none are executed. In addition, strange characters appear on the MAPI socket.
Reproducible: Always
### Steps to Reproduce:
See attached Python script
### Actual Results:
No tables created, no content inserted, strange characters appear.
### Expected Results:
Statements are executed.
## Comment 19791
Date: 2014-05-01 16:55:18 +0200
From: @hannesmuehleisen
Created attachment 285
Python script to provoke the issue
> Attached file: [test.py](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3481_test.py_285) (text/x-python, 919 bytes)
> Description: Python script to provoke the issue
## Comment 19792
Date: 2014-05-01 16:57:02 +0200
From: @hannesmuehleisen
MAPI Response looks like this:
&4 f (transaction start)
&3 (create table)
[Strange special characters]
## Comment 19793
Date: 2014-05-05 10:21:33 +0200
From: @njnes
This is because we get an 'SQLException:sql.affectedRows:failed' in the second copy into.
## Comment 19794
Date: 2014-05-05 10:45:35 +0200
From: @njnes
tablet (csv reader) always does a read_more which its not supposed to do, its should only try to read more if it needs it (ie empty (non read) buffer).
## Comment 19798
Date: 2014-05-05 11:16:11 +0200
From: @sjoerdmullender
I fixed this issue in changeset [d633b25eb73e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d633b25eb73e) I don't know why this wasn't recorded in the tracker.
| Cannot run multiple COPY INTO statements in one 's'-command | https://api.github.com/repos/MonetDB/MonetDB/issues/3481/comments | 0 | 2020-11-30T12:39:42Z | 2024-06-27T12:03:11Z | https://github.com/MonetDB/MonetDB/issues/3481 | 753,445,403 | 3,481 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-04-29 19:49:31 +0200
From: andres Ferraro <<aferraro>>
To: SQL devs <<bugs-sql>>
Version: 11.17.13 (Jan2014-SP1)
CC: @njnes
Last updated: 2014-05-05 10:48:06 +0200
## Comment 19786
Date: 2014-04-29 19:49:31 +0200
From: andres Ferraro <<aferraro>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/30.0.1599.66 Safari/537.36
Build Identifier:
When I execute this query:
select sub.* from (select sum(t.onboard_at_departure) as tot, p.number as numb from total t, passengers p where t.trip=p.trip group by p.number) sub order by sub.tot DESC limit 1;
The result is OK, but when I put that on the FROM of other query like this:
select * from (select sub.* from (select sum(t.onboard_at_departure) as tot, p.number as numb from total t, passengers p where t.trip=p.trip group by p.number) sub order by sub.tot DESC limit 1);
I get this error:
syntax error, unexpected ORDER, expecting INTERSECT or EXCEPT or UNION or ')' in: "select * from (select sub.* from (select sum(t.onboard_at_departure) as tot, p.number as numb from total t, passengers p where t.trip=p.trip group by p.number) sub order"
Reproducible: Always
### Steps to Reproduce:
1.execute: select * from (select sub.* from (select sum(t.onboard_at_departure) as tot, p.number as numb from total t, passengers p where t.trip=p.trip group by p.number) sub order by sub.tot DESC limit 1);
2.
3.
### Actual Results:
get this error:
syntax error, unexpected ORDER, expecting INTERSECT or EXCEPT or UNION or ')' in: "select * from (select sub.* from (select sum(t.onboard_at_departure) as tot, p.number as numb from total t, passengers p where t.trip=p.trip group by p.number) sub order"
### Expected Results:
the result must be the same that running this query:
select sub.* from (select sum(t.onboard_at_departure) as tot, p.number as numb from total t, passengers p where t.trip=p.trip group by p.number) sub order by sub.tot DESC limit 1;
The data I'm using is the voc_dump.sql
## Comment 19795
Date: 2014-05-05 10:48:06 +0200
From: @njnes
this is intended. The need for order by in a sub query, doesn't fit a relational model.
| subquery on FROM fails on ORDER BY | https://api.github.com/repos/MonetDB/MonetDB/issues/3480/comments | 0 | 2020-11-30T12:39:39Z | 2024-06-28T13:16:52Z | https://github.com/MonetDB/MonetDB/issues/3480 | 753,445,380 | 3,480 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-04-29 11:45:33 +0200
From: andrei <<andrei.ciobotar>>
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: @njnes
Last updated: 2014-05-05 18:10:38 +0200
## Comment 19785
Date: 2014-04-29 11:45:33 +0200
From: andrei <<andrei.ciobotar>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/33.0.1750.117 Safari/537.36
Build Identifier:
"timestamp (0000-00-00 00:00:00) has incorrect format" in jan2014-sp1
"INSERT INTO: NOT NULL constraint violated for column a" in feb2013-sp6
monetdb doesn't seem to accept timestamps < 1 for the year, month or day fields. Error reporting also seems incomplete/inconsistent and the system seems to interpret these timestamps as NULL values in older versions.
Reproducible: Always
### Steps to Reproduce:
1. sql>create table timestamp_test ("ts" timestamp);
2. sql>insert into timestamp_test values('0000-00-00 00:00:00');
### Actual Results:
timestamp (0000-00-00 00:00:00) has incorrect format
### Expected Results:
inserted the row
mserver5 --version
MonetDB 5 server v11.17.13 "Jan2014-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 23.5GiB available memory, 8 available cpu cores
Libraries:
libpcre: 8.12 2011-01-15 (compiled with 8.12)
openssl: OpenSSL 1.0.1 14 Mar 2012 (compiled with OpenSSL 1.0.1 14 Mar 2012)
libxml2: 2.7.8 (compiled with 2.7.8)
Compiled by: root@dev.monetdb.org (x86_64-pc-linux-gnu)
Compilation: gcc -O3 -fomit-frame-pointer -pipe -fstack-protector --param=ssp-buffer-size=4 -Wformat -Wformat-security -D_FORTIFY_SOURCE=2
Linking : /usr/bin/ld -m elf_x86_64 -Wl,-Bsymbolic-functions -Wl,-z,relro
## Comment 19796
Date: 2014-05-05 10:49:34 +0200
From: @njnes
there was no year 0.
## Comment 19797
Date: 2014-05-05 11:04:34 +0200
From: andrei <<andrei.ciobotar>>
(In reply to comment 1)
> there was no year 0.
Thanks for the reply. We're not trying to reference non-existing dates but this makes migrating from systems with NO_ZERO_IN_DATE = false problematic as '0000-00-00' is frequently used as a dummy date for space and indexing purposes (instead of NULL). I was assuming there was no practical reason for forcing stricly positive integers for year fields, but if it's more trouble than it's worth for a future version, please close this.
## Comment 19799
Date: 2014-05-05 15:44:34 +0200
From: @njnes
I'm not sure what you mean with "for space and indexing purposes", but when migrating to a new platform its good to remove any workarounds for limitations
(performance or other) from your schema.
## Comment 19800
Date: 2014-05-05 16:25:28 +0200
From: andrei <<andrei.ciobotar>>
"MySQL permits you to store a “zero” value of '0000-00-00' as a “dummy date.” This is in some cases more convenient than using NULL values, and uses less data and index space." taken from the mysql reference. Regardless, it's a moot point if this was an intentional design decision in monetdb.
## Comment 19801
Date: 2014-05-05 18:10:38 +0200
From: @njnes
In MonetDB the storage requirement for NULL and not-NULL is the same and yes
the idea is to only support valid 'values' (date or others) and NULL. No other
special cases. The idea is that more special cases require more special treatment etc.
| Timestamp doesn't accept zero values for year/month/day | https://api.github.com/repos/MonetDB/MonetDB/issues/3479/comments | 0 | 2020-11-30T12:39:36Z | 2024-06-28T13:16:51Z | https://github.com/MonetDB/MonetDB/issues/3479 | 753,445,352 | 3,479 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-04-28 21:41:03 +0200
From: Bryan <<monetdb>>
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: @njnes
Last updated: 2014-07-09 14:59:30 +0200
## Comment 19782
Date: 2014-04-28 21:41:03 +0200
From: Bryan <<monetdb>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/34.0.1847.116 Safari/537.36
Build Identifier:
We have had intermittent crashes of the database for the last few months, using All Feb2013 and Jan2014 versions. We have found a dataset that is 100% repeatable and hence are filing this bug.
We load the data using COPY INTO from FILE
If we then attempt to query the table just loaded with a where clause, mclient (and JDBC) terminates with a "connection termination" message and the server crashes. This happens both on Windows and Linux (Ubuntu 12.04.2 LTS)
If we perform the exact same operation but drop all indexes on the loaded table BEFORE executing the COPY INTO from FILE the same query executes correctly without server crash
Reproducible: Always
### Steps to Reproduce:
1. create schema the bug #2. load the attached SQL dump (thebug.sql.gz)
3. insert into fact_bug select * from data; (mimicking our COPY INTO dataset)
4. select count(1) from fact_bug where equipment_status_id = 1;
### Actual Results:
Database server crash
### Expected Results:
successful query results
If you drop the fact_bug table created above in step 2
create it without any indexes
perform step 3, you can query the dataset without crashing the server
## Comment 19783
Date: 2014-04-28 21:42:10 +0200
From: Bryan <<monetdb>>
Created attachment 284
The dataset to reproduce bug
> Attached file: [thebug.sql.gz](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3478_thebug.sql.gz_284) (application/x-gzip, 281435 bytes)
> Description: The dataset to reproduce bug
## Comment 19784
Date: 2014-04-28 21:44:20 +0200
From: Bryan <<monetdb>>
step 1 should be "create schema thebug;"
## Comment 19822
Date: 2014-05-23 22:28:26 +0200
From: Bryan <<monetdb>>
Just tested this in Jan2014SP2 and unfortunately the bug still exists.
## Comment 19891
Date: 2014-07-09 14:59:30 +0200
From: @njnes
i'm unable to reproduce a crash.
| Database crashes after load of data when indexes exist | https://api.github.com/repos/MonetDB/MonetDB/issues/3478/comments | 0 | 2020-11-30T12:39:34Z | 2024-06-28T13:16:50Z | https://github.com/MonetDB/MonetDB/issues/3478 | 753,445,327 | 3,478 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-04-24 21:50:26 +0200
From: Nik Schuiling <<nik>>
To: clients devs <<bugs-clients>>
Version: 11.17.13 (Jan2014-SP1)
Last updated: 2014-05-22 09:52:22 +0200
## Comment 19776
Date: 2014-04-24 21:50:26 +0200
From: Nik Schuiling <<nik>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/34.0.1847.116 Safari/537.36
Build Identifier:
The following code causes the ODBC driver to raise a "unexpected end of input" exception:
using (OdbcCommand command = new OdbcCommand("INSERT INTO TOTAL_COUNTS VALUES (?, ?)", node.OdbcConnection, transaction))
{
command.Parameters.Add("?", OdbcType.NVarChar);
command.Parameters.Add("?", OdbcType.Int);
command.Prepare();
command.Parameters[0].Value = "157428";
command.Parameters[1].Value = 3337;
command.ExecuteNonQuery();
}
Schema of TOTAL_COUNTS table: CREATE TABLE TOTAL_COUNTS (ITEM varchar(7), SUPPORT int);
ODBC driver log is attached.
I've tried different OdbcTypes for parameter 0, but didn't succeed.
Reproducible: Always
### Steps to Reproduce:
Run described code.
Part of ODBC error log:
SQLPrepareW 5452d48
SQLPrepare: "INSERT INTO TOTAL_COUNTS VALUES (?, ?)"
SQLFreeStmt 5452d48 SQL_RESET_PARAMS
SQLSetStmtAttrW 5452d48 SQL_ATTR_QUERY_TIMEOUT 1e
addStmtError 5452d48 HYC00 Optional feature not implemented 0
SQLGetInfoW 544da78 SQL_CONVERT_VARCHAR
SQLBindParameter 5452d48 1 1 SQL_C_WCHAR SQL_VARCHAR 7 0
SQLBindParameter 5452d48 2 1 SQL_C_SLONG SQL_INTEGER 10 0
SQLExecute 5452d48
SQLExecute 5452d48 execute 3 ('157428\000\000\000\000à´',3337)
addStmtError 5452d48 HY000 unexpected end of input
0
SQLGetDiagRecW Stmt 5452d48 1 0
SQLGetDiagRecW: [MonetDB][ODBC Driver 11.17.13]unexpected end of input
SQLGetDiagRecW Stmt 5452d48 2 0
SQLGetDiagRecW Stmt 5452d48 1 1024
SQLGetDiagRecW: [MonetDB][ODBC Driver 11.17.13]unexpected end of input
SQLGetDiagRecW Stmt 5452d48 2 1024
SQLGetInfoW 544da78 SQL_DRIVER_NAME
addDbcError 544da78 01004 String data, right truncated 0
SQLGetConnectAttrW 544da78 SQL_ATTR_CONNECTION_DEAD
## Comment 19777
Date: 2014-04-24 21:50:59 +0200
From: Nik Schuiling <<nik>>
Created attachment 282
Full OBDC driver log
> Attached file: [odbc.log](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3477_odbc.log_282) (application/octet-stream, 4122 bytes)
> Description: Full OBDC driver log
## Comment 19779
Date: 2014-04-25 16:05:23 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [9f4ee3a5da21](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9f4ee3a5da21) 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=9f4ee3a5da21](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=9f4ee3a5da21)
Changeset description:
Stop at null bytes.
This hopefully fixes bug #3477.
## Comment 19802
Date: 2014-05-06 10:14:50 +0200
From: @sjoerdmullender
Nik, can you test the (hopefully) fixed ODBC driver? I haven't been able to reproduce the problem, but I did implement what I think may well be a fix.
You can find ODBC driver installers at our nightly test page:
http://monetdb.cwi.nl/testweb/web/status.php?branch=Jan2014&n=1
Click on the <x> files link near the top and then download the driver(s) you need.
## Comment 19803
Date: 2014-05-06 14:35:12 +0200
From: Nik Schuiling <<nik>>
(In reply to comment 3)
> Nik, can you test the (hopefully) fixed ODBC driver? I haven't been able to
> reproduce the problem, but I did implement what I think may well be a fix.
>
> You can find ODBC driver installers at our nightly test page:
> http://monetdb.cwi.nl/testweb/web/status.php?branch=Jan2014&n=1
>
> Click on the <x> files link near the top and then download the driver(s) you
> need.
Hi Sjoerd, fix confirmed. Thanks :).
| ODBC driver raises "unexpected end of input" for prepared string parameter from .NET application | https://api.github.com/repos/MonetDB/MonetDB/issues/3477/comments | 0 | 2020-11-30T12:39:31Z | 2024-06-27T12:03:07Z | https://github.com/MonetDB/MonetDB/issues/3477 | 753,445,299 | 3,477 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-04-21 23:38:36 +0200
From: Boris Aronshtam <<baronshtam>>
To: SQL devs <<bugs-sql>>
Version: 11.15.15 (Feb2013-SP4)
CC: @njnes, vera.matei
Last updated: 2015-08-28 13:41:40 +0200
## Comment 19765
Date: 2014-04-21 23:38:36 +0200
From: Boris Aronshtam <<baronshtam>>
The following test-case creates user1 and user2. User1 creates a table u1 and grants some access to it. But after user1 revokes access, user2 is still able to access u1.
-- admin
create role user1_role;
create schema user1_schema authorization user1_role;
create user "user1" with password 'password' name 'User1' schema "user1_schema";
create user "user2" with password 'password' name 'User2' schema "pub";
grant user1_role to user1;
grant pub_role to user2;
-- user1
set role user1_role;
create table u1(c1 int, c2 int);
insert into u1 values(1,101),(2,102);
grant SELECT on u1 to user2; -- ALL | INSERT | DELETE | [ UPDATE | SELECT | REFERENCES ] opt_column_list
-- user2
select * from user1_schema.u1;
insert into user1_schema.u1 values(3,103); -- insufficient privileges
-- user1
grant INSERT on u1 to user2;
-- user2
insert into user1_schema.u1 values(3,103); -- OK now
-- user1
revoke SELECT on u1 from user2;
revoke INSERT on u1 from user2;
-- user2
select * from user1_schema.u1; -- BUG: user2 still can select
insert into user1_schema.u1 values(3,103); -- BUG: user2 still can insert
## Comment 20994
Date: 2015-07-13 22:39:32 +0200
From: vera <<vera.matei>>
The revoking takes place only if the the user signs out first. Test has been added to the test suite of the default branch.
## Comment 21078
Date: 2015-08-01 15:18:14 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [30d12a4105a0](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=30d12a4105a0) 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=30d12a4105a0](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=30d12a4105a0)
Changeset description:
Bug fixes
Bug #3759 fixed types of returned bats (ie use timestamps instead of lng) in modules/mal/clients.c
Bug #3769 fixed crash (triggered assert) by properly looking for referenced groupby columns in push aggr down
optimizer in rel_optimizer.c
Bug #3760 fixed correlated subquery handling in rel_select.c, ie merge only distinct expressions.
Bug #3761 fixed by properly adding the zero_or_one aggregation only when needed (ie when the cardinality of
the outer query is a column/table, ie when we expect one value per row for the subquery).
Bug #3364 fixed in sql_privileges.c, ie allow set role when role_id == auth_id (ie default role).
Bug #3365 fixed in sql_privileges.c, the sql_grant_table_privs always added 'all' privileges in one go
in rel_schema.c properly pass the login_id or role_id based on the 'from current_{role/user}'
part of the grant statement
Bug #3476 was already fixed by previous revoke / schema fixes.
in gdk/gdk_select.c fixed problem with anti select and nils (stop before first nil instead of last)
## Comment 21182
Date: 2015-08-28 13:41:40 +0200
From: @sjoerdmullender
Jul2015 has been released.
| Cannot revoke object access | https://api.github.com/repos/MonetDB/MonetDB/issues/3476/comments | 0 | 2020-11-30T12:39:29Z | 2024-06-27T12:03:06Z | https://github.com/MonetDB/MonetDB/issues/3476 | 753,445,274 | 3,476 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-04-18 17:39:52 +0200
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: @mlkersten
Duplicates: #3434
Last updated: 2014-04-23 11:49:44 +0200
## Comment 19761
Date: 2014-04-18 17:39:52 +0200
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:28.0) Gecko/20100101 Firefox/28.0
Build Identifier:
DROP TABLE [IF EXISTS]
is handy for SQL scripting.
Should be considered for inclusion.
Reproducible: Always
### Steps to Reproduce:
1.Current the SQL syntax: DROP TABLE IF EXISTS MyTable is not supported
### Actual Results:
syntax error, unexpected IF in: "drop table if"
See also 3434
## Comment 19762
Date: 2014-04-18 18:14:01 +0200
From: @mlkersten
This feature request is already covered by 3434
*** This bug has been marked as a duplicate of bug #34 ***
## Comment 19766
Date: 2014-04-23 11:49:44 +0200
From: @sjoerdmullender
*** This bug has been marked as a duplicate of bug #3434 ***
| DROP TABLE IF EXISTS syntax support | https://api.github.com/repos/MonetDB/MonetDB/issues/3475/comments | 0 | 2020-11-30T12:39:26Z | 2024-06-28T13:40:18Z | https://github.com/MonetDB/MonetDB/issues/3475 | 753,445,248 | 3,475 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-04-14 17:37:50 +0200
From: @sjoerdmullender
To: SQL devs <<bugs-sql>>
Version: 11.17.13 (Jan2014-SP1)
CC: @mlkersten, @njnes
Last updated: 2014-10-31 14:13:51 +0100
## Comment 19752
Date: 2014-04-14 17:37:50 +0200
From: @sjoerdmullender
The bulk and scalar versions of mkey.rotate_xor_hash differ for types whose ATOMstorage type is different from the type itself and equal to one of the numeric types.
This has dire consequences for SQL since the function is used in the calculation and maintenance of the index column for multi-column primary keys. It is easy to construct a table with duplicate primary keys:
create table x ( a clob, b clob, c timestamp);
alter table x add primary key (a,b,c);
create table y ( a clob, b clob, c timestamp);
insert into y values ('FIAM','HHZ', '2010-04-25T14:00:00.000');
insert into x (select * from y);
-- this query should fail but doesn't:
insert into x values ('FIAM','HHZ', '2010-04-25T14:00:00.000');
## Comment 20249
Date: 2014-10-04 23:27:33 +0200
From: @mlkersten
Added the test duplicate_primary_keys.Bug-3474.sql
## Comment 20265
Date: 2014-10-08 15:01:40 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [362c555dffad](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=362c555dffad) 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=362c555dffad](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=362c555dffad)
Changeset description:
make sure we use the correct atom hash function for external types.
Fixes bug #3474, a cleanup of all the hash functions is still needed
## Comment 20269
Date: 2014-10-08 16:10:27 +0200
From: @njnes
todo - upgrade indices (dump restore or recreate indices/keys which include types such as date, timestamp etc)
## Comment 20291
Date: 2014-10-14 17:37:52 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [56c438d0b5e4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=56c438d0b5e4) 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=56c438d0b5e4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=56c438d0b5e4)
Changeset description:
Fixed mkey module to use the same hash function whichever way it is called.
This should fix bug #3474 properly (except for upgrade issues).
## Comment 20340
Date: 2014-10-31 14:13:51 +0100
From: @sjoerdmullender
Oct2014 has been released.
| bulk and scalar versions of mkey.rotate_xor_hash differ | https://api.github.com/repos/MonetDB/MonetDB/issues/3474/comments | 0 | 2020-11-30T12:39:24Z | 2024-06-27T12:03:04Z | https://github.com/MonetDB/MonetDB/issues/3474 | 753,445,220 | 3,474 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-04-14 16:07:40 +0200
From: Richard Hughes <<richard.monetdb>>
To: SQL devs <<bugs-sql>>
Version: 11.17.13 (Jan2014-SP1)
CC: @njnes
Last updated: 2014-05-22 09:52:24 +0200
## Comment 19750
Date: 2014-04-14 16:07:40 +0200
From: Richard Hughes <<richard.monetdb>>
Created attachment 280
Stack traces of leak paths
I ran mserver5 under libtcmalloc's leak checker and produced the attached output. I was throwing arbitrary read-only queries at it from our front-end and I'm afraid I wasn't keeping track of the complete list; that was stupid of me. I do know that they were all of the form "select ... from (select ... from src where ...) as t(a,b,...) group by ..."
I eyeballed the stack traces and came up with the patch I'm going to attach shortly. Please re-do my analysis - I really have no idea what I'm doing.
> Attached file: [mserver5.15319._main_-end.dot](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3473_mserver5.15319._main_-end.dot_280) (application/msword, 7666 bytes)
> Description: Stack traces of leak paths
## Comment 19751
Date: 2014-04-14 16:08:30 +0200
From: Richard Hughes <<richard.monetdb>>
Created attachment 281
Proposed patch
> Attached file: [monetdb-sql-memory-leaks.diff](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3473_monetdb-sql-memory-leaks.diff_281) (text/plain, 1946 bytes)
> Description: Proposed patch
## Comment 19753
Date: 2014-04-14 22:19:59 +0200
From: @njnes
thanks for the patch. I just applied it to the jan2014 branch.
## Comment 19758
Date: 2014-04-17 23:46:57 +0200
From: Richard Hughes <<richard.monetdb>>
I feel disgustingly ungrateful questioning you like this, but did something go wrong with the merge? I can't see it in hg, either on default or Jan2014.
http://dev.monetdb.org/hg/MonetDB/log/tip/sql/backends/monet5/sql_gencode.c
## Comment 19759
Date: 2014-04-18 10:04:23 +0200
From: @njnes
Your so right, I seem to have missed a hg push. Now its there (on
Jan2014).
| Various memory leaks in SQL compilation | https://api.github.com/repos/MonetDB/MonetDB/issues/3473/comments | 0 | 2020-11-30T12:39:21Z | 2024-06-27T12:03:03Z | https://github.com/MonetDB/MonetDB/issues/3473 | 753,445,196 | 3,473 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-04-13 08:34:12 +0200
From: @skinkie
To: Documentation maintainers <<bugs-docs>>
Version: unspecified
Last updated: 2014-04-13 09:38:31 +0200
## Comment 19745
Date: 2014-04-13 08:34:12 +0200
From: @skinkie
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/32.0.1700.55 Safari/537.36
Build Identifier:
https://www.monetdb.org/Documentation/Manuals/SQLreference/CopyInto
COPY subquery INTO file_name [ [USING] DELIMITERS field_separator [',' record_separator [ ',' string_quote ]]] [ NULL AS null_string ]
1) the best I can do right now is:
copy (select stabielid, bagidadresseerbaarobject, bagidnummeraanduiding, binnenlandshuisletter, binnenlandshuisnummer, huisnummertoevoeging, binnenlandsplaats, binnenlandspostbusnummer, binnenlandspostcode, binnenlandsstraatnaam, buitenlandsland, buitenlandspostcodewoonplaats, buitenlandsregio, buitenlandsstraathuisnummer, indicatiebinnenlands, registratietijdstip from nhr.adres) to '/tmp/neo4j_adres.csv' delimiter ' ' null as '';
2) I would like to be able to type \t for tabs, like I can do with the loading.
3) documentation states 'into' and 'using' this does not seem to work, neither the full delimiter syntax.
Reproducible: Always
MonetDB 5 server v11.17.14 (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 125.9GiB available memory, 16 available cpu cores
Libraries:
libpcre: 8.34 2013-12-15 (compiled with 8.34)
openssl: OpenSSL 1.0.1f 6 Jan 2014 (compiled with OpenSSL 1.0.1f 6 Jan 2014)
libxml2: 2.9.1 (compiled with 2.9.1)
Compiled by: root@localhost (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
...yes I should update my openssl ;)
| currentt syntax for copy into file is different than website states | https://api.github.com/repos/MonetDB/MonetDB/issues/3472/comments | 0 | 2020-11-30T12:39:19Z | 2024-06-28T13:16:50Z | https://github.com/MonetDB/MonetDB/issues/3472 | 753,445,171 | 3,472 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-04-11 17:57:56 +0200
From: Martin van Dinther <<martin.van.dinther>>
To: clients devs <<bugs-clients>>
Version: 11.17.9 (Jan2014)
CC: ccummins, @njnes, @drstmane, @yzchang
Last updated: 2014-10-31 14:13:39 +0100
## Comment 19743
Date: 2014-04-11 17:57:56 +0200
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:28.0) Gecko/20100101 Firefox/28.0
Build Identifier:
it outputs 49 instead of 1
Other numeric values produce outputs between 48 and 57
Negative numeric values produce output 45
Reproducible: Always
### Steps to Reproduce:
1. execute SQL query: SELECT 1 AS Expected_1;
2.
3.
### Actual Results:
49
### Expected Results:
1
Query: SELECT 1 AS wrong1, 2 AS wrong2, 3 AS wrong3, 4 AS wrong4, 9 AS wrong9, 10 AS wrong10, 99 AS wrong99, 0 AS wrong0, -1 AS wrong_1;
produces output:
wrong1 wrong2 wrong3 wrong4 wrong9 wrong10 wrong99 wrong0 wrong_1
49 50 51 52 57 49 57 48 45
Query: SELECT -1 AS wrong1, -2 AS wrong2, -3 AS wrong3, -4 AS wrong4, -9 AS wrong9, -10 AS wrong10, -99 AS wrong99, -0 AS wrong0;
produces output:
wrong1 wrong2 wrong3 wrong4 wrong9 wrong10 wrong99 wrong0
45 45 45 45 45 45 45 48
So the output numbers increase (to 57) and decrease (to 45).
However query:
SELECT 0 + 1 AS one;
produces correct output: 1
## Comment 19744
Date: 2014-04-11 20:03:55 +0200
From: @njnes
I'm unable to repeat this bug. Anymore details on how to reproduce?
## Comment 19746
Date: 2014-04-13 12:54:11 +0200
From: @yzchang
Just FYI, I can't reproduce it neither (MonetDB v11.17.14).
## Comment 19747
Date: 2014-04-13 13:51:25 +0200
From: @drstmane
Same here, i.e., all works fine with Jan2014-SP1 on my 64-bit Fedora 20 laptop:
$ mclient
Welcome to mclient, the MonetDB/SQL interactive terminal (Jan2014-SP1)
Database: MonetDB v11.17.13 (Jan2014-SP1), 'demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>SELECT 1 AS wrong1, 2 AS wrong2, 3 AS wrong3, 4 AS wrong4, 9 AS wrong9, 10 AS wrong10, 99 AS wrong99, 0 AS wrong0, -1 AS wrong_1;
+--------+--------+--------+--------+--------+---------+---------+--------+---------+
| wrong1 | wrong2 | wrong3 | wrong4 | wrong9 | wrong10 | wrong99 | wrong0 | wrong_1 |
+========+========+========+========+========+=========+=========+========+=========+
| 1 | 2 | 3 | 4 | 9 | 10 | 99 | 0 | -1 |
+--------+--------+--------+--------+--------+---------+---------+--------+---------+
1 tuple (0.627ms)
sql>SELECT -1 AS wrong1, -2 AS wrong2, -3 AS wrong3, -4 AS wrong4, -9 AS wrong9, -10 AS wrong10, -99 AS wrong99, -0 AS wrong0;
+--------+--------+--------+--------+--------+---------+---------+--------+
| wrong1 | wrong2 | wrong3 | wrong4 | wrong9 | wrong10 | wrong99 | wrong0 |
+========+========+========+========+========+=========+=========+========+
| -1 | -2 | -3 | -4 | -9 | -10 | -99 | 0 |
+--------+--------+--------+--------+--------+---------+---------+--------+
1 tuple (1.472ms)
sql>SELECT 1 AS Expected_1;
+------------+
| expected_1 |
+============+
| 1 |
+------------+
1 tuple (0.469ms)
sql>SELECT 0 + 1 AS one;
+------+
| one |
+======+
| 1 |
+------+
1 tuple (39.836ms)
sql>
## Comment 19748
Date: 2014-04-13 15:07:41 +0200
From: @yzchang
I saw Martin van Dinther's GUI client (SQuirreL?) displaying wrong results like he reported.
Corrupted database or bug in de client?
@Martin van Dinther: would you please check if your monetdb server is giving wrong results by execute your queries using e.g. mclient?
## Comment 19749
Date: 2014-04-13 15:30:26 +0200
From: @drstmane
Aha --- the original bug report did not mention Squirrel ... ;-)
well, then it seems like that for some reason Squirrel shows the ASCII codes of the single-character numbers (character '1' has ASCII code 49) than the numbers themselves ...
## Comment 19760
Date: 2014-04-18 17:24:54 +0200
From: Martin van Dinther <<martin.van.dinther>>
1. start mserver5
2. start SQuirreL connecting to demo database using JDBC driver from monetdb-jdbc-2.10.jar
3. execute SQL query: SELECT 1
this returns value: 49
with column name: single_value
and column data type: tinyint
4. the same result and data type when executing SQL: SELECT 1 / 1
5. however when you execute SQL: SELECT 1 + 0
this returns value: 1
with column name: sql_add_single_value
and column data type: smallint
So there is a difference is in the column data type meta data.
I tested this also via:
CREATE TABLE test_tinyint (val tinyint NOT NULL);
INSERT INTO test_tinyint values (1);
INSERT INTO test_tinyint values (2);
INSERT INTO test_tinyint values (9);
INSERT INTO test_tinyint values (10);
INSERT INTO test_tinyint values (0);
INSERT INTO test_tinyint values (-1);
SELECT * from test_tinyint
-- this returns follonwing output in SQuirreL
val
49
50
57
49
48
45
I checked java/src/nl/cwi/monetdb/jdbc/MonetResultSet.java for method:
public short getShort(int columnIndex) but couldn't see any issue.
There is no public short getTiny(int columnIndex) method in JDBC API, so I guess SQuirreL is using getShort() or getInt() methods for this column.
I will have to write a small Java program (testing getShort(), getInt() and getString()) to find out where it goes wrong in the fetching method. Will do that later.
## Comment 19764
Date: 2014-04-18 19:01:42 +0200
From: Martin van Dinther <<martin.van.dinther>>
Have written a small Java program testing getShort(), getInt() and getString() with the test_tinyint table. No issues found for these methods.
However using the method getByte() produced the same results as SQuirreL.
I conclude that SQuirreL is probably using getByte() method when the data type of the column is tinyint as both represent a 1 byte value.
We have currently implemented in MonetResultSet.java
public byte getByte(int columnIndex) throws SQLException {
String bytes = getString(columnIndex);
if (bytes == null || bytes.length() == 0) {
return (byte)0;
} else {
return bytes.getBytes()[0];
}
}
As we transfer data as Strings, the getByte() takes the String value and if not empty converts it to bytes array and returns the first byte.
This goes wrong for all numeric data values.
Maybe we should improve this code by detecting what the data type of result column is (see types[]) and when it is an integer (tinyint, short, int, bigint) or decimal (incl. real, float, double) convert the String first to an integer (or double) and next to a byte value (if it matches int value between -128 and +127).
## Comment 19807
Date: 2014-05-09 19:11:49 +0200
From: Martin van Dinther <<martin.van.dinther>>
I have created a fix for it and tested it succesfully.
Now all columns with resulttype tinyint return the expected numeric value in SQuirreL.
Tried to check in the fix into branch Jan2014, but got:
abort: push creates new remote head fe586d5e7439 on branch 'Jan2014'!
## Comment 19808
Date: 2014-05-09 21:14:41 +0200
From: @drstmane
This it not the complete Mercurial error message.
It continues as follows:
"(pull and merge or see "hg help push" for details about pushing new heads)"
Please do NOT considere the second option, i.e., forcing push to push new heads!
The reason for the error message is that apparently there are new remote changes in the branch that you committed to locally and now want to push to.
To resolve the situation, you first need to pull the new remote changes.
To prevent the need to merge these changes after a pull, please use
hg pull --rebase
This command will rebase your local commits on top of the incoming changesets. This means that you don't need to create a separate changeset to merge your changes with the incoming changes.
To enable the rebase extensions, add the following to your configuration file:
[extensions]
rebase =
In case you did already pull the new remote changes without "--rebase", you will have to merge the two heads (the one you created locally, and the one created remotely) using
hg merge
For general and very useful information and guidelines about (using) Mercurial, please consult https://www.monetdb.org/Assets/MonetDB-Mercurial.html
among others section "Pull and Update".
In fact, the entire document is considered "mandatory literature" for any/all MonetDB developer(s); this includes studying the document and following advice given!
;-)
## Comment 19813
Date: 2014-05-16 16:09:49 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [2ccadb9bd115](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=2ccadb9bd115) 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=2ccadb9bd115](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=2ccadb9bd115)
Changeset description:
fix for bug #3471
## Comment 19862
Date: 2014-06-30 14:20:08 +0200
From: @sjoerdmullender
*** Bug #3503 has been marked as a duplicate of this bug. ***
## Comment 20332
Date: 2014-10-31 14:13:39 +0100
From: @sjoerdmullender
Oct2014 has been released.
| JDBC driver: incorrect output result of SQL query: SELECT 1 ; | https://api.github.com/repos/MonetDB/MonetDB/issues/3471/comments | 0 | 2020-11-30T12:39:16Z | 2024-06-27T12:03:00Z | https://github.com/MonetDB/MonetDB/issues/3471 | 753,445,138 | 3,471 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-04-11 10:32:07 +0200
From: @hannesmuehleisen
To: clients devs <<bugs-clients>>
Version: -- development
CC: martin.van.dinther
Last updated: 2018-05-31 18:39:28 +0200
## Comment 19740
Date: 2014-04-11 10:32:07 +0200
From: @hannesmuehleisen
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/34.0.1847.116 Safari/537.36
Build Identifier:
setClob() currently requires a length if a stream reader is passed. Since the stream reader returns -1 if the end of stream is reached, we can simply read until that happens.
Reproducible: Always
### Steps to Reproduce:
PreparedStatement stmt = con.prepareStatement("insert into clobtest values(?);");
stmt.setClob(1, new StringReader("ABCDEDFG"));
stmt .execute();
### Actual Results:
SQLFeatureNotSupportedException
### Expected Results:
Parameter is set
## Comment 19741
Date: 2014-04-11 11:45:48 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [155ec42a591b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=155ec42a591b) 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=155ec42a591b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=155ec42a591b)
Changeset description:
JDBC Connector: setClob() fixes for Bug #3470
## Comment 19742
Date: 2014-04-11 11:46:19 +0200
From: @hannesmuehleisen
Found an issue in the other setClob() method while I was at it, it did not work at all.
## Comment 26056
Date: 2018-01-04 16:37:49 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset 361fc7902a94, made by Martin van Dinther <martin.van.dinther@monetdbsolutions.com> in the monetdb-java repo, refers to this bug.
For complete details, see https://dev.monetdb.org/hg/monetdb-java?cmd=changeset;node=361fc7902a94
Changeset description:
Extend PreparedStatement test with calling setClob(int, Clob) and setClob(int, Reader) methods.
This completes the fix for bug #3470
## Comment 26471
Date: 2018-05-31 18:39:28 +0200
From: Martin van Dinther <<martin.van.dinther>>
A new version (2.28) of the MonetDB JDBC driver is released, see
https://www.monetdb.org/downloads/Java/
which includes the fix.
| Support setClob without length restrictions | https://api.github.com/repos/MonetDB/MonetDB/issues/3470/comments | 0 | 2020-11-30T12:39:13Z | 2024-06-27T12:02:59Z | https://github.com/MonetDB/MonetDB/issues/3470 | 753,445,117 | 3,470 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-04-09 18:00:08 +0200
From: Edward Clarkson <<edward.clarkson>>
To: clients devs <<bugs-clients>>
Version: 11.17.13 (Jan2014-SP1)
Last updated: 2014-05-22 09:52:34 +0200
## Comment 19739
Date: 2014-04-09 18:00:08 +0200
From: Edward Clarkson <<edward.clarkson>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_2) AppleWebKit/537.75.14 (KHTML, like Gecko) Version/7.0.3 Safari/537.75.14
Build Identifier:
[I selected ODBC, but I know it's present in mclient and wouldn't be surprised if it were on all client interfaces]
Windows network paths look can look like \\network-share\path\to\file, but this format is rejected as not an absolute path by the client, e.g.,:
[MonetDB][ODBC Driver 11.17,13] COPY INTO: filename must have absolute path: \\vmware-host\Documents\Datasets\baseballdata.csv
You can work around this by mapping the path to a drive letter, but that seems unnecessary, and on systems the user doesn't have the right admin rights, not possible.
Reproducible: Always
### Steps to Reproduce:
1. COPY INTO dummy FROM '\\dummyserver\test';
### Actual Results:
Path rejected prior to SQL parsing as an invalid absolute path.
### Expected Results:
SQL parsed (above statement would be rejected b/c the table or file doesn't exist).
## Comment 19768
Date: 2014-04-24 15:10:14 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [ae468c8d6e55](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ae468c8d6e55) 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=ae468c8d6e55](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ae468c8d6e55)
Changeset description:
Accept UNC paths as absolute path in Windows.
This fixes bug #3469.
## Comment 19787
Date: 2014-04-30 10:26:07 +0200
From: @sjoerdmullender
I believe this is fixed now.
| Absolute network paths considered invalid for COPY INTO ... FROM statement. | https://api.github.com/repos/MonetDB/MonetDB/issues/3469/comments | 0 | 2020-11-30T12:39:11Z | 2024-06-27T12:02:58Z | https://github.com/MonetDB/MonetDB/issues/3469 | 753,445,097 | 3,469 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-04-08 17:11:07 +0200
From: a.mckinley
To: SQL devs <<bugs-sql>>
Version: 11.17.13 (Jan2014-SP1)
CC: @njnes
Last updated: 2014-05-22 09:52:30 +0200
## Comment 19734
Date: 2014-04-08 17:11:07 +0200
From: a.mckinley
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/33.0.1750.117 Safari/537.36
Build Identifier:
If you create a local temporary table in a user session, close the session and then open a new one, it is not possible to create a table of the same name.
Reproducible: Always
### Steps to Reproduce:
Execute the following python script twice in a row to reproduce.
import monetdb.sql
import sys
dbh = monetdb.sql.Connection(port=int(sys.argv[1]),database=sys.argv[2],hostname=sys.argv[3],autocommit=True)
cursor = dbh.cursor()
cursor.execute('create local temporary table input (term int, p float) on commit preserve rows;');
cursor.execute('insert into input values ( 1,0.1);');
dbh.close()
### Actual Results:
monetdb.exceptions.OperationalError: 42S01!CREATE TABLE: name 'input' already in use
### Expected Results:
The local temporary table name should be re-usable in a new session.
## Comment 19735
Date: 2014-04-09 12:49:22 +0200
From: @njnes
bug was fixed. At transaction destroy (called at the end of a session) we now properly cleanup all local temporary tables
| Local temporary table persists across sessions | https://api.github.com/repos/MonetDB/MonetDB/issues/3468/comments | 0 | 2020-11-30T12:39:08Z | 2024-06-27T12:02:57Z | https://github.com/MonetDB/MonetDB/issues/3468 | 753,445,073 | 3,468 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-04-08 16:22:25 +0200
From: Edward Clarkson <<edward.clarkson>>
To: clients devs <<bugs-clients>>
Version: 11.17.13 (Jan2014-SP1)
CC: @njnes
Last updated: 2015-01-29 14:07:32 +0100
## Comment 19733
Date: 2014-04-08 16:22:25 +0200
From: Edward Clarkson <<edward.clarkson>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_2) AppleWebKit/537.75.14 (KHTML, like Gecko) Version/7.0.3 Safari/537.75.14
Build Identifier:
See the following query sequence:
sql>CREATE TABLE test (a int);
operation successful (5.230ms)
sql>SELECT a AS "" FROM test;
+--+
| |
+==+
+--+
0 tuples (3.320ms)
In contrast:
sql>SELECT a AS "@" FROM test;
+---+
| @ |
+===+
+---+
0 tuples (1.672ms)
This is the simplest replication: this is more severe/inconvenient when selecting multiple columns: ALL fields after one with a '' are then blanked). So even if I'm missing something w/r/t escaping ''--which then should be documented somewhere--it's still not right to blank other field names.
Note that all the *results* are fine, it's just the field names that are bad--which causes a lot of problems in client code that grabs result set data by field name rather than position.
Reproducible: Always
### Steps to Reproduce:
1. Start the MonetDB server.
2. CREATE TABLE test (a int);
3. SELECT a AS "" FROM test;
### Actual Results:
An empty result set with an empty field name.
### Expected Results:
An empty result set with a field named "".
## Comment 19738
Date: 2014-04-09 13:18:12 +0200
From: @njnes
The '' is part of the mapi protocol. This means we should disallow this within names or somehow escape it. Tested with python also gives a similar problem.
## Comment 20555
Date: 2015-01-14 14:39:51 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [6e51939fcef0](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=6e51939fcef0) 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=6e51939fcef0](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=6e51939fcef0)
Changeset description:
Escape in column name.
This fixes bug #3467.
## Comment 20599
Date: 2015-01-29 14:07:32 +0100
From: @sjoerdmullender
Oct2014-SP2 has been released.
| Field aliases with '#' character excise field names in result set. | https://api.github.com/repos/MonetDB/MonetDB/issues/3467/comments | 0 | 2020-11-30T12:39:06Z | 2024-06-27T12:02:57Z | https://github.com/MonetDB/MonetDB/issues/3467 | 753,445,054 | 3,467 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-04-07 11:41:35 +0200
From: Christian Braun <<hcb>>
To: GDK devs <<bugs-common>>
Version: 11.17.13 (Jan2014-SP1)
CC: @njnes, @swingbit, shuren1314
Last updated: 2015-05-07 12:38:07 +0200
## Comment 19732
Date: 2014-04-07 11:41:35 +0200
From: Christian Braun <<hcb>>
User-Agent: Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:28.0) Gecko/20100101 Firefox/28.0
Build Identifier:
Part of a daily ETL routine i run a lot of statements like:
UPDATE x.t1 SET a = (SELECT a FROM x.t2 as t2 where t1.id = t2.id) where a IS NULL;
One UPDATE just failed with:
ERROR = !GDKerror:!ERROR: HEAPextend: failed to extend to 4521984 for 01/77/54/1775440.tail: MT_mremap() failed
On a later rerun the statement succeeded.
From the log file:
2014-04-06 05:29:23 MSG merovingian[9061]: proxying client (local) for database 'db' to mapi:monetdb:///var/monetdb/dbfarm/db/.mapi.sock?database=db
2014-04-06 05:29:23 MSG merovingian[9061]: target connection is on local UNIX domain socket, passing on filedescriptor instead of proxying
2014-04-06 05:29:27 ERR db[9073]: = gdk_posix.c:558: MT_mremap(bat/01/77/54/1775440.tail,7b1330081000,655360,4521984): p == MAP_FAILED
2014-04-06 05:29:29 MSG merovingian[9061]: proxying client (local) for database 'db' to mapi:monetdb:///var/monetdb/dbfarm/db/.mapi.sock?database=db
2014-04-06 05:29:29 MSG merovingian[9061]: target connection is on local UNIX domain socket, passing on filedescriptor instead of proxying
2014-04-06 05:29:29 MSG merovingian[9061]: proxying client (local) for database 'db' to mapi:monetdb:///var/monetdb/dbfarm/db/.mapi.sock?database=db
2014-04-06 05:29:29 MSG merovingian[9061]: target connection is on local UNIX domain socket, passing on filedescriptor instead of proxying
2014-04-06 05:29:29 MSG db[9073]: GDKmmap(3866624) fails, try to free up space [memory in use=1760358320,virtual memory in use=18446738982644412336]
2014-04-06 05:29:29 MSG db[9073]: GDKmmap(3866624) result [mem=1516209584,vm=18446733116116534704]
2014-04-06 05:29:29 MSG db[9073]: GDKmmap: recovery ok. Continuing..
2014-04-06 05:29:29 MSG merovingian[9061]: proxying client (local) for database 'db' to mapi:monetdb:///var/monetdb/dbfarm/db/.mapi.sock?database=db
2014-04-06 05:29:29 MSG merovingian[9061]: target connection is on local UNIX domain socket, passing on filedescriptor instead of proxying
Reproducible: Didn't try
## Comment 19754
Date: 2014-04-16 08:59:36 +0200
From: @njnes
remap failing can be caused by running out of memory and/or out of disk space.
Is this somehow repeatable using a small(er) example?
## Comment 19757
Date: 2014-04-16 13:19:48 +0200
From: Christian Braun <<hcb>>
So far the error did not repeat. There was/is plenty of disk space available. The machine has 256GB memory. Both tables used in the update statement are not that big. About 15 million rows each. Columns are all bigint. Running out of memory seems unlikely to me.
## Comment 20115
Date: 2014-09-01 18:07:45 +0200
From: @swingbit
I get the same problem on a CREATE statement, using the most recent code (v11.19.0 - Oct2014).
I exclude that memory / disk space isn't enough for this data/query combination.
NB: the same works fine on Feb2013.
Initial data dump is about 3MB.
During the query (a rather normal aggregate), two BATextend to the size of 510GB are attempted, which is not normal.
A 1-file test to reproduce the error is available here:
https://drive.google.com/file/d/0BwkuZ5tb4WEnT21wLVFWRndzcHM
## Comment 20116
Date: 2014-09-01 18:08:55 +0200
From: @swingbit
For completeness, here the output of my test script:
$ mclient -p52500 -d x < bug3466.sql
DROP TABLE: no such table 'term_obj'
auto commit mode: off
operation successful
164280 affected row
auto commit mode: off
DROP TABLE: no such table 'obj_region'
auto commit mode: off
operation successful
29219 affected row
auto commit mode: on
GDK reported error.
HEAPextend: failed to extend to 548010065920 for 06/642.tail: MT_mremap() failed
HEAPextend: failed to extend to 548010065920 for 05/540.tail: MT_mremap() failed
## Comment 20117
Date: 2014-09-01 18:47:57 +0200
From: @swingbit
And the relevant bit from the log:
2014-09-01 18:12:24 ERR x[32451]: = /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/gdk/gdk_posix.c:428: MT_mremap(./bat/06/604.tail,7f2d46390000,29818880,548010065920): GDKextendf() failed
2014-09-01 18:12:24 ERR x[32451]: = /opt/spinque/MonetDBServer/MonetDB.Spinque_Oct2014/src/gdk/gdk_posix.c:428: MT_mremap(./bat/05/501.tail,7f2d44720000,29818880,548010065920): GDKextendf() failed
2014-09-01 18:12:50 ERR x[32451]: !ERROR: bm_subcommit: commit failed
2014-09-01 18:12:50 ERR x[32451]: !ERROR: logger_exit: logger_commit failed
2014-09-01 18:12:50 ERR x[32451]: !FATAL: write-ahead logging failure, disk full?
2014-09-01 18:12:51 MSG x[32451]: !ERROR: GDKsave: error -1 on: name=01/105, ext=tail, mode=0
2014-09-01 18:12:51 MSG x[32451]: !OS: No space left on device
## Comment 20118
Date: 2014-09-01 22:38:22 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [d4c521efe6b9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d4c521efe6b9) 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=d4c521efe6b9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=d4c521efe6b9)
Changeset description:
Better size extrapolation for result bat in mergejoin.
This hopefully fixes bug #3466. It certainly makes that Roberto's
example runs on my system (where it didn't in the original version).
## Comment 20119
Date: 2014-09-02 09:33:43 +0200
From: @swingbit
I confirm that I no longer get this problem (updated to changeset b3341f69aab6)
## Comment 20120
Date: 2014-09-02 10:28:16 +0200
From: shuren <<shuren1314>>
I get this error:
" GDKerror:!ERROR: HEAPextend: failed to extend to 4397727744 for 42\4226.theap: MT_mremap() failed"
when I use the batch import command(COPY INTO…).
I import 200 thousand/2 million/20 million rows every batch.It will give the above error when the data of this table exceeds 300 million rows（maybe 280 million rows）.
My PC have 8G memory and enough disk space.
Data size of a record is less than 30B.
## Comment 20121
Date: 2014-09-02 11:06:28 +0200
From: shuren <<shuren1314>>
(In reply to comment 7)
> I confirm that I no longer get this problem (updated to changeset
> b3341f69aab6)
I am a freshman to use monetdb.
I get this error in my first apply of monetdb :
" GDKerror:!ERROR: HEAPextend: failed to extend to 4397727744 for 42\4226.theap: MT_mremap() failed"
when I use the batch import command(COPY INTO…).
I import 200 thousand/2 million/20 million rows every batch.It will give the above error when the data of this table exceeds 300 million rows（maybe 280 million rows）.
My PC have 8G memory and enough disk space.
Data size of a record is less than 30B.
expect your reply.
## Comment 20720
Date: 2015-03-18 16:38:30 +0100
From: @sjoerdmullender
I'm closing this report as fixed since it seems that my fix did indeed fix the original problem.
As to the problem described in comment 8 and 9, I would say that is a different problem and should get a different bug report.
However, first make sure what version of MonetDB you are running. In particular, are you running a 64 bit version? If not, then that's your problem right there: a 32 bit version cannot cope with large (more than a few GB) databases.
| UPDATE statements fails with "GDKerror: MT_mremap() failed" | https://api.github.com/repos/MonetDB/MonetDB/issues/3466/comments | 0 | 2020-11-30T12:39:03Z | 2024-06-27T12:02:56Z | https://github.com/MonetDB/MonetDB/issues/3466 | 753,445,010 | 3,466 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-04-04 17:41:26 +0200
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: -- development
Last updated: 2017-07-17 16:07:20 +0200
## Comment 19730
Date: 2014-04-04 17:41:26 +0200
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:28.0) Gecko/20100101 Firefox/28.0
Build Identifier:
An ORDER BY clause in the creation of a VIEW is currently not supported.
It is very desirable to have this supported, as it allows you to predefine the ordering per view. This allows it to be used for meaningful (ordered) reporting.
Reproducible: Always
### Steps to Reproduce:
CREATE VIEW user_tables AS
SELECT * FROM sys.tables WHERE NOT system AND query IS NULL
ORDER BY name;
### Actual Results:
Error: CREATE VIEW: ORDER BY not supported
SQLState: 42000
### Expected Results:
The view to be created and the ordering applied during querying.
## Comment 25079
Date: 2017-03-02 15:00:55 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [2bac5d524bcb](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=2bac5d524bcb) 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=2bac5d524bcb](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=2bac5d524bcb)
Changeset description:
Remove the restriction to use an ORDER BY clause in a CREATE VIEW statement.
Added tests to prove this works correctly.
This implements request: bug #3465
| Request: add support for CREATE VIEW with ORDER BY clause | https://api.github.com/repos/MonetDB/MonetDB/issues/3465/comments | 0 | 2020-11-30T12:39:00Z | 2024-06-27T12:02:55Z | https://github.com/MonetDB/MonetDB/issues/3465 | 753,444,984 | 3,465 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-04-03 18:17:42 +0200
From: Qing Guo <<qguo.29>>
To: clients devs <<bugs-clients>>
Version: 11.17.13 (Jan2014-SP1)
Last updated: 2014-04-30 10:27:05 +0200
## Comment 19727
Date: 2014-04-03 18:17:42 +0200
From: Qing Guo <<qguo.29>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/33.0.1750.152 Safari/537.36
Build Identifier:
When compiling MonetDB with "--disable-shared", runing monetdb and mclient will always crush due to an error: "fail to open a communication channel".
The error will not happen when not using "--disable-shared" at compilation.
I tested different versions of MonetDBs on different platforms: amd64 with ubuntu 12, amd64 with fc19, and arm with ubuntu 12.
Reproducible: Always
### Steps to Reproduce:
1.configure with "--disable-shared" and make
2.create dbfarm using monetdbd
3.create a database "voc" using monetdb, and start it
### Actual Results:
database 'voc' started up, but failed to open up a communication channel
## Comment 19769
Date: 2014-04-24 15:25:30 +0200
From: @sjoerdmullender
We do not support building using static libraries. Why do you want to do this?
| Compile with --disable-shared | https://api.github.com/repos/MonetDB/MonetDB/issues/3464/comments | 0 | 2020-11-30T12:38:58Z | 2024-06-27T12:02:54Z | https://github.com/MonetDB/MonetDB/issues/3464 | 753,444,955 | 3,464 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-03-29 20:49:00 +0100
From: Edward Clarkson <<edward.clarkson>>
To: SQL devs <<bugs-sql>>
Version: 11.17.13 (Jan2014-SP1)
CC: edward.clarkson, @njnes
Last updated: 2014-05-22 09:52:22 +0200
## Comment 19722
Date: 2014-03-29 20:49:00 +0100
From: Edward Clarkson <<edward.clarkson>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_2) AppleWebKit/537.74.9 (KHTML, like Gecko) Version/7.0.2 Safari/537.74.9
Build Identifier:
The SELECT query in the reproduction instructions causes the server to terminate, and appears to be the result of using a 1) SERIAL column in an aggregate function with 2) the DISTINCT modifier and 3) a column alias as the aggregate. The following queries return an empty result set as expected (using the test table defined in the reproduction instructions):
[count DISTINCT int instead of serial]
SELECT a as t, count(distinct b) FROM test GROUP BY t;
[don't aggregate by column alias]
SELECT a, count(distinct c) FROM test GROUP BY a;
[don't count DISTINCT]
SELECT a as t, count(c) FROM test GROUP BY t;
The following also terminates:
SELECT a as t, sum(distinct b) FROM test GROUP BY t;
Reproducible: Always
### Steps to Reproduce:
1. Start the server using the default demo database.
2. In the M5 client, execute:
CREATE TABLE test (a int, b int, c serial);
SELECT a as t, count(distinct c) FROM test GROUP BY t;
### Actual Results:
Server terminates.
### Expected Results:
Empty result set.
This is a stock windows install, but just in case this is helpful:
C:\Program Files\MonetDB\MonetDB5>M5server.bat --version
MonetDB 5 server v11.17.9 "Jan2014" (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 5.9GiB available memory, 4 available cpu cores
Libraries:
libpcre: 8.13 2011-08-16
openssl: OpenSSL 1.0.1b 26 Apr 2012
libxml2: 2.7.8
Compiled by: monet@LAB03 (x86_64-pc-winnt)
Compilation: cl -GF -W3 -WX -MD -nologo -Ox -Zi
Linking : cl -GF -W3 -WX -MD -nologo -Ox -Zi
## Comment 19723
Date: 2014-03-30 12:22:42 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [f3b9839c7ed6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=f3b9839c7ed6) 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=f3b9839c7ed6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=f3b9839c7ed6)
Changeset description:
fixed bug #3463, make sure we also handle aliases in the aggr list when reducing
groupby expressions
## Comment 19726
Date: 2014-03-31 15:59:28 +0200
From: Edward Clarkson <<edward.clarkson>>
Wow, that was fast, thanks Niels!
FWIW, I am really pleased with working with MonetDB the past several months. Really impressive OSS project.
| Crash on SELECT with SERIAL aggregation and GROUP BY column alias's | https://api.github.com/repos/MonetDB/MonetDB/issues/3463/comments | 0 | 2020-11-30T12:38:55Z | 2024-06-27T12:02:53Z | https://github.com/MonetDB/MonetDB/issues/3463 | 753,444,940 | 3,463 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-03-28 17:31:18 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: @njnes
Last updated: 2014-05-22 09:52:29 +0200
## Comment 19715
Date: 2014-03-28 17:31:18 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:27.0) Gecko/20100101 Firefox/27.0
Build Identifier:
The SQL query:
SELECT * FROM SYS.ARGS
WHERE FUNC_ID NOT IN (SELECT * FROM SYS.FUNCTIONS)
is accepted by MonetDB and even executed!
However the where-clause is invalid as
SELECT * FROM SYS.FUNCTIONS
returns multiple columns which cannot be mapped to 1 col: FUNC_ID
This invalid SQL query should be detected by the SQL parser/executor and reported as an SQL error.
Instead MonetDB executes it returns 3401 rows, same data as when executing:
SELECT * FROM SYS.ARGS;
Reproducible: Always
### Steps to Reproduce:
1. start mserver5
2. start SQuirreL connecting to demo database
3. execute SQL query:
SELECT * FROM SYS.ARGS
WHERE FUNC_ID NOT IN (SELECT * FROM SYS.FUNCTIONS)
### Actual Results:
MonetDB executes it returns 3401 rows.
Same data as when executing: SELECT * FROM SYS.ARGS;
### Expected Results:
An SQL syntax error (invalid subquery as it returns more than 1 col)
The correct SQL query:
SELECT * FROM SYS.ARGS
WHERE FUNC_ID NOT IN (SELECT ID FROM SYS.FUNCTIONS)
works fine and returns 0 rows as expected.
## Comment 19718
Date: 2014-03-29 14:20:50 +0100
From: @njnes
properly check correct types (ie also number of columns should match)
## Comment 19721
Date: 2014-03-29 14:24:04 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [c5338ccd7803](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=c5338ccd7803) 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=c5338ccd7803](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=c5338ccd7803)
Changeset description:
fixed bugs 3458, 3461 and 3462
3461 added missing select
3462 check number of columns
| Invalid SQL (IN with subquery which returns multiple columns) is accepted | https://api.github.com/repos/MonetDB/MonetDB/issues/3462/comments | 0 | 2020-11-30T12:38:53Z | 2024-06-27T12:02:52Z | https://github.com/MonetDB/MonetDB/issues/3462 | 753,444,923 | 3,462 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-03-28 17:15:59 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: @njnes
Last updated: 2014-05-22 09:52:32 +0200
## Comment 19714
Date: 2014-03-28 17:15:59 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:27.0) Gecko/20100101 Firefox/27.0
Build Identifier:
Execute following SQL query on a new demo database
SELECT * FROM SYS.ARGS WHERE FUNC_ID NOT IN (SELECT ID FROM SYS.FUNCTIONS) OR FUNC_ID NOT IN (SELECT * FROM SYS.FUNCTIONS)
Reproducible: Always
### Steps to Reproduce:
1. start mserver5
2. start SQuirreL connecting to demo database
3. execute SQL query:
SELECT * FROM SYS.ARGS
WHERE FUNC_ID NOT IN (SELECT ID FROM SYS.FUNCTIONS)
OR FUNC_ID NOT IN (SELECT * FROM SYS.FUNCTIONS)
### Actual Results:
builtin opt gdk_dbpath = /ufs/dinther/MonetDBFebr2014/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
MonetDB 5 server v11.17.9 "Jan2014"
Serving database 'demo', using 8 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs dynamically linked
Found 7.733 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
>./start_mserver5: line 1: 11671 Segmentation fault ./MonetDBFebr2014/bin/mserver5 -d10
### Expected Results:
No server crash, but a SQL syntax error as the construction
FUNC_ID NOT IN (SELECT * FROM SYS.FUNCTIONS)
is not valid.
SELECT * FROM SYS.ARGS WHERE FUNC_ID NOT IN (SELECT ID FROM SYS.FUNCTIONS)
works fine (reports 0 rows).
SELECT * FROM SYS.ARGS WHERE FUNC_ID NOT IN (SELECT * FROM SYS.FUNCTIONS)
also executes (many rows returned), but should have given an SQL syntax error (I will report that as a separate bug)
Only the combination of the two where-clauses with an OR lets the server crash.
## Comment 19717
Date: 2014-03-29 14:20:02 +0100
From: @njnes
fixed crash added missing select operator
## Comment 19720
Date: 2014-03-29 14:23:54 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [c5338ccd7803](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=c5338ccd7803) 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=c5338ccd7803](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=c5338ccd7803)
Changeset description:
fixed bugs 3458, 3461 and 3462
3461 added missing select
3462 check number of columns
| mserver5 crash on SQL: SELECT * FROM SYS.ARGS WHERE FUNC_ID NOT IN (SELECT ID FROM SYS.FUNCTIONS) OR FUNC_ID NOT IN (SELECT * FROM SYS.FUNCTIONS) | https://api.github.com/repos/MonetDB/MonetDB/issues/3461/comments | 0 | 2020-11-30T12:38:51Z | 2024-06-27T12:02:51Z | https://github.com/MonetDB/MonetDB/issues/3461 | 753,444,899 | 3,461 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-03-28 14:55:32 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: clients devs <<bugs-clients>>
Version: 11.21.19 (Jul2015-SP4)
Last updated: 2016-08-04 18:27:25 +0200
## Comment 19713
Date: 2014-03-28 14:55:32 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:27.0) Gecko/20100101 Firefox/27.0
Build Identifier:
In java/src/nl/cwi/monetdb/jdbc/MonetDatabaseMetaData.java the methods supportsConvert(), supportsConvert(int fromType, int toType) are NOT correctly implemented. They both return false which not correct.
In ODBC driver code (clients/odbc/driver/SQLGetInfo.c) these are implemented see: SQLGetInfo() for:
case SQL_CONVERT_BIGINT:
case SQL_CONVERT_BINARY:
case SQL_CONVERT_BIT:
case SQL_CONVERT_CHAR:
case SQL_CONVERT_DATE:
case SQL_CONVERT_DECIMAL:
case SQL_CONVERT_DOUBLE:
case SQL_CONVERT_FLOAT:
case SQL_CONVERT_INTEGER:
case SQL_CONVERT_INTERVAL_DAY_TIME:
case SQL_CONVERT_INTERVAL_YEAR_MONTH:
case SQL_CONVERT_LONGVARBINARY:
case SQL_CONVERT_LONGVARCHAR:
case SQL_CONVERT_NUMERIC:
case SQL_CONVERT_REAL:
case SQL_CONVERT_SMALLINT:
case SQL_CONVERT_TIME:
case SQL_CONVERT_TIMESTAMP:
case SQL_CONVERT_TINYINT:
case SQL_CONVERT_VARBINARY:
case SQL_CONVERT_VARCHAR:
nValue = SQL_CVT_BIGINT |
SQL_CVT_BINARY |
SQL_CVT_BIT |
SQL_CVT_CHAR |
SQL_CVT_DATE |
SQL_CVT_DECIMAL |
SQL_CVT_DOUBLE |
SQL_CVT_FLOAT |
SQL_CVT_INTEGER |
SQL_CVT_INTERVAL_DAY_TIME |
SQL_CVT_INTERVAL_YEAR_MONTH |
SQL_CVT_LONGVARBINARY |
SQL_CVT_LONGVARCHAR |
SQL_CVT_NUMERIC |
SQL_CVT_REAL |
SQL_CVT_SMALLINT |
SQL_CVT_TIME |
SQL_CVT_TIMESTAMP |
SQL_CVT_TINYINT |
SQL_CVT_VARBINARY |
SQL_CVT_VARCHAR;
len = sizeof(SQLUINTEGER);
break;
case SQL_CONVERT_FUNCTIONS:
nValue = SQL_FN_CVT_CAST | SQL_FN_CVT_CONVERT;
len = sizeof(SQLUINTEGER);
break;
MonetDatabaseMetaData.java needs to be updated for the mentioned methods.
Reproducible: Always
### Steps to Reproduce:
1. start mserver5
2. start SQuirreL connecting to demo database
3. in Objects panel, select tab Metadata, search for property "supportsConvert". it shows value: false
### Actual Results:
both methods return false
### Expected Results:
method supportsConvert() should return true.
method supportsConvert(int fromType, int toType) should return true or false depending on the combination of fromType and toType.
SQL query which uses convert function:
SELECT DISTINCT type, convert(type, char(6)) as type_str, convert(type, double) as type_dbl, sql, convert(sql, int) as sql_int, convert(sql, char(5)) as sql_str
FROM SYS.FUNCTIONS
SQL query which uses cast function:
SELECT DISTINCT type, cast(type as char(5)) as type_str, cast(type as double) as type_dbl, sql, cast(sql as int) as sql_int, cast(sql as char(5)) as sql_str
FROM SYS.FUNCTIONS
Both SQL queries are accepted by MonetDB and produce output:
type type_str type_dbl sql sql_int sql_str
1 1 1 false 0 0
3 3 3 false 0 0
1 1 1 true 1 1
4 4 4 true 1 1
2 2 2 true 1 1
3 3 3 true 1 1
## Comment 22157
Date: 2016-05-19 19:28:38 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [67d578e84e0b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=67d578e84e0b) 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=67d578e84e0b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=67d578e84e0b)
Changeset description:
Implemented MonetDatabaseMetaData.supportsConvert() and
MonetDatabaseMetaData.supportsConvert(int fromType, int toType) methods.
It used to always return false. Now it returns true for the supported conversions.
This fixes Bug #3460.
## Comment 22248
Date: 2016-07-14 17:44:46 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [76af7753bd2e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=76af7753bd2e) 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=76af7753bd2e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=76af7753bd2e)
Changeset description:
Test script to test scalar functions convert(fromType, toType) and cast(fromType as toType)
for most MonetDB SQL data types and data values
See also Bug #3460.
## Comment 22282
Date: 2016-08-04 18:27:25 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [517709436cea](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=517709436cea) 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=517709436cea](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=517709436cea)
Changeset description:
Run convert-function-test.Bug-3460 serial (1 thread) to get stable output
| incomplete implementation of JDBC driver supportsConvert(), supportsConvert(int fromType, int toType) methods in MonetDatabaseMetaData.java | https://api.github.com/repos/MonetDB/MonetDB/issues/3460/comments | 0 | 2020-11-30T12:38:48Z | 2024-06-27T12:02:50Z | https://github.com/MonetDB/MonetDB/issues/3460 | 753,444,875 | 3,460 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-03-28 14:10:45 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: clients devs <<bugs-clients>>
Version: 11.17.9 (Jan2014)
Last updated: 2014-10-31 14:14:24 +0100
## Comment 19711
Date: 2014-03-28 14:10:45 +0100
From: Martin van Dinther <<martin.van.dinther>>
In java/src/nl/cwi/monetdb/jdbc/MonetDatabaseMetaData.java the methods getNumericFunctions(), getStringFunctions(), getSystemFunctions(), getTimeDateFunctions() and getSQLKeywords() are NOT correctly implemented. They all return "" which not correct.
In ODBC driver code (clients/odbc/driver/SQLGetInfo.c) these are implemented see: SQLGetInfo() for:
case SQL_NUMERIC_FUNCTIONS:
case SQL_STRING_FUNCTIONS:
case SQL_SYSTEM_FUNCTIONS:
case SQL_TIMEDATE_FUNCTIONS:
case SQL_KEYWORDS:
Step 1: MonetDatabaseMetaData.java needs to be updated for the mentioned methods.
Step 2: Ideally the list of supported functions should be retrieved from the server metadata (such as SYS.FUNCTIONS table). That way we do not need to manually sync the three (and possibly more) hard coded places, but it will become dynamic and in-sync with the SQL database server. This implementation however will require more programming and testing work.
## Comment 19899
Date: 2014-07-11 16:08:31 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [ec61f772de56](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ec61f772de56) 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=ec61f772de56](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ec61f772de56)
Changeset description:
Added implementation for methods getNumericFunctions(), getStringFunctions(), getSystemFunctions(), getTimeDateFunctions() and getSQLKeywords().
This fixes bug #3459
## Comment 19941
Date: 2014-08-01 18:46:10 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [e20766848db4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e20766848db4) 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=e20766848db4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=e20766848db4)
Changeset description:
Added implementation for JDBC Database Meta data methods: getNumericFunctions(), getStringFunctions(), getSystemFunctions(), getTimeDateFunctions() and getSQLKeywords().
This fixes bug #3459
## Comment 20369
Date: 2014-10-31 14:14:24 +0100
From: @sjoerdmullender
Oct2014 has been released.
| incomplete implementation of JDBC driver getNumericFunctions(), getStringFunctions(), getSystemFunctions(), getTimeDateFunctions(), getSQLKeywords() methods in MonetDatabaseMetaData.java | https://api.github.com/repos/MonetDB/MonetDB/issues/3459/comments | 0 | 2020-11-30T12:38:46Z | 2024-06-27T12:02:49Z | https://github.com/MonetDB/MonetDB/issues/3459 | 753,444,856 | 3,459 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-03-28 12:55:29 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.17.13 (Jan2014-SP1)
CC: @hannesmuehleisen, @njnes
Last updated: 2014-05-22 09:52:27 +0200
## Comment 19709
Date: 2014-03-28 12:55:29 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:27.0) Gecko/20100101 Firefox/27.0
Build Identifier:
MonetDB 5 server v11.17.9 "Jan2014"
Serving database 'demo', using 8 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs dynamically linked
Found 7.733 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
>./start_mserver5: line 1: 7467 Segmentation fault ./MonetDBFebr2014/bin/mserver5
Reproducible: Always
### Steps to Reproduce:
1. start mserver5
2. start SQuirreL connecting to demo database
3. execute SQL query: SELECT COUNT(*) FROM SYS.TABLES HAVING COUNT(*) > 0
### Actual Results:
MonetDB 5 server v11.17.9 "Jan2014"
Serving database 'demo', using 8 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs dynamically linked
Found 7.733 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
>./start_mserver5: line 1: 7467 Segmentation fault ./MonetDBFebr2014/bin/mserver5
### Expected Results:
1 row, 1 col with value 40
query
SELECT COUNT(*) FROM SYS.TABLES
works fine
but adding the HAVING-clause causes the server to crash immediately (Segmentation fault).
The SQL query is correct SQL (as it works on DB2, Oracle, etc).
## Comment 19710
Date: 2014-03-28 13:02:32 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [db58be30cce8](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=db58be30cce8) 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=db58be30cce8](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=db58be30cce8)
Changeset description:
Added test for bug #3458.
## Comment 19712
Date: 2014-03-28 14:32:52 +0100
From: @hannesmuehleisen
Good one, confirmed.
## Comment 19716
Date: 2014-03-29 14:17:52 +0100
From: @njnes
fixed, ie. properly initialize a select relational operator
## Comment 19719
Date: 2014-03-29 14:23:50 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [c5338ccd7803](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=c5338ccd7803) 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=c5338ccd7803](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=c5338ccd7803)
Changeset description:
fixed bugs 3458, 3461 and 3462
3461 added missing select
3462 check number of columns
| mserver5 crash on SQL: SELECT COUNT(*) FROM SYS.TABLES HAVING COUNT(*) > 0 | https://api.github.com/repos/MonetDB/MonetDB/issues/3458/comments | 0 | 2020-11-30T12:38:44Z | 2024-06-27T12:02:48Z | https://github.com/MonetDB/MonetDB/issues/3458 | 753,444,835 | 3,458 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-03-21 09:55:29 +0100
From: @bartscheers
To: SQL devs <<bugs-sql>>
Version: 11.17.13 (Jan2014-SP1)
CC: @njnes
Last updated: 2014-05-22 09:52:34 +0200
## Comment 19704
Date: 2014-03-21 09:55:29 +0100
From: @bartscheers
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:27.0) Gecko/20100101 Firefox/27.0
Build Identifier:
After the client kernel crashes or client machine is switched off by pressing power button, the original connection remains established on server side. This may cause unexpected huge growth of the sql_logs directory of the corresponding database.
Reproducible: Always
### Steps to Reproduce:
1. mclient -hmyremote.org -dmydb -ume
2. check netstat to see established tcp connection
3. push client's power button
4. repeat step 2
### Actual Results:
tcp 0 0 myremote.org:52000 myclient.:48421 ESTABLISHED 18707/mserver5
### Expected Results:
Connection should be closed properly
## Comment 19705
Date: 2014-03-23 13:45:09 +0100
From: @njnes
is this still the case after the keep alive changes?
## Comment 19724
Date: 2014-03-31 12:50:51 +0200
From: @bartscheers
Yes, this works in the Jan2014 branch (hg id f3b9839c7ed6)!
## Comment 19725
Date: 2014-03-31 14:05:30 +0200
From: @sjoerdmullender
I guess we can close it then.
| When kernel of remote client crashes, the connection remains established on server side | https://api.github.com/repos/MonetDB/MonetDB/issues/3457/comments | 0 | 2020-11-30T12:38:41Z | 2024-06-27T12:02:47Z | https://github.com/MonetDB/MonetDB/issues/3457 | 753,444,807 | 3,457 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-03-20 17:18:01 +0100
From: Radovan Bičiště <<radovan.biciste>>
To: clients devs <<bugs-clients>>
Version: 11.17.13 (Jan2014-SP1)
CC: @njnes
Last updated: 2014-05-22 09:52:25 +0200
## Comment 19703
Date: 2014-03-20 17:18:01 +0100
From: Radovan Bičiště <<radovan.biciste>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:27.0) Gecko/20100101 Firefox/27.0
Build Identifier:
Insert fails with syntax error when running process from KETTLE (http://ketttle.pentaho.com). Issue seems to be related to certain combination of input values bound to a prepared statement.
Log from the tool:
2014/03/20 16:58:12 - Produkty - MonetDB.0 - Prepared statement : INSERT INTO produkty_stage (produkt_id, pdk, sukl, id_zbozi, nazev, brand, atc1, atc2, atc3, ean, vyrobce) VALUES ( ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)
2014/03/20 16:58:12 - Produkty - MonetDB.0 - ERROR (version 5.0.1-stable, build 1 from 2013-11-15_16-08-58 by buildguy) : Because of an error, this step can't continue:
2014/03/20 16:58:12 - Produkty - MonetDB.0 - ERROR (version 5.0.1-stable, build 1 from 2013-11-15_16-08-58 by buildguy) : org.pentaho.di.core.exception.KettleException:
2014/03/20 16:58:12 - Produkty - MonetDB.0 - Error inserting row into table [produkty_stage] with values: [50071486], [2861681], [-], [50071486], [EUCERIN EVEN BRIGHTER DENNÍ KRÉM 50ML PROMO PACK], [BEIE], [-], [-], [-], [9005800222462], [BEIERSDORF S.R.O.]
2014/03/20 16:58:12 - Produkty - MonetDB.0 -
2014/03/20 16:58:12 - Produkty - MonetDB.0 - Error inserting/updating row
2014/03/20 16:58:12 - Produkty - MonetDB.0 - syntax error, unexpected ',' in: "exec 2(,"
2014/03/20 16:58:12 - Produkty - MonetDB.0 -
2014/03/20 16:58:12 - Produkty - MonetDB.0 -
2014/03/20 16:58:12 - Produkty - MonetDB.0 - at org.pentaho.di.trans.steps.tableoutput.TableOutput.writeToTable(TableOutput.java:445)
2014/03/20 16:58:12 - Produkty - MonetDB.0 - at org.pentaho.di.trans.steps.tableoutput.TableOutput.processRow(TableOutput.java:128)
2014/03/20 16:58:12 - Produkty - MonetDB.0 - at org.pentaho.di.trans.step.RunThread.run(RunThread.java:60)
2014/03/20 16:58:12 - Produkty - MonetDB.0 - at java.lang.Thread.run(Thread.java:744)
2014/03/20 16:58:12 - Produkty - MonetDB.0 - Caused by: org.pentaho.di.core.exception.KettleDatabaseException:
2014/03/20 16:58:12 - Produkty - MonetDB.0 - Error inserting/updating row
2014/03/20 16:58:12 - Produkty - MonetDB.0 - syntax error, unexpected ',' in: "exec 2(,"
2014/03/20 16:58:12 - Produkty - MonetDB.0 -
2014/03/20 16:58:12 - Produkty - MonetDB.0 - at org.pentaho.di.core.database.Database.insertRow(Database.java:1193)
2014/03/20 16:58:12 - Produkty - MonetDB.0 - at org.pentaho.di.trans.steps.tableoutput.TableOutput.writeToTable(TableOutput.java:284)
2014/03/20 16:58:12 - Produkty - MonetDB.0 - ... 3 more
2014/03/20 16:58:12 - Produkty - MonetDB.0 - Caused by: java.sql.SQLException: syntax error, unexpected ',' in: "exec 2(,"
2014/03/20 16:58:12 - Produkty - MonetDB.0 - at nl.cwi.monetdb.jdbc.MonetConnection$ResponseList.executeQuery(MonetConnection.java:2535)
2014/03/20 16:58:12 - Produkty - MonetDB.0 - at nl.cwi.monetdb.jdbc.MonetConnection$ResponseList.processQuery(MonetConnection.java:2284)
2014/03/20 16:58:12 - Produkty - MonetDB.0 - at nl.cwi.monetdb.jdbc.MonetStatement.internalExecute(MonetStatement.java:508)
2014/03/20 16:58:12 - Produkty - MonetDB.0 - at nl.cwi.monetdb.jdbc.MonetStatement.execute(MonetStatement.java:349)
2014/03/20 16:58:12 - Produkty - MonetDB.0 - at nl.cwi.monetdb.jdbc.MonetPreparedStatement.execute(MonetPreparedStatement.java:247)
2014/03/20 16:58:12 - Produkty - MonetDB.0 - at nl.cwi.monetdb.jdbc.MonetPreparedStatement.executeUpdate(MonetPreparedStatement.java:287)
2014/03/20 16:58:12 - Produkty - MonetDB.0 - at org.pentaho.di.core.database.Database.insertRow(Database.java:1153)
2014/03/20 16:58:12 - Produkty - MonetDB.0 - ... 4 more
2014/03/20 16:58:12 - cl_crm_dw2monet_produkty - ERROR (version 5.0.1-stable, build 1 from 2013-11-15_16-08-58 by buildguy) : Errors detected!
2014/03/20 16:58:12 - ceos.cl.dwh.oracle - Statement canceled!
Reproducible: Always
### Steps to Reproduce:
1. run process for transfer data from Oracle to Monet
2. watch the error happen.
### Actual Results:
Syntax error
### Expected Results:
Insert of row.
## Comment 19736
Date: 2014-04-09 13:07:46 +0200
From: @njnes
Could you tell us what the table definition of produkty_stage is? The values it self and the insert statement do not tell us that.
## Comment 19737
Date: 2014-04-09 13:13:03 +0200
From: Radovan Bičiště <<radovan.biciste>>
Thank you for looking into it.
The definition of the table is as follows:
CREATE TABLE drmaxtst.produkty_stage
(
produkt_id bigint,
pdk varchar(25),
forma varchar(25),
atc1 varchar(25),
atc2 varchar(25),
atc3 varchar(25),
ean varchar(25),
sukl varchar(25),
alias_vyrobce varchar(85),
stat_puvodu varchar(15),
doplnek varchar(55),
zbozi_id varchar(25),
pdk_sukl_ean varchar(255),
kod_pzt varchar(20),
nadzasoba varchar(6),
drz_sukl varchar(8),
datum_vytvoreni timestamp,
skupina int,
akce varchar(8),
procento decimal(14,4),
apa varchar(12),
id_zbozi bigint,
ciselnik varchar(6),
pozlist varchar(6),
brand varchar(35),
k_skupina varchar(100),
k_podskupina varchar(100),
nazev_vyrobce varchar(100),
nazev_korporace varchar(100),
atc1_popis varchar(100),
atc2_popis varchar(100),
atc3_popis varchar(100),
typ_vydeje varchar(5),
preference varchar(1),
category_group varchar(10),
subcategory1 varchar(3),
subcategory2 varchar(4),
subcategory3 varchar(6),
datum_prirazeni_vyrobci timestamp,
subcategory4 varchar(8),
subcategory5 varchar(8),
cenova_agresivita varchar(15),
kategorizace int,
kategorie varchar(10),
kategorie_ii varchar(2),
kategorie_iii varchar(3),
kategorie_iv varchar(4),
kategorie_v varchar(6),
kategorie_vi varchar(10),
kategorie_jmeno varchar(200),
kategorie_farmis varchar(50),
vyrobce varchar(100),
id_cenova_agresivita varchar(20),
nazev_orig varchar(150),
nazev varchar(150)
);
## Comment 19780
Date: 2014-04-25 16:22:31 +0200
From: Radovan Bičiště <<radovan.biciste>>
Eventuallly I had to fix it myself. Problem was in MonetPreparedStament on line 983:
int dot = xStr.indexOf(".");
if (dot != 0)
xStr = xStr.substring(0, Math.min(xStr.length(), dot + 1 + scale[i]));
works for me as
int dot = xStr.indexOf(".");
if (dot >= 0)
xStr = xStr.substring(0, Math.min(xStr.length(), dot + 1 + scale[i]));
If the number does not contain decimal point ".", the method returns empty string because method indexOf returns -1.
## Comment 19781
Date: 2014-04-25 16:37:42 +0200
From: Radovan Bičiště <<radovan.biciste>>
Created attachment 283
MonetPreparedStatement.java
Modified class.
> Attached file: [MonetPreparedStatement.java](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3456_MonetPreparedStatement.java_283) (text/x-java, 85819 bytes)
> Description: MonetPreparedStatement.java
## Comment 19788
Date: 2014-04-30 10:54:29 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [53d0352eaa3b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=53d0352eaa3b) 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=53d0352eaa3b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=53d0352eaa3b)
Changeset description:
Fix for bug #3456. Bumped version number.
## Comment 19789
Date: 2014-04-30 10:54:50 +0200
From: @sjoerdmullender
Patch applied. Thanks.
| Insert fails | https://api.github.com/repos/MonetDB/MonetDB/issues/3456/comments | 0 | 2020-11-30T12:38:39Z | 2024-06-27T12:02:46Z | https://github.com/MonetDB/MonetDB/issues/3456 | 753,444,779 | 3,456 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-03-17 17:02:08 +0100
From: Richard Hughes <<richard.monetdb>>
To: clients devs <<bugs-clients>>
Version: 11.17.13 (Jan2014-SP1)
Last updated: 2014-05-22 09:52:35 +0200
## Comment 19701
Date: 2014-03-17 17:02:08 +0100
From: Richard Hughes <<richard.monetdb>>
User-Agent: Opera/9.80 (Windows NT 6.1; WOW64; Edition United Kingdom Local) Presto/2.12.388 Version/12.16
Build Identifier:
The world's most trivial .NET test application using the x64 MonetDB ODBC driver fails with an overflow exception deep in the bowels of the CLR when converting a string column to .NET types.
OdbcDataReader.Item[Int32] calls SQLColAttribute(..., SQL_DESC_CONCISE_TYPE, ...) as part of its operation. The MonetDB implementation stores SQL_WCHAR (-8) as an int into the caller-supplied buffer, which .NET then treats as an int64 needing to be truncated to int32. 0xfffffff8 doesn't fit in an int32, so the interop code generates an overflow exception. It would appear Microsoft didn't read their own documentation: "Please note that some drivers may only write the lower 32-bit or 16-bit of a buffer and leave the higher-order bit unchanged."
The fix is trivial:
diff -r [5ca1f6398fec](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=5ca1f6398fec) clients/odbc/driver/SQLColAttribute.c
--- a/clients/odbc/driver/SQLColAttribute.c Fri Mar 14 12:55:23 2014 +0100
+++ b/clients/odbc/driver/SQLColAttribute.c Mon Mar 17 15:57:23 2014 +0000
@@ -117,7 +117,7 @@
break;
case SQL_DESC_CONCISE_TYPE: /* SQL_COLUMN_TYPE */
if (NumericAttributePtr)
- *(int *) NumericAttributePtr = rec->sql_desc_concise_type;
+ *(SQLLEN *) NumericAttributePtr = rec->sql_desc_concise_type;
break;
case SQL_DESC_COUNT:
if (NumericAttributePtr)
It would be nice if you could go through all the other cases in that switch to see which others can store negative numbers.
Reproducible: Always
### Steps to Reproduce:
1. Build the attached
2. Run clrmonet "select 'Hello, world!'"
### Actual Results:
Crash, overflow exception
### Expected Results:
Hello, world!
## Comment 19702
Date: 2014-03-17 17:03:14 +0100
From: Richard Hughes <<richard.monetdb>>
Created attachment 279
Test client
> Attached file: [clrmonet.cs](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3455_clrmonet.cs_279) (application/octet-stream, 801 bytes)
> Description: Test client
## Comment 19728
Date: 2014-04-04 11:21:30 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [dc9fc070e99b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=dc9fc070e99b) 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=dc9fc070e99b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=dc9fc070e99b)
Changeset description:
The NumericAttributePtr argument of SQLColAttribute is a pointer to SQLLEN.
Some instances of filling in the value used it as a pointer to int.
This fixes bug #3455.
## Comment 19729
Date: 2014-04-04 11:22:18 +0200
From: @sjoerdmullender
The pointer is now always used as a pointer to SQLLEN.
| String columns unusable from 64-bit .NET via ODBC | https://api.github.com/repos/MonetDB/MonetDB/issues/3455/comments | 0 | 2020-11-30T12:38:36Z | 2024-06-27T12:02:45Z | https://github.com/MonetDB/MonetDB/issues/3455 | 753,444,752 | 3,455 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-03-17 12:38:14 +0100
From: @mlkersten
To: SQL devs <<bugs-sql>>
Version: 11.15.11 (Feb2013-SP3)
CC: @njnes
Last updated: 2014-03-23 13:46:19 +0100
## Comment 19700
Date: 2014-03-17 12:38:14 +0100
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:27.0) Gecko/20100101 Firefox/27.0
Build Identifier:
After fixing some compilation issues, scenario00.sql can trigger the assertion error in
sql_mvc.c:465: mvc_reset: Assertion `m->session->active == 0' failed
when scenario00.sql statements are thrown at the server as one block.
Stepwise, it works fine,...seems a race/concurrency conflict.
Reproducible: Always
## Comment 19706
Date: 2014-03-23 13:46:19 +0100
From: @njnes
this was/is in the datacell startup code
| Assertion error Datacell | https://api.github.com/repos/MonetDB/MonetDB/issues/3454/comments | 0 | 2020-11-30T12:38:34Z | 2024-06-28T07:19:44Z | https://github.com/MonetDB/MonetDB/issues/3454 | 753,444,720 | 3,454 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-03-10 15:57:56 +0100
From: @dnedev
To: buildtools devs <<bugs-buildtools>>
Version: 11.17.9 (Jan2014)
Last updated: 2014-03-11 09:47:29 +0100
## Comment 19686
Date: 2014-03-10 15:57:56 +0100
From: @dnedev
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_2) AppleWebKit/537.74.9 (KHTML, like Gecko) Version/7.0.2 Safari/537.74.9
Build Identifier:
The configure tool checks for the presence of libatomic_ops, but even if the library is not available, building of the core modules is enable: gdk, monet5, sql.
Reproducible: Always
### Steps to Reproduce:
1. Set-up a build system on OS X
2. Install/provide all required libs, but libatomic_ops
3. Configure a build
4. Make MonetDB
### Actual Results:
Build fails.
### Expected Results:
Core modules dependent on libatomic_ops should be disabled by configure.
atomic_ops was correctly reported as not found.
After providing the library, cleaning the build and re-running configure the build of the core modules is successful.
## Comment 19687
Date: 2014-03-10 15:58:49 +0100
From: @dnedev
I got to try this after cleaning up my system and (systematically) re-trying to build MonetDB on OS X.
## Comment 19688
Date: 2014-03-10 16:34:09 +0100
From: @sjoerdmullender
libatomic_ops is optional. If it's not available, we use a different mechanism to do atomic operations, and if atomic operations are not possible, we fall back to using the pthread library.
So, if building fails if libatomic_ops is not available, it's something else. I would like to see a log of the build process.
## Comment 19689
Date: 2014-03-10 16:36:48 +0100
From: @dnedev
Created attachment 278
Log from the failed build
> Attached file: [build.log](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3453_build.log_278) (application/octet-stream, 16894 bytes)
> Description: Log from the failed build
## Comment 19690
Date: 2014-03-10 16:38:09 +0100
From: @dnedev
pthread is indeed marked as enabled for the build by configure.
## Comment 19691
Date: 2014-03-10 17:13:01 +0100
From: @sjoerdmullender
The errors I see are:
Undefined symbols for architecture x86_64:
"_iconv", referenced from:
_main in mclient.o
_doFile in mclient.o
"_iconv_open", referenced from:
_main in mclient.o
This has nothing to do with libatomic_ops.
## Comment 19692
Date: 2014-03-10 17:17:41 +0100
From: @dnedev
I thought so too, but the only thing that I did to make it build was to install libatomic_ops from MacPorts and it came with no dependencies. Thus my conjecture that it issues was with libatomic_ops (or related) on my OS X system. My be something to do with phtread and Clan/LLVM.
## Comment 19693
Date: 2014-03-11 08:27:56 +0100
From: @sjoerdmullender
Turned out to be a missing --with-libiconv-prefix=/opt/local argument.
## Comment 19698
Date: 2014-03-11 09:47:29 +0100
From: @dnedev
To avoid any further issues, I'll also add --with-libiconv-prefix=/opt/local flag to the new manual for building MonetDB on OS X with MacPorts.
| Configure enables building of core modules even without required libatomic_ops installed | https://api.github.com/repos/MonetDB/MonetDB/issues/3453/comments | 0 | 2020-11-30T12:38:31Z | 2024-06-28T07:19:43Z | https://github.com/MonetDB/MonetDB/issues/3453 | 753,444,689 | 3,453 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-03-05 13:14:06 +0100
From: @dnedev
To: clients devs <<bugs-clients>>
Version: 11.17.9 (Jan2014)
Last updated: 2014-03-11 09:19:35 +0100
## Comment 19671
Date: 2014-03-05 13:14:06 +0100
From: @dnedev
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_2) AppleWebKit/537.74.9 (KHTML, like Gecko) Version/7.0.2 Safari/537.74.9
Build Identifier:
ODBC driver build fails on OS X Mavericks due to a conflicting types for the SQLColAttribute in the UnixODBC library.
The error is the following:
ibtool: compile: gcc -DHAVE_CONFIG_H -I. -I../../../../clients/odbc/driver -I../../.. -I../../../../clients/odbc/driver -I../../mapilib -I../../../../clients/odbc/driver/../../mapilib -I../../../common/options -I../../../../clients/odbc/driver/../../../common/options -I/opt/local/include -DLIBMONETODBC -g -O2 -I/opt/local/include -c ../../../../clients/odbc/driver/SQLColAttributes.c -fno-common -DPIC -o .libs/libMonetODBC_la-SQLColAttributes.o
../../../../clients/odbc/driver/SQLColAttribute.c:243:1: error: conflicting types for 'SQLColAttribute'
SQLColAttribute(SQLHSTMT StatementHandle,
^
/opt/local/include/sql.h:609:24: note: previous declaration is here
SQLRETURN SQL_API SQLColAttribute(SQLHSTMT StatementHandle,
^
../../../../clients/odbc/driver/SQLColAttribute.c:272:1: error: conflicting types for 'SQLColAttributeA'
SQLColAttributeA(SQLHSTMT StatementHandle,
^
/opt/local/include/sqlucode.h:377:19: note: previous declaration is here
SQLRETURN SQL_API SQLColAttributeA(
^
../../../../clients/odbc/driver/SQLColAttribute.c:290:1: error: conflicting types for 'SQLColAttributeW'
SQLColAttributeW(SQLHSTMT StatementHandle,
^
/opt/local/include/sqlucode.h:29:19: note: previous declaration is here
SQLRETURN SQL_API SQLColAttributeW(
^
3 errors generated.
make[7]: *** [libMonetODBC_la-SQLColAttribute.lo] Error 1
Reproducible: Always
### Steps to Reproduce:
1. Download latest sources
2. Configure
3. Make
### Actual Results:
Build succeeds
### Expected Results:
Build fails
The error was eventually traced to an API change in the unixODBC library:
In the MonetDB/clients/odbc/driver/SQLColAttribute.c for the SQLColAttribute, SQLColAttributeA, SQLColAttributeW functions, the last parameter (NumericAttributePtr) is of type LENP_OR_POINTER_T.
In the latest unixODBC that parameter is of type (SQLLEN *).
I am running (Mac) OS X 10.6
Xcode 5.0.2 (Apple LLVM version 5.0 (clang-500.2.79) (based on LLVM 3.3svn))
unixODBC 2.3.1_0 installed from MacPorts 2.2.1
## Comment 19672
Date: 2014-03-05 13:14:56 +0100
From: @dnedev
I'll try to identify in which version of the unixODBC library was the change introduced.
## Comment 19673
Date: 2014-03-05 13:18:46 +0100
From: @sjoerdmullender
There is a check in configure.ag for just this thing. The type LENP_OR_POINTER_T doesn't actually exist but is defined to be either SQLLEN * or SQLPOINTER depending on the configure test. It seems the test doesn't work as expected.
Can you check your config.log file for the relevant test, or else attach the file here?
## Comment 19674
Date: 2014-03-05 13:24:51 +0100
From: @dnedev
Created attachment 276
config.log
Attaching the config.log
> Attached file: [config.log](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3452_config.log_276) (application/octet-stream, 411925 bytes)
> Description: config.log
## Comment 19675
Date: 2014-03-05 13:52:27 +0100
From: @dnedev
There is indeed an error for this test in the config.log
configure:21028: checking checking whether SQLColAttribute uses SQLLEN * or SQLPOINTER
configure:21053: gcc -c -g -O2 conftest.c >&5
conftest.c:68:14: fatal error: 'sql.h' file not found
include <sql.h>
^
1 error generated.
## Comment 19676
Date: 2014-03-05 14:14:16 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [0c281aca4f71](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=0c281aca4f71) 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=0c281aca4f71](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=0c281aca4f71)
Changeset description:
Set CPPFLAGS before running the compiler.
This hopefully fixes bug #3452.
## Comment 19677
Date: 2014-03-05 14:15:10 +0100
From: @sjoerdmullender
Can you try out the latest version in Mercurial?
## Comment 19678
Date: 2014-03-05 14:35:49 +0100
From: @dnedev
Unfortunately I cannot currently build the default branch on OS X, due to another issue:
clang: error: argument unused during compilation: '-pthread'
But applied the change over the MonetDB-11.17.11 build I used originally. After cleaning, it still failed with the same error and the 'sql.h' file not found is still present in the config.log.
## Comment 19679
Date: 2014-03-05 15:00:19 +0100
From: @sjoerdmullender
Did you recreate configure after changing configure.ag?
## Comment 19680
Date: 2014-03-05 15:06:22 +0100
From: @dnedev
Yep. Executed in that order:
make clean
patch
../configure --prefix=/Users/dnedev/monetdb/installation --with-unixodbc=/opt/local --with-sphinxclient=/opt/local --with-readline=/opt/local
make -j4
## Comment 19681
Date: 2014-03-05 15:12:32 +0100
From: @sjoerdmullender
Sorry, that doesn't patch configure, just configure.ag. You need to run bootstrap to create configure.ac from configure.ag, and configure from configure.ac. Whether bootstrap actually works when using the tar ball sources, I don't know.
I'd recommend trying:
hg clone -u Jan2014 http://dev.monetdb.org/MonetDB
cd MonetDB
./bootstrap
./configure ...
make ...
## Comment 19682
Date: 2014-03-05 15:15:39 +0100
From: @sjoerdmullender
./bootstrap does work from the tar ball.
## Comment 19683
Date: 2014-03-05 15:28:32 +0100
From: @dnedev
Got it. Cleaning the build, re-running bootstrap, re-configuring does it. The ODBC driver now builds successfully.
Just to note - the clang -pthread issue is also present on the Jan2014 branch.
## Comment 19695
Date: 2014-03-11 09:19:35 +0100
From: @sjoerdmullender
Jan2014-SP1 has been released.
| ODBC driver build fails on Mac OS X due to a conflicting types for the SQLColAttribute with the unixODBC library | https://api.github.com/repos/MonetDB/MonetDB/issues/3452/comments | 0 | 2020-11-30T12:38:28Z | 2024-06-27T12:02:41Z | https://github.com/MonetDB/MonetDB/issues/3452 | 753,444,657 | 3,452 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-03-04 23:25:25 +0100
From: Dann Corbit <<dcorbit>>
To: clients devs <<bugs-clients>>
Version: 11.21.19 (Jul2015-SP4)
CC: dcorbit, martin.van.dinther
Last updated: 2017-05-18 19:01:58 +0200
## Comment 19664
Date: 2014-03-04 23:25:25 +0100
From: Dann Corbit <<dcorbit>>
User-Agent: Mozilla/5.0 (Windows NT; Win64; x64; rv:26.0) Gecko/20100101 Firefox/26.0 Waterfox/26.0
Build Identifier:
When I create tables using MonetDB, the tables should be created as user tables and not as system tables.
So, for instance, if I create the tables for the voc shipping application using voc_dump.sql, all the new tables are registered as system tables. This is not correct behavior. In order to see these tables via inquiry, it is necessary to set the system tables flag, which should not be necessary. http://msdn.microsoft.com/en-us/library/ms711831(v=vs.85).aspx
Reproducible: Always
### Steps to Reproduce:
1. Create a database
2. Populate the database using voc_dump.sql from your site
3. Try to collect the table names using ODBC inquiry. The tables do not show up unless you request system tables.
### Actual Results:
User tables only show up via table inquiry when you ask for system tables
### Expected Results:
User tables show up when you ask for user tables
I can send you screen shots that demonstrate the problem, but I imagine you can easily reproduce it yourselves.
C:\Program Files\MonetDB\MonetDB5>mserver5 --version
MonetDB 5 server v11.17.9 "Jan2014" (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 31.9GiB available memory, 12 available cpu cores
Libraries:
libpcre: 8.13 2011-08-16
openssl: OpenSSL 1.0.1b 26 Apr 2012
libxml2: 2.7.8
Compiled by: monet@LAB03 (x86_64-pc-winnt)
Compilation: cl -GF -W3 -WX -MD -nologo -Ox -Zi
Linking : cl -GF -W3 -WX -MD -nologo -Ox -Zi
## Comment 19665
Date: 2014-03-04 23:29:09 +0100
From: Dann Corbit <<dcorbit>>
Created attachment 273
This shows the result set of requesting user tables
> Attached file: [request_normal_tables.png](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3451_request_normal_tables.png_273) (image/png, 6508 bytes)
> Description: This shows the result set of requesting user tables
## Comment 19666
Date: 2014-03-04 23:31:04 +0100
From: Dann Corbit <<dcorbit>>
Created attachment 274
This shows the result of requesting system tables.
Note the presence of table "voyages" for instance, which is a user table and not a system table.
> Attached file: [request_system_tables.png](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3451_request_system_tables.png_274) (image/png, 9075 bytes)
> Description: This shows the result of requesting system tables.
## Comment 19667
Date: 2014-03-04 23:42:27 +0100
From: Dann Corbit <<dcorbit>>
This bug has been present for quite a long time, but I was too lazy to report it previously.
## Comment 19668
Date: 2014-03-05 07:40:11 +0100
From: @grobian
From your screenshot it looks like you're using the "monetdb" superuser account. I'd argue that because of that, what you see is expected behaviour. Create/use a regular user for normal operations.
## Comment 19669
Date: 2014-03-05 07:53:56 +0100
From: Dann Corbit <<dcorbit>>
Created attachment 275
image that shows creation of sa table in SQL Server
In the attached image, you can see by looking at the query tab that I am logged in as 'sa' which is the system administrator.
I created a table with this account and as you can see, it is placed in the user tables and not in the system tables.
Food for thought.
> Attached file: [sa.png](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3451_sa.png_275) (image/png, 77248 bytes)
> Description: image that shows creation of sa table in SQL Server
## Comment 19670
Date: 2014-03-05 07:55:37 +0100
From: Dann Corbit <<dcorbit>>
Use Oracle and log in as the system account and create tables. They will not be created as system tables.
Use Microsoft SQL Server and log in as 'sa' and create tables. They will not be created as system tables.
Normally, a system table is something like the INFORMATION_SCHEMA tables, or database replication tables -- things of that nature.
However, if you have this unusual definition for system tables, that is OK. This sort of thing is not controlled by the ANSI and ISO SQL standards.
I would suggest that the unusual behavior be documented and that the sample VOC database have a user creation script at the top.
I suspect that I am not the only one who will be puzzled by this unusual definition.
## Comment 19731
Date: 2014-04-04 18:53:28 +0200
From: @sjoerdmullender
(In reply to comment 0)
> Steps to Reproduce:
> 1. Create a database
> 2. Populate the database using voc_dump.sql from your site
> 3. Try to collect the table names using ODBC inquiry. The tables do not
> show up unless you request system tables.
Which ODBC function is being used for this inquiry?
## Comment 25203
Date: 2017-04-06 17:35:40 +0200
From: Martin van Dinther <<martin.van.dinther>>
MonetDB uses schemas "sys" and "tmp" to store its system tables (and system views).
However it is also possible and allowed (but not recommended) to create user tables and views in those MonetDB system schemas.
So if your tables/views are created in "sys" or "tmp" schemas you need to provide the TableType argument "'TABLE','VIEW'" to SQLTables() to list only the user tables and views.
NB: MonetDB supports many TableTypes, to list them run query:
SELECT * FROM sys.table_types;
It also includes: 'SYSTEM TABLE' and 'SYSTEM VIEW' which can be used in the TableType argument of SQLTables().
Turn on ODBC logging and check the log file for calls of ODBC function SQLTables() to see which argument values your application really makes.
To list the user created table and views in MonetDB run query:
SELECT * FROM sys.tables WHERE NOT system;
This should list only user tables and views. If that is not the case then it is a bug.
The VOC example documentation has been updated. It includes steps to create a dedicated voc user and voc schema and sets that schema as the default schema for the voc user. Before the voc tables are created you need to login with the new voc user credentials. See: https://www.monetdb.org/Documentation/UserGuide/Tutorial
## Comment 25204
Date: 2017-04-06 20:50:32 +0200
From: Dann Corbit <<dcorbit>>
It is clear enough now how to cleanly differentiate a MonetDB system table and also how to create a user schema.
You can safely close the issue, I think.
| When user tables are created, all of these tables are reported as system tables | https://api.github.com/repos/MonetDB/MonetDB/issues/3451/comments | 0 | 2020-11-30T12:38:25Z | 2024-06-28T07:19:43Z | https://github.com/MonetDB/MonetDB/issues/3451 | 753,444,621 | 3,451 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-03-03 11:16:34 +0100
From: andrei <<andrei.ciobotar>>
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: andrei.ciobotar, @njnes, @swingbit
Last updated: 2014-09-01 11:27:01 +0200
## Comment 19657
Date: 2014-03-03 11:16:34 +0100
From: andrei <<andrei.ciobotar>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/33.0.1750.117 Safari/537.36
Build Identifier:
I'm attempting to run queries on a table with a little under 40 million rows and am receiving the following error:
GDK reported error.
BATproject: does not match always
Table structure is
field1 bigint,
field2 date,
field3 int,
field4 int,
field5 int
The only query so far that does not return the error seems to be select count(*), everything else, including a simple select * from throws the error immediately.
It is important to note that the error is thrown only for this specific large table. The same query on the same table but with less data (in a different database) runs flawlessly.
Reproducible: Always
### Steps to Reproduce:
not available, seems to be data specific
MonetDB 5 server v11.17.9 "Jan2014" (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 23.5GiB available memory, 12 available cpu cores
Libraries:
libpcre: 8.31 2012-07-06 (compiled with 8.31)
openssl: OpenSSL 1.0.1e 11 Feb 2013 (compiled with OpenSSL 1.0.1e 11 Feb 2013)
libxml2: 2.9.1 (compiled with 2.9.1)
Compiled by: root@dev.monetdb.org (x86_64-pc-linux-gnu)
Compilation: gcc -O3 -fomit-frame-pointer -pipe -Wp,-D_FORTIFY_SOURCE=2
Linking : /usr/bin/ld -m elf_x86_64
2014-02-26 14:51:27 MSG merovingian[25370]: Merovingian 1.7 (Jan2014) starting
2014-02-26 14:51:27 MSG merovingian[25370]: monitoring dbfarm /var/lib/monetdb/farms/ht
2014-02-26 14:51:27 MSG merovingian[25370]: accepting connections on TCP socket 0.0.0.0:50000
2014-02-26 14:51:27 MSG merovingian[25370]: accepting connections on UNIX domain socket /tmp/.s.monetdb.50000
2014-02-26 14:51:27 MSG discovery[25370]: listening for UDP messages on 0.0.0.0:50000
2014-02-26 14:51:27 MSG control[25370]: accepting connections on UNIX domain socket /tmp/.s.merovingian.50000
2014-02-26 14:51:40 MSG control[25370]: (local): created database 'ht_v1'
2014-02-26 14:51:46 MSG control[25370]: (local): served status list
2014-02-26 14:51:46 MSG control[25370]: (local): released database 'ht_v1'
2014-02-26 14:51:56 MSG merovingian[25370]: starting database 'ht_v1', up min/avg/max: 0s/0s/0s, crash average: 0.00 0.00 0.00 (0-0=0)
2014-02-26 14:51:56 MSG merovingian[25370]: proxying client (local) for database 'ht_v1' to mapi:monetdb:///var/lib/monetdb/farms/ht/ht_v1/.mapi.sock?database=ht_v1
2014-02-26 14:51:56 MSG merovingian[25370]: target connection is on local UNIX domain socket, passing on filedescriptor instead of proxying
2014-02-26 14:51:56 MSG ht_v1[25396]: arguments: /usr/bin/mserver5 --dbpath=/var/lib/monetdb/farms/ht/ht_v1 --set merovingian_uri=mapi:monetdb://monetdb1.srv.ht.net:50000/ht_v1 --set mapi_open=false --set mapi_port=0 --set mapi_usock=/var/lib/monetdb/farms/ht/ht_v1/.mapi.sock --set monet_vault_key=/var/lib/monetdb/farms/ht/ht_v1/.vaultkey --set gdk_nr_threads=12 --set max_clients=64 --set sql_optimizer=default_pipe --set monet_daemon=yes
2014-02-26 14:51:56 MSG ht_v1[25396]: MonetDB 5 server v11.17.9 "Jan2014"
2014-02-26 14:51:56 MSG ht_v1[25396]: Serving database 'ht_v1', using 12 threads
2014-02-26 14:51:56 MSG ht_v1[25396]: Compiled for x86_64-pc-linux-gnu/64bit with 64bit OIDs dynamically linked
2014-02-26 14:51:56 MSG ht_v1[25396]: Found 23.539 GiB available main-memory.
2014-02-26 14:51:56 MSG ht_v1[25396]: Copyright (c) 1993-July 2008 CWI.
2014-02-26 14:51:56 MSG ht_v1[25396]: Copyright (c) August 2008-2014 MonetDB B.V., all rights reserved
2014-02-26 14:51:56 MSG ht_v1[25396]: Visit http://www.monetdb.org/ for further information
2014-02-26 14:51:56 MSG ht_v1[25396]: Listening for UNIX domain connection requests on mapi:monetdb:///var/lib/monetdb/farms/ht/ht_v1/.mapi.sock
2014-02-26 14:51:56 MSG ht_v1[25396]: MonetDB/SQL module loaded
2014-02-26 14:51:56 MSG merovingian[25370]: proxying client (local) for database 'ht_v1' to mapi:monetdb:///var/lib/monetdb/farms/ht/ht_v1/.mapi.sock?database=ht_v1
2014-02-26 14:51:56 MSG merovingian[25370]: target connection is on local UNIX domain socket, passing on filedescriptor instead of proxying
2014-02-26 14:51:56 MSG ht_v1[25396]: SQL catalog created, loading sql scripts once
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 09_like.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 10_math.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 11_times.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 12_url.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 13_date.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 14_inet.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 15_querylog.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 16_tracelog.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 17_compress.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 18_dictionary.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 19_cluster.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 20_vacuum.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 21_dependency_functions.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 22_clients.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 23_skyserver.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 24_zorder.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 25_debug.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 26_sysmon.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 39_analytics.sql
2014-02-26 14:51:56 MSG ht_v1[25396]: loading sql script: 75_storagemodel.sql
## Comment 19659
Date: 2014-03-03 13:19:35 +0100
From: andrei <<andrei.ciobotar>>
Some more debug info:
mapi_query_part:44:select * from tbl;
fetch next block: start at:4298
got next block: length:56
text:!GDK reported error.
!BATproject: does not match always
got complete block:
text:!GDK reported error.
!BATproject: does not match always
read_line:!GDK reported error.
allocating new result set
got complete block:
text:!BATproject: does not match always
read_line:!BATproject: does not match always
fetch next block: start at:4354
got next block: length:0
text:
got complete block:
text:
read_line:
GDK reported error.
BATproject: does not match always
closing result set
## Comment 19660
Date: 2014-03-03 14:56:32 +0100
From: andrei <<andrei.ciobotar>>
I've been checking the merovingian log, apparently the database containing the table has crashed last night:
2014-03-02 00:46:01 MSG merovingian[25370]: database 'db1' (1260) was killed by signal SIGSEGV
## Comment 19661
Date: 2014-03-03 15:29:11 +0100
From: andrei <<andrei.ciobotar>>
[ 3259.213832] mserver5[1756]: segfault at 1000002d0 ip 00007f6fe9d34164 sp 00007fff48bd07f0 error 4 in libpthread-2.17.so[7f6fe9d2b000+17000]
Running saucy salamander 13.10.
## Comment 19663
Date: 2014-03-04 15:05:30 +0100
From: andrei <<andrei.ciobotar>>
as a last note, a downgrade to 11.15.17 resolved these issues
## Comment 19699
Date: 2014-03-12 10:50:42 +0100
From: @njnes
does switching of mitosis help?
## Comment 19755
Date: 2014-04-16 09:05:13 +0200
From: @njnes
andrei do you still have these problems. A select * from Table, is a very simple and frequently used query, for which no other reports exists. This seems to indicate this is cause by some form of data corruption (ie the columns seem no longer aligned). I guessing a save route to test the new version would be a
dump/restore, but given the 40M rows is will be a slow process.
## Comment 19756
Date: 2014-04-16 12:29:26 +0200
From: andrei <<andrei.ciobotar>>
hi niels, thanks for the answer.
we've downgraded monetdb to feb2013 sp5 and the issues seem to have been resolved for the most part. however, i suspect the root of our problems to be some sort of concurrency issue.
in order to feed data into the database, we've been using a number of php processes that read from an instruction queue and execute the insert/update queries on the monetdb database. this led to some strange errors like primary key violations (e.g. two rows with the same primary key -- we've confirmed that it was indeed a bug because inserting a third copy of that same row did trigger a primary key constraint error) and numerous instances of duplicate/missing rows. i suspect this was because of some issue with the rollback mechanism. the higher the number of processes, the more severe the symptoms were (we're averaging around 100 queries per second for some tables). in order to go around the problem and avoid rollbacks/concurrency errors, we've ended up organizing the php processes on a per table basis (i.e. one process for every table that does the inserts/updates) which seems to have worked. this was all on the 2013 sp5 version.
on the newer 2014 sp1 version, using multiple processes for every table seems to trigger this batproject error. what we did manage to do was run a where query and make three dumps of the data (data was organized by male, female and unknown) that we loaded into a new table -- this seems to have worked with the exception of ten missing rows (which can be due to the fact that count(*) was delivering an incorrect number, i'm not sure how it's calculated).
mitosis flags had no effect. i'm aware that it's frustrating to have a bug report without any steps to reproduce the problem but i hope this at least offers a hint to the potential source of the issue. starting today, we're going to run two different monetdb versions in parallel on different servers (feb 2013 sp5 and 2014 sp1) and execute the same queries on them (using one php process for every table, in order to avoid concurrency errors) in an effort to narrow down the list of potential points of failure. i'm happy to provide any extra info that i can.
## Comment 20111
Date: 2014-09-01 11:27:01 +0200
From: @swingbit
I also get the same error using the latest checkout (unreleased Oct2014 branch).
Unfortunately it's not easy to make it reproducible. I get it during a rather complex indexing process via JDBC and not always, although most of the times.
It does look like a concurrency issue.
The same on Feb2013 has no problem at all.
| BATproject error when running most queries | https://api.github.com/repos/MonetDB/MonetDB/issues/3450/comments | 0 | 2020-11-30T12:38:21Z | 2024-06-27T12:02:39Z | https://github.com/MonetDB/MonetDB/issues/3450 | 753,444,581 | 3,450 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-03-01 13:51:17 +0100
From: Jim Ohlstein <<jim>>
To: clients devs <<bugs-clients>>
Version: 11.17.9 (Jan2014)
CC: jim, @drstmane
Last updated: 2015-08-28 13:41:36 +0200
## Comment 19653
Date: 2014-03-01 13:51:17 +0100
From: Jim Ohlstein <<jim>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/33.0.1750.117 Safari/537.36
Build Identifier:
On testing the download of MonetDB-11.17.9-x86_64-FreeBSD-8.2-bin.tar.bz2 and trying to access a new database using "mclient" I get an error message as follows:
monetdbd: internal error while starting mserver, please refer to the logs
Reproducible: Always
### Steps to Reproduce:
1. monetdbd start /usr/local/monetdb/var/monetdb5/dbfarm
2. monetdb create voc
3. monetdb release voc
4.mclient -u monetdb -d voc
### Actual Results:
mclient -u monetdb -d voc
password: <monetdb>
monetdbd: internal error while starting mserver, please refer to the logs
### Expected Results:
access the database without error.
Entoire contents of log:
cat /usr/local/monetdb/var/log/monetdb/merovingian.log
2014-03-01 13:38:10 MSG merovingian[91347]: Merovingian 1.7 (Jan2014) starting
2014-03-01 13:38:10 MSG merovingian[91347]: monitoring dbfarm /usr/local/monetdb/var/monetdb5/dbfarm
2014-03-01 13:38:10 MSG merovingian[91347]: accepting connections on TCP socket 10.0.2.6:50000
2014-03-01 13:38:10 MSG merovingian[91347]: accepting connections on UNIX domain socket /tmp/.s.monetdb.50000
2014-03-01 13:38:10 MSG discovery[91347]: listening for UDP messages on 0.0.0.0:50000
2014-03-01 13:38:10 MSG control[91347]: accepting connections on UNIX domain socket /tmp/.s.merovingian.50000
2014-03-01 13:38:15 MSG control[91347]: (local): created database 'voc'
2014-03-01 13:38:27 MSG control[91347]: (local): served status list
2014-03-01 13:38:27 MSG control[91347]: (local): released database 'voc'
2014-03-01 13:38:38 MSG merovingian[91347]: starting database 'voc', up min/avg/max: 0s/0s/0s, crash average: 0.00 0.00 0.00 (0-0=0)
2014-03-01 13:38:38 MSG voc[91351]: arguments: /usr/local/monetdb/bin/mserver5 --dbpath=/usr/local/monetdb/var/monetdb5/dbfarm/voc --set merovingian_uri=mapi:monetdb://monetdb_jail:50000/voc --set mapi_open=false --set mapi_port=0 --set mapi_usock=/usr/local/monetdb/var/monetdb5/dbfarm/voc/.mapi.sock --set monet_vault_key=/usr/local/monetdb/var/monetdb5/dbfarm/voc/.vaultkey --set gdk_nr_threads=12 --set max_clients=64 --set sql_optimizer=default_pipe --set monet_daemon=yes
2014-03-01 13:38:38 MSG merovingian[91347]: database 'voc' (91351) has exited with exit status 1
2014-03-01 13:38:38 ERR merovingian[91347]: client error: database 'voc' appears to shut itself down after starting, check monetdbd's logfile for possible hints
2014-03-01 13:38:38 MSG merovingian[91347]: starting database 'voc', up min/avg/max: 0s/0s/0s, crash average: 0.00 0.00 0.00 (0-0=0)
2014-03-01 13:38:38 MSG voc[91352]: arguments: /usr/local/monetdb/bin/mserver5 --dbpath=/usr/local/monetdb/var/monetdb5/dbfarm/voc --set merovingian_uri=mapi:monetdb://monetdb_jail:50000/voc --set mapi_open=false --set mapi_port=0 --set mapi_usock=/usr/local/monetdb/var/monetdb5/dbfarm/voc/.mapi.sock --set monet_vault_key=/usr/local/monetdb/var/monetdb5/dbfarm/voc/.vaultkey --set gdk_nr_threads=12 --set max_clients=64 --set sql_optimizer=default_pipe --set monet_daemon=yes
2014-03-01 13:38:38 MSG merovingian[91347]: database 'voc' (91352) has exited with exit status 1
2014-03-01 13:38:39 ERR merovingian[91347]: client error: database 'voc' appears to shut itself down after starting, check monetdbd's logfile for possible hints
## Comment 19654
Date: 2014-03-01 14:56:53 +0100
From: Jim Ohlstein <<jim>>
Follow-up:
Downloaded package does start up as expected on FreeBSD 8.4 amd64 and FreeBSD 9.2 amd64.
Compiled version of 11.17.9 has same crash on FreeBSD 10.0 amd64. Same error shows if compiled with clang 3.3 or gcc46.
Previous version (11.15.19) compiles and runs on FreeBSD 10.0 amd64.
Will attache more detailed logs shortly.
## Comment 19655
Date: 2014-03-01 15:03:49 +0100
From: Jim Ohlstein <<jim>>
Created attachment 272
log file
> Attached file: [merovingian.log](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3449_merovingian.log_272) (text/plain, 225592 bytes)
> Description: log file
## Comment 19656
Date: 2014-03-03 10:46:55 +0100
From: @drstmane
Jim,
do I understand the following correctly:
- after installing MonetDB-11.17.9-x86_64-FreeBSD-8.2-bin.tar.bz2 on a "virgin" (wrt. MonetDB, i.e., no old versions or databases left behind) mserver5 starts fine (on a new virgin database) on FreeBSD 8.4 & 9.2 but fails as reported on FreeBSD 10.0
- compiling both 11.15.19 (Feb2013-SP6) and 11.17.9 (Jan2014) from source tarballs works fine on FreeBSD10.0, however, starting the server on a new virgin database only works fine with 11.15.19 (Feb2013-SP6), but fails as reported with 11.17.9 (Jan2014)
?
Thanks!
Stefan
## Comment 19658
Date: 2014-03-03 12:12:01 +0100
From: Jim Ohlstein <<jim>>
- after installing MonetDB-11.17.9-x86_64-FreeBSD-8.2-bin.tar.bz2 on a "virgin" (wrt. MonetDB, i.e., no old versions or databases left behind) mserver5 starts fine (on a new virgin database) on FreeBSD 8.4 & 9.2 but fails as reported on FreeBSD 10.0
Yes. These tests were done on VirtualBox VM's for 8.4 and 9.2. For 10.0 they were done in a fresh jail on a 10.0-STABLE amd64 machine (r260997) and confirmed in a VirtualBOX VM running 10-0 RELEASE with a "GENERIC" kernel (FreeBSD freebsd-10.0-amd64 10.0-RELEASE FreeBSD 10.0-RELEASE 0 r260789: Thu Jan 16 22:34:59 UTC 2014 root@snap.freebsd.org:/usr/obj/usr/src/sys/GENERIC amd64)
- compiling both 11.15.19 (Feb2013-SP6) and 11.17.9 (Jan2014) from source tarballs works fine on FreeBSD10.0, however, starting the server on a new virgin database only works fine with 11.15.19 (Feb2013-SP6), but fails as reported with 11.17.9 (Jan2014)
Correct. These tests were also done in fresh jails. I only tested these on FreeBSD 10-STABLE amd64.
I did patch common/utils/mutils.c on BOTH versions to force compilation, without which an error is thrown during compilation:
libtool: link: cc -g -O2 -D_REENTRANT -o .libs/mserver5 mserver5.o monet_version.o -pthread -Wl,--export-dynamic ../../monetdb5/tools/.libs/libmonetdb5.so -L/usr/lib -L/usr/local/lib /root/MonetDB-11.15.19/gdk/.libs/libbat.so /root/MonetDB-11.15.19/clients/mapilib/.libs/libmapi.so -lreadline /usr/local/lib/libxml2.so -lz -llzma ../../gdk/.libs/libbat.so /root/MonetDB-11.15.19/common/stream/.libs/libstream.so /usr/local/lib/libiconv.so -lm -lbz2 -lkvm /usr/local/lib/libpcre.so -lssl -lcrypto -lumem -pthread -Wl,-rpath -Wl,/usr/local/lib
/root/MonetDB-11.15.19/gdk/.libs/libbat.so: undefined reference to `backtrace_symbols'
/root/MonetDB-11.15.19/gdk/.libs/libbat.so: undefined reference to `backtrace'
cc: error: linker command failed with exit code 1 (use -v to see invocation)
gmake[5]: *** [mserver5] Error 1
The same error is seen on both versions and the same patch allowed compilation of both versions.
The patch, as follows, had no effect on function of compiled versions of 11.15.19.
cat patch-common-utils-mutils-c
--- common/utils/mutils.c.orig 2014-02-02 18:51:49.389290318 +0100
+++ common/utils/mutils.c 2014-02-02 18:55:21.300389407 +0100
@@ -315,26 +315,7 @@
ifdef HAVE_EXECINFO_H
-/* Obtain a backtrace and print it to stdout. */
-void
-print_trace (void)
-{
- void *array[10];
- size_t size;
- char **strings;
- size_t i;
-
- size = backtrace (array, 10);
- strings = backtrace_symbols (array, size);
-
- printf ("Obtained " SZFMT " stack frames.\n", size);
-
- for (i = 0; i < size; i++)
- printf ("%s\n", strings[i]);
-
- free (strings);
-}
-else
+/* Backtrace not enabled. */
void
print_trace(void)
{
## Comment 19770
Date: 2014-04-24 15:44:33 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [de2ffe1f09ba](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=de2ffe1f09ba) 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=de2ffe1f09ba](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=de2ffe1f09ba)
Changeset description:
Check for existence of backtrace function.
This should fix the compilation problem noted in bug #3449.
## Comment 21126
Date: 2015-08-12 14:19:47 +0200
From: @sjoerdmullender
Closing since we now run mserver5 on FreeBSD 10.1 ourselved, and that seems to work.
Note, the distributed binaries are still built on an ancient FreeBSD system.
## Comment 21180
Date: 2015-08-28 13:41:36 +0200
From: @sjoerdmullender
Jul2015 has been released.
| mserver crash on start - Freebsd 10 amd64 | https://api.github.com/repos/MonetDB/MonetDB/issues/3449/comments | 0 | 2020-11-30T12:38:18Z | 2024-06-27T12:02:38Z | https://github.com/MonetDB/MonetDB/issues/3449 | 753,444,549 | 3,449 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-27 17:45:14 +0100
From: Jim Ohlstein <<jim>>
To: buildtools devs <<bugs-buildtools>>
Version: -- development
Last updated: 2019-04-30 12:36:04 +0200
## Comment 19644
Date: 2014-02-27 17:45:14 +0100
From: Jim Ohlstein <<jim>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/33.0.1750.117 Safari/537.36
Build Identifier:
If libatomic_ops is installed, configure defaults to using it even if it is not desired (perhaps for testing, etc). There is no way to turn this off
Reproducible: Always
### Steps to Reproduce:
1.Install libatomic_ops
2.Run ./configure
### Actual Results:
libatomic_ops is enabled by defauls
### Expected Results:
The same, but I'd like there to be a configure option to turn it off.
Will port proposed patches shortly
## Comment 19648
Date: 2014-02-27 17:58:04 +0100
From: Jim Ohlstein <<jim>>
Created attachment 269
This patch is for configure.ag and works if bootstrap is used
> Attached file: [patch-configure-ag](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3448_patch-configure-ag_269) (text/plain, 1713 bytes)
> Description: This patch is for configure.ag and works if bootstrap is used
## Comment 19649
Date: 2014-02-27 17:59:29 +0100
From: Jim Ohlstein <<jim>>
Created attachment 270
This patch is for configure.ac and can be applied directly to source tarball download
> Attached file: [patch-configure-ac](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3448_patch-configure-ac_270) (text/plain, 1667 bytes)
> Description: This patch is for configure.ac and can be applied directly to source tarball download
## Comment 19650
Date: 2014-02-27 18:00:16 +0100
From: Jim Ohlstein <<jim>>
Created attachment 271
This patch is for configure and can be applied directly to source tarball download
> Attached file: [patch-configure](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3448_patch-configure_271) (text/plain, 4184 bytes)
> Description: This patch is for configure and can be applied directly to source tarball download
## Comment 26830
Date: 2019-01-22 10:16:02 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [d7810a904660](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d7810a904660) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=d7810a904660](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=d7810a904660)
Changeset description:
Implemented --with(out)-atomic-ops configure option.
This fixes bug #3448.
| Make lbatomic_ops optional if installed | https://api.github.com/repos/MonetDB/MonetDB/issues/3448/comments | 0 | 2020-11-30T12:38:16Z | 2024-06-27T12:02:37Z | https://github.com/MonetDB/MonetDB/issues/3448 | 753,444,526 | 3,448 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-27 17:38:47 +0100
From: Jim Ohlstein <<jim>>
To: buildtools devs <<bugs-buildtools>>
Version: 11.17.9 (Jan2014)
CC: @drstmane
Last updated: 2014-02-27 18:05:31 +0100
## Comment 19643
Date: 2014-02-27 17:38:47 +0100
From: Jim Ohlstein <<jim>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/33.0.1750.117 Safari/537.36
Build Identifier:
If libgeos.so is present, configure ignores "--disable-geom" when passed to it.
Reproducible: Always
### Steps to Reproduce:
1.Install geos
2.run ./configure --disable-geom
3.
### Actual Results:
Available features/extensions:
...
curl = yes
geos = yes
getaddrinfo = yes
...
### Expected Results:
Available features/extensions:
...
curl = yes
geos = no
getaddrinfo = yes
...
## Comment 19645
Date: 2014-02-27 17:50:15 +0100
From: @drstmane
While possibly misleading at first glance,
this is correct and intended behavior:
Even with --disable-geom, the general availability of the geos library on the given system is not altered.
Thus, configure in either case correctly reports that it did find the geos library.
However, configure also correctly reports that the geom module (component) is enabled or disabled as requested:
.../configure --enable-geom
[...]
* Enabled/disabled components:
[...]
geom is enabled
[...]
vs.
.../configure --disable-geom
[...]
* Enabled/disabled components:
[...]
geom is disabled (by command line option)
[...]
## Comment 19646
Date: 2014-02-27 17:56:40 +0100
From: @drstmane
Please also compare
(default)
.../configure [--with-goes[=auto]]
[...]
* Enabled/disabled components:
[...]
geom is enabled
[...]
* Available features/extensions:
[...]
geos = yes
[...]
vs.
.../configure --withou-goes (or --with-goes=no)
[...]
* Enabled/disabled components:
[...]
geom is disabled (geos library required for geom module)
[...]
* Available features/extensions:
[...]
geos = no (by command line option)
[...]
## Comment 19647
Date: 2014-02-27 17:58:00 +0100
From: @drstmane
Sorry, without typos it's
"--with-geos"
and
"--without-geos"
## Comment 19651
Date: 2014-02-27 18:05:31 +0100
From: Jim Ohlstein <<jim>>
Thank you. Sorry for the noise
| --disable-geom has no effect | https://api.github.com/repos/MonetDB/MonetDB/issues/3447/comments | 0 | 2020-11-30T12:38:13Z | 2024-06-28T07:19:43Z | https://github.com/MonetDB/MonetDB/issues/3447 | 753,444,497 | 3,447 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-27 12:11:44 +0100
From: @swingbit
To: SQL devs <<bugs-sql>>
Version: 11.15.19 (Feb2013-SP6)
CC: @njnes
Last updated: 2014-05-22 09:52:26 +0200
## Comment 19642
Date: 2014-02-27 12:11:44 +0100
From: @swingbit
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/32.0.1700.102 Safari/537.36
Build Identifier:
create table t(a int);
insert into t values (1),(3),(2),(3),(3),(3);
select * from t;
-- remove the 3s
delete from t where a = 3;
select * from t;
-- Oops, the 3s are back when the table is set read-only
alter table t set read only;
select * from t;
-- Oops, the 3s are gone again when the table is read/write again
alter table t set read write;
select * from t;
Reproducible: Always
$ mserver5 --version
MonetDB 5 server v11.15.20 (64-bit, 64-bit oids)
This is an unreleased version
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2013 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: 8.33 2013-05-28 (compiled with 8.32)
openssl: OpenSSL 1.0.1e 11 Feb 2013 (compiled with OpenSSL 1.0.1e-fips 11 Feb 2013)
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 19652
Date: 2014-02-28 11:22:10 +0100
From: @swingbit
For completeness: this is not just related to deletions, but to all updates
## Comment 19707
Date: 2014-03-26 11:35:55 +0100
From: @njnes
deletes are now properly kept even with read only. When there are updates the set read only will now fail, unless the updates have be flushed (ie are consumed by the base/central columns etc).
## Comment 19708
Date: 2014-03-26 11:48:58 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [5e7f1c747884](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=5e7f1c747884) 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=5e7f1c747884](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=5e7f1c747884)
Changeset description:
fixed bug #3446, ie readonly needs to use the 'tid' column and only
set to read only when there are no (unflushed) updates.
| SET READ ONLY forgets previous changes | https://api.github.com/repos/MonetDB/MonetDB/issues/3446/comments | 0 | 2020-11-30T12:38:11Z | 2024-06-27T12:02:35Z | https://github.com/MonetDB/MonetDB/issues/3446 | 753,444,470 | 3,446 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-24 07:19:54 +0100
From: Patrick TJ McPhee <<ptjm>>
To: clients devs <<bugs-clients>>
Version: -- development
CC: @hannesmuehleisen
Last updated: 2016-12-21 13:07:48 +0100
## Comment 19638
Date: 2014-02-24 07:19:54 +0100
From: Patrick TJ McPhee <<ptjm>>
Created attachment 268
Patch to mclient, stethoscope, tomograph to allow database to be specified in dotmonetdb file
The attached patch adds support for specifying the database name in .monetdb with the syntax:
database=mydbname
This is convenient in situations where a user typically connects to the same database all the time.
I list the OS as FreeBSD/amd64 above, but I've used this change for the past few releases under Linux i386, FreeBSD i386 and amd64, and NetBSD amd64.
> Attached file: [dotmonetdb-database.patch](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3445_dotmonetdb-database.patch_268) (text/plain, 3468 bytes)
> Description: Patch to mclient, stethoscope, tomograph to allow database to be specified in dotmonetdb file
## Comment 19640
Date: 2014-02-24 14:49:40 +0100
From: @hannesmuehleisen
+1
## Comment 19641
Date: 2014-02-24 17:04:49 +0100
From: @sjoerdmullender
I haven't tried the patch yet, but it looks pretty good. There is, however, one thing missing.
Can you also fix the manual page. At least mclient.1 needs to be patched, possibly more.
## Comment 19684
Date: 2014-03-08 18:43:52 +0100
From: Patrick TJ McPhee <<ptjm>>
Created attachment 277
Patch to mclient.1
Additional patch which updates mclient.1. This just adds database to the list of parameters read from DOTMONETDB. I put it after language because that's where it appears in the parameter list.
> Attached file: [mclient1.patch](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3445_mclient1.patch_277) (text/plain, 652 bytes)
> Description: Patch to mclient.1
## Comment 19685
Date: 2014-03-08 19:01:13 +0100
From: Patrick TJ McPhee <<ptjm>>
I've added another patch which adds database to the list of parameters in mclient.1. I looked for other files which reference DOTMONETDB (searching for .monetdb turns up too man hits due to www.monetdb.org occurring in most files), and turned up MTest.py and process.py, but that appears to be using the file rather than testing the interface.
## Comment 19844
Date: 2014-06-17 06:52:44 +0200
From: Patrick TJ McPhee <<ptjm>>
Created attachment 287
Patch against release 11.17.17
This is a replacement patch that applies against the most recent release tarball.
> Attached file: [dotmonetdb-11.17.17.patch](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3445_dotmonetdb-11.17.17.patch_287) (text/plain, 6195 bytes)
> Description: Patch against release 11.17.17
## Comment 21255
Date: 2015-09-08 18:20:51 +0200
From: Patrick TJ McPhee <<ptjm>>
Created attachment 350
Patch against release 11.21.5
This is a patch against the most recent release version (the previous patch worked against all the releases between then and now). There's no change in functionality.
> Attached file: [dotmonetdb-11.21.5.patch](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3445_dotmonetdb-11.21.5.patch_350) (text/plain, 5766 bytes)
> Description: Patch against release 11.21.5
## Comment 21256
Date: 2015-09-08 18:22:20 +0200
From: Patrick TJ McPhee <<ptjm>>
Comment on attachment 287
Patch against release 11.17.17
This is obsolete as of release 11.21.5
## Comment 24528
Date: 2016-10-18 06:03:50 +0200
From: Patrick TJ McPhee <<ptjm>>
Created attachment 481
Patch against release 11.23.13
Functionality is unchanged. This patch applies against the most recent release.
> Attached file: [dotmonetdb-11.23.13.patch](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3445_dotmonetdb-11.23.13.patch_481) (text/plain, 5788 bytes)
> Description: Patch against release 11.23.13
## Comment 24529
Date: 2016-10-18 11:20:53 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [e9ea484670f4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e9ea484670f4) 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=e9ea484670f4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=e9ea484670f4)
Changeset description:
Add database option to .monetdb file.
This fixes bug #3445.
Thanks to Patrick McPhee for the suggestion and the code.
## Comment 24530
Date: 2016-10-18 11:22:09 +0200
From: @sjoerdmullender
This will be in the next feature release.
| Add support for database name to dotmonetdb file | https://api.github.com/repos/MonetDB/MonetDB/issues/3445/comments | 0 | 2020-11-30T12:38:08Z | 2024-06-27T12:02:34Z | https://github.com/MonetDB/MonetDB/issues/3445 | 753,444,439 | 3,445 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-21 14:43:51 +0100
From: janmartijn
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: @njnes
Last updated: 2014-03-11 09:19:33 +0100
## Comment 19632
Date: 2014-02-21 14:43:51 +0100
From: janmartijn
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:28.0) Gecko/20100101 Firefox/28.0
Build Identifier:
The query with a LEFT OUTER JOIN of two tables has problems with the AND clause after the ON clause. After starting the query the monetdb server will crash.
Reproducible: Always
### Steps to Reproduce:
1.Startup client in windows 7 connecting to local database
2.Entering query:
DROP TABLE table1;
DROP TABLE table2;
CREATE TABLE "sys"."table1" (
"taskid" INTEGER,
"cid" INTEGER);
CREATE TABLE "sys"."table2" (
"taskid" INTEGER,
"cname" VARCHAR(255));
SELECT * FROM table1 t1
LEFT OUTER JOIN table2 t2
ON (t1.taskid = t2.taskid)
AND (
t1.cid IS NULL OR
t1.cid = t2.taskid OR
t2.cname = 'test'
);
### Actual Results:
client reports : Connection terminated
### Expected Results:
With the WHERE clause same result will be reached in this case i think.
DROP TABLE table1;
DROP TABLE table2;
CREATE TABLE "sys"."table1" (
"taskid" INTEGER,
"cid" INTEGER);
CREATE TABLE "sys"."table2" (
"taskid" INTEGER,
"cname" VARCHAR(255));
SELECT * FROM table1 t1
LEFT OUTER JOIN table2 t2
ON (t1.taskid = t2.taskid)
WHERE (
t1.cid IS NULL OR
t1.cid = t2.taskid OR
t2.cname = 'test'
);
Also when you remove "OR t2.cname = 'test'" from the AND clause of the problem query it will not cause a crash
MonetDB 5 server v11.15.19 "Feb2013-SP6"
Serving database 'demo', using 8 threads
Compiled for x86_64-pc-winnt/64bit with 64bit OIDs dynamically linke
Found 7.874 GiB available main-memory.
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2013 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/JAQL module loaded
Welcome to mclient, the MonetDB/SQL interactive terminal (Feb2013-SP6)
Database: MonetDB v11.15.19 (Feb2013-SP6), 'demo'
MonetDB/SQL module loaded
os: Windows 7 uptodate
## Comment 19635
Date: 2014-02-21 17:16:04 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [1784cbdea0c8](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=1784cbdea0c8) 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=1784cbdea0c8](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=1784cbdea0c8)
Changeset description:
Test for bug #3444.
## Comment 19636
Date: 2014-02-21 19:34:39 +0100
From: @njnes
fix bug in handling the combination of OUTER and OR expressions. Normal expression are pushed down very early, we need to prevent this on OUTER/OR expressions.
## Comment 19637
Date: 2014-02-21 21:50:26 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [4b6b92785afa](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=4b6b92785afa) 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=4b6b92785afa](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=4b6b92785afa)
Changeset description:
fixed bug #3444,ie handle or and outer properly
## Comment 19694
Date: 2014-03-11 09:19:33 +0100
From: @sjoerdmullender
Jan2014-SP1 has been released.
| AND after ON () of LEFT OUTER JOIN with certain expressions will cause crash | https://api.github.com/repos/MonetDB/MonetDB/issues/3444/comments | 0 | 2020-11-30T12:38:06Z | 2024-06-27T12:02:33Z | https://github.com/MonetDB/MonetDB/issues/3444 | 753,444,398 | 3,444 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-21 13:48:04 +0100
From: @hannesmuehleisen
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: ajdamico
Last updated: 2014-03-11 09:19:36 +0100
## Comment 19629
Date: 2014-02-21 13:48:04 +0100
From: @hannesmuehleisen
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/33.0.1750.112 Safari/537.36
Build Identifier:
Dropping an index from a table crashes the server.
Reproducible: Always
### Steps to Reproduce:
1. Create a fairly large table (confirmed with ~200000 entries) with a single integer column
2. Create an index on the table
3. Drop the index again
### Actual Results:
>!FATAL: 40000!COMMIT: transation commit failed (perhaps your disk is full?) exiting (kernel error: !ERROR: BATsubselect: invalid argument: b must have a dense head.
### Expected Results:
No crash and index is gone.
## Comment 19630
Date: 2014-02-21 13:48:46 +0100
From: @hannesmuehleisen
Created attachment 267
SQL script that provokes the issue
> Attached file: [crash.sql.bz2](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3443_crash.sql.bz2_267) (application/x-bzip, 359611 bytes)
> Description: SQL script that provokes the issue
## Comment 19631
Date: 2014-02-21 14:17:21 +0100
From: @sjoerdmullender
The logger uses BUNdelete. We can't have that in "headless" code.
## Comment 19633
Date: 2014-02-21 16:21:58 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [ef0df6f6f964](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ef0df6f6f964) 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=ef0df6f6f964](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ef0df6f6f964)
Changeset description:
Revert back to using legacy functions in log_tend.
This fixes bug #3443.
## Comment 19634
Date: 2014-02-21 17:06:11 +0100
From: @sjoerdmullender
Fixed, but we still need a test.
## Comment 19696
Date: 2014-03-11 09:19:36 +0100
From: @sjoerdmullender
Jan2014-SP1 has been released.
| DROP INDEX crashes server with BATsubselect: invalid argument: b must have a dense head | https://api.github.com/repos/MonetDB/MonetDB/issues/3443/comments | 0 | 2020-11-30T12:38:03Z | 2024-06-27T12:02:32Z | https://github.com/MonetDB/MonetDB/issues/3443 | 753,444,355 | 3,443 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-20 11:15:42 +0100
From: @sjoerdmullender
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: @njnes
Last updated: 2014-03-11 09:19:38 +0100
## Comment 19592
Date: 2014-02-20 11:15:42 +0100
From: @sjoerdmullender
When using COPY INTO ... LOCKED on a non-empty table, the server reports a count which is one too high for the number of inserted values. The actual number inserted is correct, only the reporting is off.
## Comment 19627
Date: 2014-02-21 09:51:08 +0100
From: @njnes
We now return the correct slice (ie with the correct row count)
## Comment 19697
Date: 2014-03-11 09:19:38 +0100
From: @sjoerdmullender
Jan2014-SP1 has been released.
| COPY INTO ... LOCKED reports incorrect count | https://api.github.com/repos/MonetDB/MonetDB/issues/3442/comments | 0 | 2020-11-30T12:38:00Z | 2024-06-27T12:02:31Z | https://github.com/MonetDB/MonetDB/issues/3442 | 753,444,329 | 3,442 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-19 15:55:46 +0100
From: @mlkersten
To: SQL devs <<bugs-sql>>
Version: -- development
Last updated: 2016-04-11 11:45:20 +0200
## Comment 19589
Date: 2014-02-19 15:55:46 +0100
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:26.0) Gecko/20100101 Firefox/26.0
Build Identifier:
PG compatibility feature request, allow explicit exclusion of TIME ZONE in temporal type
Reproducible: Always
### Steps to Reproduce:
start transaction;
create table zones(
t timestamp,
tz timestamp with time zone,
tn timestamp without time zone
);
abort;
| Explicit WITHOUT TIME ZONE | https://api.github.com/repos/MonetDB/MonetDB/issues/3441/comments | 0 | 2020-11-30T12:37:58Z | 2024-06-28T13:40:18Z | https://github.com/MonetDB/MonetDB/issues/3441 | 753,444,297 | 3,441 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-19 15:46:41 +0100
From: @mlkersten
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: @njnes
Last updated: 2015-08-28 13:42:53 +0200
## Comment 19588
Date: 2014-02-19 15:46:41 +0100
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:26.0) Gecko/20100101 Firefox/26.0
Build Identifier:
Unexpected syntax error while the language would allow it.
Reproducible: Always
### Steps to Reproduce:
CREATE SEQUENCE ancilfiles_aid_seq AS INTEGER
START WITH 1
INCREMENT BY 1
NO MAXVALUE
NO MINVALUE;
### Actual Results:
sql>CREATE SEQUENCE ancilfiles_aid_seq AS INTEGER
more> START WITH 1
more> INCREMENT BY 1
more> NO MAXVALUE
more> NO MINVALUE;
syntax error, unexpected NOMINVALUE, expecting SCOLON in: "create sequence ancilfiles_aid_seq as integer
start with 1
increment by 1
no maxvalue
no minvalue"
sql>
### Expected Results:
sql>CREATE SEQUENCE ancilfiles_aid_seq AS INTEGER
more> START WITH 1
more> INCREMENT BY 1
more> NO MAXVALUE
more> NO MINVALUE;
## Comment 19628
Date: 2014-02-21 10:18:21 +0100
From: @njnes
2003 sql definition is indeed a lot more flexible then our current parser allows. It now has a fixed order of options while all (except the AS type) can be in any order.
## Comment 21107
Date: 2015-08-08 17:05:58 +0200
From: @njnes
improved the handling of sequence number creation. This will be available in the release after the Jul2015 one.
## Comment 21220
Date: 2015-08-28 13:42:53 +0200
From: @sjoerdmullender
Jul2015 has been released.
| Sequence type errors | https://api.github.com/repos/MonetDB/MonetDB/issues/3440/comments | 0 | 2020-11-30T12:37:55Z | 2024-06-27T12:02:30Z | https://github.com/MonetDB/MonetDB/issues/3440 | 753,444,268 | 3,440 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-18 17:37:28 +0100
From: Alex Shestakov <<shura.shestakov>>
To: clients devs <<bugs-clients>>
Version: 11.17.9 (Jan2014)
CC: shura.shestakov
Last updated: 2014-05-22 09:52:20 +0200
## Comment 19586
Date: 2014-02-18 17:37:28 +0100
From: Alex Shestakov <<shura.shestakov>>
Created attachment 266
fix_milliseconds
Python driver drops milliseconds from temporal data when reading from the database.
The attached patch fixes the problem for me.
> Attached file: [monetdb_python.patch](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3439_monetdb_python.patch_266) (application/octet-stream, 2537 bytes)
> Description: fix_milliseconds
## Comment 19771
Date: 2014-04-24 17:09:47 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [f2bbeac41f80](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=f2bbeac41f80) 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=f2bbeac41f80](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=f2bbeac41f80)
Changeset description:
Keep second fraction in Python driver.
This fixes bug #3439, albeit in a different way than the patch in the
bug report.
## Comment 19772
Date: 2014-04-24 17:17:26 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [a1c694f27eeb](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=a1c694f27eeb) 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=a1c694f27eeb](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=a1c694f27eeb)
Changeset description:
Added test for bug #3439, fixed typo in fix for same.
## Comment 19774
Date: 2014-04-24 18:45:06 +0200
From: Alex Shestakov <<shura.shestakov>>
The original patch also fixed a bug with parsing timestamps having negative timezone offsets (in _extract_timezone() function).
The new fix doesn't do it.
Also in the original patch I replaced python parsing code using string.split() with datettime.strptime() assuming that it will work significantly faster.
## Comment 19775
Date: 2014-04-24 19:25:37 +0200
From: @sjoerdmullender
(In reply to comment 3)
> The original patch also fixed a bug with parsing timestamps having negative
> timezone offsets (in _extract_timezone() function).
Now also fixed.
> The new fix doesn't do it.
>
> Also in the original patch I replaced python parsing code using
> string.split() with datettime.strptime() assuming that it will work
> significantly faster.
I'll need to look into that claim.
## Comment 19778
Date: 2014-04-24 22:10:34 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [4bd5465d5ced](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=4bd5465d5ced) 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=4bd5465d5ced](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=4bd5465d5ced)
Changeset description:
Redid fix for bug #3439, much more like attached patch.
The patch wasn't entirely correct (times don't have fractions), and
not complete (also did date in the same way).
The advantage of this patch is that it actually simplifies the code,
something I didn't appreciate earlier.
The (slight) disadvantage is that it expects the format of the time
zone to be exactly [+-]HH:MM, unlike the original which was slightly
more liberal (in intent, anyway -- there was a bug in it) but I guess
that's OK.
| Python driver drops milliseconds from timestamps | https://api.github.com/repos/MonetDB/MonetDB/issues/3439/comments | 0 | 2020-11-30T12:37:52Z | 2024-06-27T12:02:29Z | https://github.com/MonetDB/MonetDB/issues/3439 | 753,444,229 | 3,439 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-18 14:08:40 +0100
From: movingon
To: GDK devs <<bugs-common>>
Version: 11.17.9 (Jan2014)
CC: @njnes
Last updated: 2014-09-01 04:58:34 +0200
## Comment 19585
Date: 2014-02-18 14:08:40 +0100
From: movingon
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/27.0.1453.110 Safari/537.36 CoolNovo/2.0.9.20
Build Identifier:
error when using the substring function.
(in case utf8 data)
Reproducible: Sometimes
### Steps to Reproduce:
1. sample table create
--------------------------
CREATE TABLE TestTable(Column1 varchar(24), Column2 varchar(22), Column3 BIGINT);
2. data insert:
-> 200,000 row data (Repeated data).
-> first column data: UTF8
----------------------------------
insert into aaa values ('한글', '20140201', 100);
insert into aaa values ('잘 처리해 주세요 ^^', '20140202', 100);
insert into aaa values ('감사합니다 ~~.', '20140203', 100);
...
3. SELECT Column1, SUBSTRING(Column2, 1, 4) AS Year_of_Column2, SUM(Column3) AS Sum_of_Column3 FROM TestTable GROUP BY Column1, Year_of_Column2 ;
### Actual Results:
error: GDK reported error.
BATfetchjoin(tmp_433,tmp_534) does not hit always (|bn|=0 != 2=|l|) => can't use
fetchjoin.
### Expected Results:
SQL is performed successfully and multiple data output.
3. SELECT Column1, SUBSTRING(Column2, 1, 4) AS Year_of_Column2, SUM(Column3) AS Sum_of_Column3 FROM TestTable GROUP BY Column1, Year_of_Column2 ;
=> not use the substring function, performed successfully
3. SELECT Column1, Column2 AS Year_of_Column2, SUM(Column3) AS Sum_of_Column3 FROM TestTable GROUP BY Column1, Year_of_Column2 ;
## Comment 19892
Date: 2014-07-09 15:31:39 +0200
From: @njnes
could you try with the current released (Jan 2014 sp2) version?
## Comment 20110
Date: 2014-09-01 04:58:34 +0200
From: movingon
i'm try the released version (Jan 2014 sp2) .
SQL is performed successfully.
thank you very much ~~
have a nice day ^^.
| error when using the substring function. | https://api.github.com/repos/MonetDB/MonetDB/issues/3438/comments | 0 | 2020-11-30T12:37:50Z | 2024-06-27T12:02:28Z | https://github.com/MonetDB/MonetDB/issues/3438 | 753,444,203 | 3,438 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-10 13:18:20 +0100
From: @swingbit
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: @njnes
Last updated: 2015-05-21 15:01:54 +0200
## Comment 19571
Date: 2014-02-10 13:18:20 +0100
From: @swingbit
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/32.0.1700.102 Safari/537.36
Build Identifier:
Given the following table (~ 2M tuples - data is available but cannot post it here as it is too large):
sql>\d _dict
CREATE TABLE "spinque"."_dict" (
"idstr" CHARACTER LARGE OBJECT,
"id" INTEGER,
"type" CHARACTER LARGE OBJECT
);
and the following query:
select idstr, count(*) as c from _dict group by idstr having count(*) > 1;
It takes (Feb2013 SP6, Fedora 20, 16GB ram):
- almost 8 minutes with default optimizer pipe
- 14 seconds with no_mitosis_pipe.
I am aware that mitosis helps in many cases and also that its benefit are not trivial to predict in the general case. However, I run quite consistently in similar issues, for which mitosis is usually the cause.
The TRACE for both optimizer pipelines show that the default pipeline spends a ridiculous amount of time on mat.packIncrement() operations. I hope this gives a clue on how to improve this.
I attach both traces to this bug report.
Reproducible: Always
$ mserver5 --version
MonetDB 5 server v11.15.20 (64-bit, 64-bit oids)
This is an unreleased version
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2013 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: 8.33 2013-05-28 (compiled with 8.32)
openssl: OpenSSL 1.0.1e 11 Feb 2013 (compiled with OpenSSL 1.0.1e-fips 11 Feb 2013)
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 19572
Date: 2014-02-10 13:19:10 +0100
From: @swingbit
Created attachment 264
trace with default_pipe
> Attached file: [CORE-201_dict.trace](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3437_CORE-201_dict.trace_264) (text/plain, 32522 bytes)
> Description: trace with default_pipe
## Comment 19573
Date: 2014-02-10 13:19:48 +0100
From: @swingbit
Created attachment 265
trace with no_mitosis_pipe
> Attached file: [CORE-201_dict.trace_nomitosis](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3437_CORE-201_dict.trace_nomitosis_265) (text/plain, 13671 bytes)
> Description: trace with no_mitosis_pipe
## Comment 19574
Date: 2014-02-10 14:19:01 +0100
From: @swingbit
A couple more points about this, hoping that it helps finding the issue.
Besides the mat.packIncrement() time, I also notice that the total time spent on group.subgroupdone() is *one order of magnitude* higher in the default_pipe case.
$ grep subgroupdone CORE-201_dict.trace_nomitosis
| 14555360 | (X_17=<tmp_3766>[21908277],r1_17=<tmp_3775>[21908157],r2_17=<tmp_3576>[21908157]) := group.subgroupdone(X_16=<tmp_3744>[21908277]); |
$ grep subgroupdone CORE-201_dict.trace
| 1998838 | (X_106=<tmp_3766>[2738534],X_107=<tmp_3765>[2738534],X_108=<tmp_3713>[2738534]) := group.subgroupdone(X_98=<tmp_3705>[2738534]); |
| 2172497 | (X_126=<tmp_3714>[2738534],X_127=<tmp_3763>[2738534],X_128=<tmp_3777>[2738534]) := group.subgroupdone(X_103=<tmp_3776>[2738534]); |
| 2182481 | (X_130=<tmp_3524>[2738534],X_131=<tmp_3724>[2738490],X_132=<tmp_3757>[2738490]) := group.subgroupdone(X_104=<tmp_3651>[2738534]); |
| 2249392 | (X_110=<tmp_3620>[2738534],X_111=<tmp_3754>[2738534],X_112=<tmp_100>[2738534]) := group.subgroupdone(X_99=<tmp_3114>[2738534]); |
| 2283462 | (X_122=<tmp_3751>[2738534],X_123=<tmp_3771>[2738534],X_124=<tmp_3744>[2738534]) := group.subgroupdone(X_102=<tmp_3522>[2738534]); |
| 2289781 | (X_118=<tmp_701>[2738534],X_119=<tmp_3755>[2738534],X_120=<tmp_3775>[2738534]) := group.subgroupdone(X_101=<tmp_3546>[2738534]); |
| 2289672 | (X_114=<tmp_3716>[2738534],X_115=<tmp_3746>[2738534],X_116=<tmp_3576>[2738534]) := group.subgroupdone(X_100=<tmp_3730>[2738534]); |
| 2298757 | (X_134=<tmp_3737>[2738539],X_135=<tmp_3521>[2738539],X_136=<tmp_3773>[2738539]) := group.subgroupdone(X_105=<tmp_3774>[2738539]); |
| 96588639 | (X_12=<tmp_3773>[21908233],r1_17=<tmp_3757>[21908157],X_138=<tmp_3777>[21908157]) := group.subgroupdone(X_11=<tmp_3766>[21908233]); |
Notice that the last group.subgroupdone() above - which is performed on the result of the (expensive) mat.packIncrement() sequence - is by itself way more expensive than the single group.subgroupdone() of the no_mitosis_pipe case.
## Comment 19575
Date: 2014-02-10 16:03:43 +0100
From: @swingbit
In case it helps, a data dump to be used with the query above can be found (not indefinitely) here: http://www.spinque.com/tmp/bug-3437_data.sql.gz
## Comment 19579
Date: 2014-02-12 13:05:11 +0100
From: @njnes
Roberto
Basic assumption for the group by with mitosis is that we expect a limited amount of results. If this is not the case (which I expect for your queries) indeed its better to fallback to no mitosis. This is allready automatically done for primary key inserts/updates.
Current relational plans do not know if the group by result size is large or not
(except for the special case of primary key), ie its hard to automatically switch off mitosis.
## Comment 19580
Date: 2014-02-12 13:09:32 +0100
From: @njnes
besides the solution to automatically disable mitosis for these large result cases we could also try to switch to a value based partitioning solution.
## Comment 19581
Date: 2014-02-12 13:21:21 +0100
From: @swingbit
In this case the result of that aggregation had 120 tuples, so very small. Should that have worked well with mitosis then?
## Comment 19582
Date: 2014-02-12 13:24:59 +0100
From: @njnes
The intermediate per part group results must be much bigger than your 120 because else the mat appends wouldn't be expensive.
## Comment 19583
Date: 2014-02-12 13:28:34 +0100
From: @swingbit
That's true. The intermediates are indeed very large. This is an histogram on an almost-unique attribute.
It outputs almost 2M tuples with count=1 and 120 tuples with count=2. And then the HAVING clause keeps only the 120 tuples.
## Comment 20879
Date: 2015-05-21 15:01:54 +0200
From: @njnes
In default the problem isn't as big anymore. But still we should (oneday) implement some of the improvements (ie partition, handle a sorted case efficiently) for large group by results. Moved to a feature request ..
| Aggregation with mitosis horribly slow | https://api.github.com/repos/MonetDB/MonetDB/issues/3437/comments | 0 | 2020-11-30T12:37:46Z | 2024-06-28T07:19:43Z | https://github.com/MonetDB/MonetDB/issues/3437 | 753,444,162 | 3,437 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-06 12:27:51 +0100
From: Ken Leese <<ken.leese>>
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: @drstmane
Last updated: 2014-02-20 15:44:26 +0100
## Comment 19560
Date: 2014-02-06 12:27:51 +0100
From: Ken Leese <<ken.leese>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:26.0) Gecko/20100101 Firefox/26.0
Build Identifier:
If a file referenced by a COPY ... INTO statement is encoded in UTF-8 but contains a leading Byte Order Mark, the BOM is not ignored and the COPY statement succeeds.
However:
- A subsequent attempt to SELECT using MCLIENT fails.
- A subsequent attempt to SELECT using an ODBC client indicates data corruption
Reproducible: Always
### Steps to Reproduce:
Using MCLIENT:
CREATE TABLE "accent" ("City" VARCHAR(37) ,"Population" INTEGER);
COPY 2 RECORDS INTO "accent" FROM 'D:\cj1\save\accent.csv' DELIMITERS ',','\n','"' NULL AS '';
SELECT * FROM "accent";
### Actual Results:
sql>SELECT * FROM "accent";
+-------------+------------+
| City | Population |
+=============+============+
|2 tuples (0.554ms)
write error
### Expected Results:
Table rows should be displayed
This may reflect an error in COPY INTO, which should either skip the initial Byte Order Mark (preferable) or alternatively reject the erroneous data.
Additional error checking during the SELECT using MLCLIENT is also desirable.
## Comment 19561
Date: 2014-02-06 12:29:15 +0100
From: Ken Leese <<ken.leese>>
Created attachment 262
Attached file contains leading Windows BOM byte
> Attached file: [accent.csv](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3436_accent.csv_262) (application/octet-stream, 38 bytes)
> Description: Attached file contains leading Windows BOM byte
## Comment 19562
Date: 2014-02-06 12:34:04 +0100
From: Ken Leese <<ken.leese>>
Created attachment 263
Screen shot from ODBC command line debug tool
> Attached file: [LeadingBOM.png](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3436_LeadingBOM.png_263) (image/png, 49409 bytes)
> Description: Screen shot from ODBC command line debug tool
## Comment 19563
Date: 2014-02-06 12:47:56 +0100
From: Ken Leese <<ken.leese>>
Note that a LINUX or OS X installation of MonetDB might still be assaulted by a file containing a leading BOM.
## Comment 19564
Date: 2014-02-06 12:49:16 +0100
From: @drstmane
I guess, simply skipping/ignoring the BOM is no good option, as then remainder of the file might be interpreted using the wrong byte order.
Thus, we'd either categorically need to reject CSV files starting with a BOM, or interpret a BOM entirely as intended --- the latter might require considerable changes in the bulk loader code ...
## Comment 19565
Date: 2014-02-06 12:51:35 +0100
From: @drstmane
Well, having said that, if we restrict ourselves (and our bulk loader) to UTF-8, thing might be simpler, and ignoring a UTF-8 BOM ("EF BB BF") might be OK ...
## Comment 19566
Date: 2014-02-06 14:34:18 +0100
From: @sjoerdmullender
BOM markers don't make any sense in UTF-8 (see e.g. http://en.wikipedia.org/wiki/Byte_order_markUTF-8). There are no byte-order dependencies in UTF-8 since it is a byte sequence. BOM markers only make sense for the 16 bit encodings of Unicode.
Having said that, the UTF-8-encoded version of the BOM marker does occur in many UTF-8 files, so we should skip it. We do skip it in mclient, but if you specify a file in a COPY INTO command, the file is read directly by the server, and the server does (apparently) not skip the BOM marker.
## Comment 19567
Date: 2014-02-06 14:55:05 +0100
From: Ken Leese <<ken.leese>>
The Byte Order Mark sequence in Windows usually differentiates a file encoded using UTF-8 from a file that is encoded in a "native" Windows codepage. Windows codepage files usually do not contain the BOM sequence.
## Comment 19568
Date: 2014-02-06 15:01:28 +0100
From: @sjoerdmullender
The server always demands UTF-8, so it should just skip the BOM when it is present.
For mclient we can do one better. With mclient you can specify the encoding it should use, and it will then transparently translate to UTF-8 when sending data to the server. However, mclient could then use the presence of the BOM as an indication that the file is not in the encoding specified, but in UTF-8 instead.
But that is an enhancement. We should first fix the BOM in the server.
## Comment 19569
Date: 2014-02-06 16:19:07 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [f349cdd547dc](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=f349cdd547dc) 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=f349cdd547dc](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=f349cdd547dc)
Changeset description:
When opening a file for a stream, check for UTF-8 BOM.
When we find a BOM, we skip it, and we mark the stream as being
UTF-8. Then we can (and do) skip conversion with iconv, so that on
(e.g.) Windows you can run mclient with some encoding set, but still
read UTF-8 encoded files, as long as they start with the BOM.
This fixes bug #3436.
## Comment 19570
Date: 2014-02-06 16:59:24 +0100
From: @sjoerdmullender
Fixed.
The enhancement mentioned was also implemented.
## Comment 19587
Date: 2014-02-18 20:53:54 +0100
From: @drstmane
The test for this bug has been failing on Windows since it has been added; cf., e.g.,
http://monetdb.cwi.nl/testweb/web/testgrid.php?serial=50496:035eedf81126&order=platform,arch,compiler&targets=Cla-Fedora-x86_64-oid32,GNU-Darwin-i386-propcheck,GNU-Darwin-powerpc-propcheck,GNU-Darwin-x86_64-oid32,GNU-Fedora-x86_64-oid32-assert-propcheck,GNU-Fedora-x86_64-oid32-dbfarm,GNU-FreeBSD-x86_64,GNU-Gentoo-powerpc-assert-dbfarm,GNU-OpenIndiana-x86_64,GNU-Ubuntu-i386-assert-propcheck-dbfarm,Int-Fedora-x86_64-oid32-assert,Int-Fedora-x86_64-oid32-propcheck,Int-Windows7-x86_64-oid32-assert,Mic-Windows7-i386-installer&module=sql&tstlimit=sql/test/BugTracker-2014
I'm not sure whether this is purely due to our testing configuration,
or whether it might indicate that the fix for this bug is not (yet) working properly.
## Comment 19626
Date: 2014-02-20 15:44:26 +0100
From: @sjoerdmullender
The problem is that on Windows we open the file in "text" mode and so only see \n as line separators, but we specify \r\n as line separator, so that doesn't match. This problem has nothing to do with this bug, so I'm closing it.
(Also, I fixed the test so that we don't get into this issue.)
| COPY INTO from file containing leading Byte Order Mark (BOM) causes corruption | https://api.github.com/repos/MonetDB/MonetDB/issues/3436/comments | 0 | 2020-11-30T12:37:43Z | 2024-06-27T12:02:25Z | https://github.com/MonetDB/MonetDB/issues/3436 | 753,444,143 | 3,436 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-05 23:53:01 +0100
From: Anthony Damico <<ajdamico>>
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: ajdamico, @hannesmuehleisen, @njnes, @drstmane
Last updated: 2014-02-20 15:02:54 +0100
## Comment 19552
Date: 2014-02-05 23:53:01 +0100
From: Anthony Damico <<ajdamico>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:26.0) Gecko/20100101 Firefox/26.0
Build Identifier:
here's an example of a join that works until you create an index on the merge-variables. this behavior shouldn't happen, right? my apologies if i'm overlooking something :) thanks for monetdb!!
Reproducible: Always
### Steps to Reproduce:
CREATE TABLE TEST1 ( one VARCHAR(10) , two INT ) ;
CREATE TABLE TEST2 ( one VARCHAR(10) , two INT ) ;
INSERT INTO TEST1 VALUES ( 'a' , 1 ) ;
INSERT INTO TEST2 VALUES ( 'a' , 1 ) ;
SELECT count(*) FROM ( SELECT one , two FROM TEST1 ) AS a INNER JOIN ( SELECT one , two FROM TEST2 ) AS B on A.one = B.one AND A.two = B.two ;
CREATE INDEX onedex ON TEST1 ( one , two ) ;
SELECT count(*) FROM ( SELECT one , two FROM TEST1 ) AS a INNER JOIN ( SELECT one , two FROM TEST2 ) AS B on A.one = B.one AND A.two = B.two ;
### Actual Results:
sql>CREATE TABLE TEST1 ( one VARCHAR(10) , two INT ) ;
operation successful (179.671ms)
sql>CREATE TABLE TEST2 ( one VARCHAR(10) , two INT ) ;
operation successful (71.839ms)
sql>
sql>INSERT INTO TEST1 VALUES ( 'a' , 1 ) ;
1 affected row (26.440ms)
sql>INSERT INTO TEST2 VALUES ( 'a' , 1 ) ;
1 affected row (25.405ms)
sql>
sql>SELECT count(*) FROM ( SELECT one , two FROM TEST1 ) AS a INNER JOIN ( SELEC
T one , two FROM TEST2 ) AS B on A.one = B.one AND A.two = B.two ;
+------+
| L1 |
+======+
| 1 |
+------+
1 tuple (1.890ms)
sql>
sql>CREATE INDEX onedex ON TEST1 ( one , two ) ;
operation successful (62.781ms)
sql>
sql>SELECT count(*) FROM ( SELECT one , two FROM TEST1 ) AS a INNER JOIN ( SELEC
T one , two FROM TEST2 ) AS B on A.one = B.one AND A.two = B.two ;
+------+
| L1 |
+======+
| 0 |
+------+
1 tuple (1.916ms)
sql>
sql>
### Expected Results:
both before and after the INDEX gets set:
+------+
| L1 |
+======+
| 1 |
+------+
## Comment 19553
Date: 2014-02-06 08:23:55 +0100
From: @hannesmuehleisen
Also confirmed on Linux, OSX, and Jan2014 branch.
## Comment 19554
Date: 2014-02-06 08:25:48 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [eed2975df8f9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=eed2975df8f9) 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=eed2975df8f9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=eed2975df8f9)
Changeset description:
Test for bug #3435
## Comment 19555
Date: 2014-02-06 08:38:17 +0100
From: @hannesmuehleisen
Issue also occurs on tables with only integer columns.
## Comment 19556
Date: 2014-02-06 09:25:40 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [33353c5ddd48](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=33353c5ddd48) 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=33353c5ddd48](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=33353c5ddd48)
Changeset description:
Expected output for Bug #3435
## Comment 19557
Date: 2014-02-06 09:39:51 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [ebad38380185](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ebad38380185) 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=ebad38380185](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ebad38380185)
Changeset description:
Enable test for bug #3435.
## Comment 19558
Date: 2014-02-06 11:03:09 +0100
From: @drstmane
Created attachment 261
diff between "good" ("OK") and "bad" ("KO") plan, explain & trace
> Attached file: [OK-KO.html](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3435_OK-KO.html_261) (text/html, 95113 bytes)
> Description: diff between "good" ("OK") and "bad" ("KO") plan, explain & trace
## Comment 19559
Date: 2014-02-06 11:05:16 +0100
From: @drstmane
attached diff between "good" ("OK") and "bad" ("KO") plan, explain, and trace
suggests that in the "bad" case, the join between the index (hash) of table test1 and the created two-column hash of table test2 find no matches.
Possibly, both hashes are created differently and thus not "compatible"/"comparable"?
## Comment 19577
Date: 2014-02-12 10:34:41 +0100
From: @njnes
its seems the stored hash is incorrect.
## Comment 19578
Date: 2014-02-12 12:59:08 +0100
From: @njnes
we now handle the index creation similar as an alter table for a key creation.
Then a key/index is created and the proper update statement is added to update the index column.
## Comment 19584
Date: 2014-02-12 14:52:39 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [cc2f9c2ac8a2](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=cc2f9c2ac8a2) 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=cc2f9c2ac8a2](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=cc2f9c2ac8a2)
Changeset description:
fix for bug #3435. Indices created on tables already holding data
are properly initialized again.
fixed a bug when loading using LOCKED. The properties weren't properly checked
## Comment 19612
Date: 2014-02-20 15:02:54 +0100
From: @sjoerdmullender
Jan2014 has been released.
| INDEX prevents JOIN from discovering matches | https://api.github.com/repos/MonetDB/MonetDB/issues/3435/comments | 0 | 2020-11-30T12:37:40Z | 2024-06-27T12:02:24Z | https://github.com/MonetDB/MonetDB/issues/3435 | 753,444,112 | 3,435 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-05 20:09:45 +0100
From: @mlkersten
To: SQL devs <<bugs-sql>>
Version: -- development
CC: martin.van.dinther
Last updated: 2019-01-22 09:50:41 +0100
## Comment 19549
Date: 2014-02-05 20:09:45 +0100
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:26.0) Gecko/20100101 Firefox/26.0
Build Identifier:
CREATE [TEMPORARY] TABLE [IF NOT EXISTS]
is widely supported and handy for SQL scripting.
Should be considered for inclusion.
Reproducible: Always
## Comment 19767
Date: 2014-04-23 11:49:44 +0200
From: @sjoerdmullender
*** Bug #3475 has been marked as a duplicate of this bug. ***
## Comment 26829
Date: 2019-01-22 09:50:41 +0100
From: @sjoerdmullender
Available since the Jul2017 release.
| if-not-exist flag | https://api.github.com/repos/MonetDB/MonetDB/issues/3434/comments | 0 | 2020-11-30T12:37:38Z | 2024-06-27T12:02:23Z | https://github.com/MonetDB/MonetDB/issues/3434 | 753,444,084 | 3,434 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-05 09:58:22 +0100
From: @sjoerdmullender
To: GDK devs <<bugs-common>>
Version: 11.17.9 (Jan2014)
Last updated: 2014-02-24 14:38:11 +0100
## Comment 19538
Date: 2014-02-05 09:58:22 +0100
From: @sjoerdmullender
The SQL frontend uses the force flag to write to read-only bats. Writing to such a bat can cause it to be extended, and extending it can cause its heaps to be relocated (i.e. get a new address). If there are views on such a bat, the views don't get updated with the new address.
The test for bug #3429 causes a crash due to this problem.
## Comment 19639
Date: 2014-02-24 14:38:11 +0100
From: @sjoerdmullender
The SQL front end should avoid getting into this situation, and now it does.
This means we can close this bug.
| appending to bat can cause view to point into freed memory | https://api.github.com/repos/MonetDB/MonetDB/issues/3433/comments | 0 | 2020-11-30T12:37:35Z | 2024-06-27T12:02:22Z | https://github.com/MonetDB/MonetDB/issues/3433 | 753,444,061 | 3,433 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-04 22:30:02 +0100
From: Ken Leese <<ken.leese>>
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: @njnes
Last updated: 2014-02-20 15:02:50 +0100
## Comment 19532
Date: 2014-02-04 22:30:02 +0100
From: Ken Leese <<ken.leese>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; Trident/7.0; rv:11.0) like Gecko
Build Identifier:
From ISO 9075 SQL/Framework Part 2:
<delimited identifier> ::= <double quote> <delimited identifier body> <double quote>
<delimited identifier body> ::= <delimited identifier part>...
<delimited identifier part> ::= <nondoublequote character> | <doublequote symbol>
<nondoublequote character> ::= !! See the Syntax Rules (below)
6) A <nondoublequote character> is any character of the source language character set other than a <double
quote>.
<doublequote symbol> ::= "" !! two consecutive double quote characters
However in file …/sql/server/sql_scan.c, the function valid_ident() does not support the standardized syntax, thus causing fatal incompatibilities with existing popular data visualization offerings. In addition, the function erroneously interprets “\” as an escape character.
Reproducible: Always
### Steps to Reproduce:
Two examples simplified from existing ODBC client requests:
-- determine database support for temporary tables
CREATE LOCAL TEMPORARY TABLE "Connect" ("COL" INTEGER) ON COMMIT PRESERVE ROWS
-- Grouping select
SELECT "voyages"."boatname" AS "none:boatname:nk" FROM "sys"."voyages" "voyages" GROUP BY "none:boatname:nk"
### Actual Results:
Erroneous syntax error diagnostics
### Expected Results:
Support for SQL-92 syntax
Correct syntax for a delimited identifier containing a doublequote is "special""identifier" (NOT "special\"identifier" which is incorrect).
Note the comment within …/sql/server/sql_scan.c:
/* Todo check if what valid 'quoted' idents are */
Support for SQL-92 delimited identifier syntax is a requirement for compatibility with existing data visualization client software.
## Comment 19547
Date: 2014-02-05 14:38:37 +0100
From: @njnes
adapted the valid_ident too allow any character except " (and monetdb specific, first char shouldn't be a % and length <= 1024)
## Comment 19548
Date: 2014-02-05 14:56:29 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [dc5a31561c13](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=dc5a31561c13) 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=dc5a31561c13](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=dc5a31561c13)
Changeset description:
fixed bug #3432, allow for any character in quoted identifiers
## Comment 19608
Date: 2014-02-20 15:02:50 +0100
From: @sjoerdmullender
Jan2014 has been released.
| MonetDB SQL syntax incompatible with SQL-92 <delimited identifier> syntax | https://api.github.com/repos/MonetDB/MonetDB/issues/3432/comments | 0 | 2020-11-30T12:37:33Z | 2024-06-27T12:02:21Z | https://github.com/MonetDB/MonetDB/issues/3432 | 753,444,037 | 3,432 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-04 22:01:21 +0100
From: Ken Leese <<ken.leese>>
To: clients devs <<bugs-clients>>
Version: 11.17.9 (Jan2014)
Last updated: 2014-02-20 15:02:32 +0100
## Comment 19531
Date: 2014-02-04 22:01:21 +0100
From: Ken Leese <<ken.leese>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; Trident/7.0; rv:11.0) like Gecko
Build Identifier:
The ODBC SqlGetInfo function with parameter SQL_NUMERIC_FUNCTIONS currently returns SQL_FN_NUM_TRUNCATE. Apparently the current MonetDB TRUNCATE function is not a NUMERIC function, but instead a STRING function.
http://msdn.microsoft.com/en-us/library/ms711743(v=vs.85).aspx describes the ODBC-compliant behaviour of the numeric TRUNCATE function.
If a numeric TRUNC() function is not available (and supported through the ODBC "{fn...}" syntax), then SqlGetInfo should not return SQL_FN_NUM_TRUNCATE.
Reproducible: Always
### Steps to Reproduce:
CREATE TABLE "test" ("built_numeric" NUMERIC(6,2));
INSERT INTO "test" VALUES (1700.2);
INSERT INTO "test" VALUES (1700.4);
INSERT INTO "test" VALUES (1760.9);
SELECT {fn TRUNCATE(built_numeric, 3)} AS "trunc_numeric" FROM "test";
### Actual Results:
Values are "170", "170", "176"
### Expected Results:
Values should be "1700.2", "1700.4", and "1760.9".
Support for numeric TRUNC() would be desirable since it would enhance compatibility with widely available data visualization offerings.
The current behaviour (in which the ODBC driver claims support for a feature not supported by the database) causes third-party ODBC client failure in a non-recoverable manner.
The failing scenario was tested after rebuilding the Windows ODBC driver from recent source that includes the enhancement described in bug #3092 "support scalar function escape"
## Comment 19535
Date: 2014-02-05 09:50:14 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [ae06f23df9b5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ae06f23df9b5) 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=ae06f23df9b5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ae06f23df9b5)
Changeset description:
Our truncate function in SQL is not the numeric kind.
This fixes bug #3431.
## Comment 19536
Date: 2014-02-05 09:53:13 +0100
From: @sjoerdmullender
An even better fix would of course be to implement the numeric truncate.
## Comment 19551
Date: 2014-02-05 23:03:04 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [08a368407b9b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=08a368407b9b) 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=08a368407b9b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=08a368407b9b)
Changeset description:
It turns out we do have a function for {fn truncate(...)}.
The function is just not called truncate, but ms_trunc.
See bug #3431 and changeset [ae06f23df9b5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ae06f23df9b5)
## Comment 19596
Date: 2014-02-20 15:02:32 +0100
From: @sjoerdmullender
Jan2014 has been released.
| SQLGetInfo returns incorrect value for SQL_FN_NUM_TRUNCATE | https://api.github.com/repos/MonetDB/MonetDB/issues/3431/comments | 0 | 2020-11-30T12:37:30Z | 2024-06-27T12:02:20Z | https://github.com/MonetDB/MonetDB/issues/3431 | 753,444,017 | 3,431 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-04 20:34:29 +0100
From: @mlkersten
To: SQL devs <<bugs-sql>>
Version: 11.15.11 (Feb2013-SP3)
CC: @njnes
Last updated: 2014-02-20 15:02:59 +0100
## Comment 19528
Date: 2014-02-04 20:34:29 +0100
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:26.0) Gecko/20100101 Firefox/26.0
Build Identifier:
sql>create schema tempo ;
operation successful (85.215ms)
a typo creates a wrong error message
sql>create temporary table temp.dummy(i int);
CREATE TABLE: no such schema 'temporary'
In Postgresql, temporary tables can only be assigned
to a specific schema;
sql>create temporary table tempo.dummy(i int);
operation successful (28.985ms)
I had expected a proper error message
sql>select * from tempo.dummy;
SELECT: no such table 'dummy'
sql>select * from tmp.dummy;
+---+
| i |
+===+
+---+
Reproducible: Always
### Steps to Reproduce:
Test file has been added
## Comment 19529
Date: 2014-02-04 20:38:13 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [8278337135e7](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=8278337135e7) 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=8278337135e7](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=8278337135e7)
Changeset description:
Add test file temporary.Bug-3430
## Comment 19541
Date: 2014-02-05 11:48:18 +0100
From: @njnes
local temp tables are indeed stored in the tmp schema, ie the schema part should not be given, and if given should give an error when not equal to tmp. For global temporary tables this is not the case.
The first create has one more side effect, ie temp gets expanded to temporary.
## Comment 19542
Date: 2014-02-05 12:09:24 +0100
From: @njnes
check for local temporary tables with incorrect schema's (!= tmp) added.
## Comment 19546
Date: 2014-02-05 13:00:19 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [be0051b026da](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=be0051b026da) 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=be0051b026da](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=be0051b026da)
Changeset description:
approved output for bug #3427 (also use auto commit mode)
fixed bug #3430, ie check if local temporary tables are stored in the 'tmp' schema
fixed bug #3428, don't overwrite the statement number, but use a temporary structure to store the
NotNil select statements. Solves the problem that the sum(a), sub(a*b) uses the a after nil
removal in the a*b.
fixed bug #3425, correctly add the timezone with the proper type (ie second interval (ie with scale 3))
## Comment 19615
Date: 2014-02-20 15:02:59 +0100
From: @sjoerdmullender
Jan2014 has been released.
| Wrong temporary handling | https://api.github.com/repos/MonetDB/MonetDB/issues/3430/comments | 0 | 2020-11-30T12:37:28Z | 2024-06-27T12:02:19Z | https://github.com/MonetDB/MonetDB/issues/3430 | 753,443,988 | 3,430 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-04 17:27:40 +0100
From: @hannesmuehleisen
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: ajdamico
Last updated: 2014-02-20 15:02:40 +0100
## Comment 19525
Date: 2014-02-04 17:27:40 +0100
From: @hannesmuehleisen
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/33.0.1750.58 Safari/537.36
Build Identifier:
The attached SQL script creates two tables, selects a lot of columns from the join result of the two, while sampling the result.
Reproducible: Always
### Steps to Reproduce:
1. create tables and load data
2. run query
3. crash
### Actual Results:
Crash
### Expected Results:
No crash
## Comment 19526
Date: 2014-02-04 17:29:48 +0100
From: @hannesmuehleisen
Created attachment 260
ddl, load, query, data
> Attached file: [createloadquerydata.zip](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3429_createloadquerydata.zip_260) (application/zip, 29435 bytes)
> Description: ddl, load, query, data
## Comment 19527
Date: 2014-02-04 18:03:33 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [785d7ada4002](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=785d7ada4002) 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=785d7ada4002](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=785d7ada4002)
Changeset description:
Fix for bug #3429
## Comment 19530
Date: 2014-02-04 20:59:05 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [b4e866a84995](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=b4e866a84995) 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=b4e866a84995](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=b4e866a84995)
Changeset description:
Test for bug #3429.
The crash in this test is due to an unrelated bug.
## Comment 19537
Date: 2014-02-05 09:55:03 +0100
From: @sjoerdmullender
The crash in the test has nothing to do with this bug.
## Comment 19603
Date: 2014-02-20 15:02:40 +0100
From: @sjoerdmullender
Jan2014 has been released.
| SAMPLE on JOIN result crashes server | https://api.github.com/repos/MonetDB/MonetDB/issues/3429/comments | 0 | 2020-11-30T12:37:26Z | 2024-06-27T12:02:18Z | https://github.com/MonetDB/MonetDB/issues/3429 | 753,443,974 | 3,429 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-04 14:51:54 +0100
From: Alex Shestakov <<shura.shestakov>>
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: @njnes
Last updated: 2014-02-20 15:02:53 +0100
## Comment 19523
Date: 2014-02-04 14:51:54 +0100
From: Alex Shestakov <<shura.shestakov>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/32.0.1700.76 Safari/537.36
Build Identifier:
A valid SQL query cannot be processed.
Reproducible: Always
### Steps to Reproduce:
create table t(
a real,
b real);
insert into t values (null, 1);
insert into t values (2, 1);
select sum(a) , sum(a * b) from t;
### Actual Results:
Error: inputs not the same size.
SQLState: 22000
ErrorCode: 0
### Expected Results:
2
Broken in the latest release candidate (11.17.3) as well as in RC1 (11.17.1).
## Comment 19524
Date: 2014-02-04 16:17:46 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [917086517922](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=917086517922) 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=917086517922](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=917086517922)
Changeset description:
Added test for bug #3428.
## Comment 19540
Date: 2014-02-05 11:38:41 +0100
From: @njnes
the combination used the wrong input (a). Fixed in sql_gencode.c
## Comment 19545
Date: 2014-02-05 13:00:18 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [be0051b026da](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=be0051b026da) 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=be0051b026da](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=be0051b026da)
Changeset description:
approved output for bug #3427 (also use auto commit mode)
fixed bug #3430, ie check if local temporary tables are stored in the 'tmp' schema
fixed bug #3428, don't overwrite the statement number, but use a temporary structure to store the
NotNil select statements. Solves the problem that the sum(a), sub(a*b) uses the a after nil
removal in the a*b.
fixed bug #3425, correctly add the timezone with the proper type (ie second interval (ie with scale 3))
## Comment 19611
Date: 2014-02-20 15:02:53 +0100
From: @sjoerdmullender
Jan2014 has been released.
| Aggregation over two columns is broken | https://api.github.com/repos/MonetDB/MonetDB/issues/3428/comments | 0 | 2020-11-30T12:37:23Z | 2024-06-27T12:02:17Z | https://github.com/MonetDB/MonetDB/issues/3428 | 753,443,955 | 3,428 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-02-01 11:20:30 +0100
From: @mlkersten
To: SQL devs <<bugs-sql>>
Version: 11.15.7 (Feb2013-SP2)
CC: @njnes
Last updated: 2014-02-20 15:03:00 +0100
## Comment 19514
Date: 2014-02-01 11:20:30 +0100
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:26.0) Gecko/20100101 Firefox/26.0
Build Identifier:
Unexpected type error using now() and current_timestamp in table creation.
Enlarged test file has been added.
Reproducible: Always
## Comment 19517
Date: 2014-02-01 15:07:43 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [9fa129eaa0d9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9fa129eaa0d9) 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=9fa129eaa0d9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=9fa129eaa0d9)
Changeset description:
current_timestamp.Bug-3427: ensure deterministic test result
(otherwise, current output will always differ from (to be provided) stable output)
## Comment 19518
Date: 2014-02-01 15:07:45 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [bdff2217856f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=bdff2217856f) 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=bdff2217856f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=bdff2217856f)
Changeset description:
current_timestamp.Bug-3427: completed test:
now() can be used for / assigned to both time & timestamp
current_timestamp can be used for / assigned to both time & timestamp
current_time can only be used for / assigned to time, but not to timestamp:
"ERROR = !types timetz(7,0) and timestamp(7,0) are not equal"
## Comment 19519
Date: 2014-02-01 15:07:47 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [394aed897576](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=394aed897576) 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=394aed897576](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=394aed897576)
Changeset description:
current_timestamp.Bug-3427: added expected stable output
## Comment 19539
Date: 2014-02-05 11:12:14 +0100
From: @njnes
We correctly give an error when a time value is assigned to a timestamp.
## Comment 19544
Date: 2014-02-05 13:00:17 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [be0051b026da](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=be0051b026da) 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=be0051b026da](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=be0051b026da)
Changeset description:
approved output for bug #3427 (also use auto commit mode)
fixed bug #3430, ie check if local temporary tables are stored in the 'tmp' schema
fixed bug #3428, don't overwrite the statement number, but use a temporary structure to store the
NotNil select statements. Solves the problem that the sum(a), sub(a*b) uses the a after nil
removal in the a*b.
fixed bug #3425, correctly add the timezone with the proper type (ie second interval (ie with scale 3))
## Comment 19616
Date: 2014-02-20 15:03:00 +0100
From: @sjoerdmullender
Jan2014 has been released.
| Consistent use of current_timestamp and now() | https://api.github.com/repos/MonetDB/MonetDB/issues/3427/comments | 0 | 2020-11-30T12:37:21Z | 2024-06-27T12:02:16Z | https://github.com/MonetDB/MonetDB/issues/3427 | 753,443,930 | 3,427 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-01-31 16:12:03 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: -- development
Last updated: 2016-04-11 11:46:55 +0200
## Comment 19511
Date: 2014-01-31 16:12:03 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:26.0) Gecko/20100101 Firefox/26.0
Build Identifier:
It would be handy to have support for (some) system event triggers, such as
CREATE TRIGGER my_logon_trigger AFTER LOGON
In Oracle you can use for instance:
CREATE OR REPLACE TRIGGER SYS.my_logon_trigger
AFTER LOGON
ON DATABASE
ENABLE
declare
v_user varchar2(30):=user;
sql_stmt1 varchar2(256) :='alter session set nls_date_format='YYYY-MM-DD HH24:MI:SS';
begin
if (v_user='Martin') THEN
execute immediate sql_stmt1;
end if;
end;
/
Note that this syntax is Oracle specific and NOT standard SQL.
Reproducible: Didn't try
this could resolve issue id=3322
| Extend SQL triggers to fire on system events such as AFTER LOGON | https://api.github.com/repos/MonetDB/MonetDB/issues/3426/comments | 0 | 2020-11-30T12:37:19Z | 2024-06-28T13:40:17Z | https://github.com/MonetDB/MonetDB/issues/3426 | 753,443,912 | 3,426 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-01-29 10:53:46 +0100
From: @mlkersten
To: SQL devs <<bugs-sql>>
Version: 11.15.11 (Feb2013-SP3)
CC: @njnes
Last updated: 2014-02-20 15:03:02 +0100
## Comment 19509
Date: 2014-01-29 10:53:46 +0100
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:26.0) Gecko/20100101 Firefox/26.0
Build Identifier:
The extraction of hour and month from a timestamp seems incorrect.
Reproducible: Always
### Steps to Reproduce:
select now(), extract(second from now());
select now(), extract(minute from now());
select now(), extract(hour from now());
select now(), extract(month from now());
select now(), extract(year from now());
### Actual Results:
sql>select now(), extract(second from now());
+----------------------------------+-------------------+
| L1 | current_timestamp |
+==================================+===================+
| 2014-01-29 10:48:26.000000+01:00 | 26.000 |
+----------------------------------+-------------------+
1 tuple (0.672ms)
sql>select now(), extract(minute from now());
+----------------------------------+-------------------+
| L1 | current_timestamp |
+==================================+===================+
| 2014-01-29 10:48:48.000000+01:00 | 48 |
+----------------------------------+-------------------+
1 tuple (0.897ms)
sql>select now(), extract(hour from now());
+----------------------------------+-------------------+
| L1 | current_timestamp |
+==================================+===================+
| 2014-01-29 10:48:58.000000+01:00 | 1 |
+----------------------------------+-------------------+
1 tuple (0.797ms)
sql>select now(), extract(month from now());
+----------------------------------+-------------------+
| L1 | current_timestamp |
+==================================+===================+
| 2014-01-29 10:49:19.000000+01:00 | 3 |
+----------------------------------+-------------------+
1 tuple (1.017ms)
sql>select now(), extract(year from now());
+----------------------------------+-------------------+
| L1 | current_timestamp |
+==================================+===================+
| 2014-01-29 10:49:29.000000+01:00 | 2014 |
+----------------------------------+-------------------+
1 tuple (0.832ms)
### Expected Results:
sql>select now(), extract(second from now());
+----------------------------------+-------------------+
| L1 | current_timestamp |
+==================================+===================+
| 2014-01-29 10:48:26.000000+01:00 | 26.000 |
+----------------------------------+-------------------+
1 tuple (0.672ms)
sql>select now(), extract(minute from now());
+----------------------------------+-------------------+
| L1 | current_timestamp |
+==================================+===================+
| 2014-01-29 10:48:48.000000+01:00 | 48 |
+----------------------------------+-------------------+
1 tuple (0.897ms)
sql>select now(), extract(hour from now());
+----------------------------------+-------------------+
| L1 | current_timestamp |
+==================================+===================+
| 2014-01-29 10:48:58.000000+01:00 | 10 |
+----------------------------------+-------------------+
1 tuple (0.797ms)
sql>select now(), extract(month from now());
+----------------------------------+-------------------+
| L1 | current_timestamp |
+==================================+===================+
| 2014-01-29 10:49:19.000000+01:00 | 1 |
+----------------------------------+-------------------+
1 tuple (1.017ms)
sql>select now(), extract(year from now());
+----------------------------------+-------------------+
| L1 | current_timestamp |
+==================================+===================+
| 2014-01-29 10:49:29.000000+01:00 | 2014 |
+----------------------------------+-------------------+
1 tuple (0.832ms)
## Comment 19510
Date: 2014-01-29 13:10:23 +0100
From: @mlkersten
sql>explain select now(), extract(day from now());
+----------------------------------------------------------------------------+
| mal |
+============================================================================+
| function user.s3_1{autoCommit=true}():void; |
| X_2 := mtime.current_timestamp(); |
| X_3 := mtime.current_timestamp(); |
| X_5 := mtime.timestamp_add_msec_interval(X_3,3600000000:lng); |
| X_6 := mtime.day(X_5); |
| X_7 := sql.resultSet(2,1,X_2); |
| sql.rsColumn(X_7,".L1","L1","timestamptz",7,0,X_2); |
| sql.rsColumn(X_7,".L1","current_timestamp","int",32,0,X_6); |
| X_19 := io.stdout(); |
| sql.exportResult(X_19,X_7); |
| end s3_1; |
| querylog.define("select now(), extract(day from now());","default_pipe") |
| optimizer.mitosis() |
| optimizer.dataflow() |
+----------------------------------------------------------------------------+
sql>declare ts timestamp;
operation successful (0.657ms)
sql>set ts = now();
operation successful (0.645ms)
sql>explain select now(), extract(day from ts);
+---------------------------------------------------------------------------------+
| mal |
+=================================================================================+
| function user.s7_1{autoCommit=true}():void; |
| X_2 := sql.mvc(); |
| X_3 := mtime.current_timestamp(); |
| X_5:timestamp := sql.getVariable(X_2,"ts"); |
| X_6 := mtime.day(X_5); |
| X_7 := sql.resultSet(2,1,X_3); |
| sql.rsColumn(X_7,".L1","L1","timestamptz",7,0,X_3); |
| sql.rsColumn(X_7,".L1","day_single_value","int",32,0,X_6); |
| X_20 := io.stdout(); |
| sql.exportResult(X_20,X_7); |
| end s7_1; |
| querylog.define("explain select now(), extract(day from ts);","default_pipe") |
| optimizer.mitosis() |
| optimizer.dataflow() |
+---------------------------------------------------------------------------------+
14 tuples (1.308ms)
## Comment 19520
Date: 2014-02-03 15:57:57 +0100
From: @sjoerdmullender
(In reply to comment 1)
> sql>explain select now(), extract(day from now());
[...]
> | X_5 := mtime.timestamp_add_msec_interval(X_3,3600000000:lng);
I think the value 3600000000 is a factor 1000 too high. Where does it come from?
## Comment 19533
Date: 2014-02-05 09:33:28 +0100
From: @njnes
that (incorrectly) comes from the second_interval. I have a patch.
## Comment 19534
Date: 2014-02-05 09:44:39 +0100
From: @njnes
extended the test ./BugTracker-2011/Tests/extract_seconds.Bug-2793.sql
## Comment 19543
Date: 2014-02-05 13:00:13 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [be0051b026da](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=be0051b026da) 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=be0051b026da](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=be0051b026da)
Changeset description:
approved output for bug #3427 (also use auto commit mode)
fixed bug #3430, ie check if local temporary tables are stored in the 'tmp' schema
fixed bug #3428, don't overwrite the statement number, but use a temporary structure to store the
NotNil select statements. Solves the problem that the sum(a), sub(a*b) uses the a after nil
removal in the a*b.
fixed bug #3425, correctly add the timezone with the proper type (ie second interval (ie with scale 3))
## Comment 19619
Date: 2014-02-20 15:03:02 +0100
From: @sjoerdmullender
Jan2014 has been released.
| Temporal extraction glitches | https://api.github.com/repos/MonetDB/MonetDB/issues/3425/comments | 0 | 2020-11-30T12:37:16Z | 2024-06-27T12:02:14Z | https://github.com/MonetDB/MonetDB/issues/3425 | 753,443,892 | 3,425 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-01-22 16:15:26 +0100
From: Anthony Damico <<ajdamico>>
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
Last updated: 2014-05-22 09:52:23 +0200
## Comment 19489
Date: 2014-01-22 16:15:26 +0100
From: Anthony Damico <<ajdamico>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:26.0) Gecko/20100101 Firefox/26.0
Build Identifier:
if a CAST AS INTEGER statement gets used on a string that starts out with numbers, the CAST will (improperly?) succeed?
Reproducible: Always
### Steps to Reproduce:
this is what you would expect:
select cast('a00asdf1' as INTEGER);
conversion of string 'a00asdf1' to type int failed.
this seems like it should not happen
select cast('00asdf1' as INTEGER);
+--------------+
| single_value |
+==============+
| 0 |
+--------------+
### Actual Results:
strings that have both numbers and letters get CAST to the leading numbers
### Expected Results:
if there are *any* letters in the string, that should prevent a CAST( xxx AS INTEGER ) from working?
thanks! monetdb rocks
## Comment 19773
Date: 2014-04-24 18:43:40 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [4708187f2a27](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=4708187f2a27) 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=4708187f2a27](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=4708187f2a27)
Changeset description:
Refuse cast from string when not whole string converted.
This fixes bug #3424.
Also added test.
| numeric values at the front of strings determines whether CAST works successfully | https://api.github.com/repos/MonetDB/MonetDB/issues/3424/comments | 0 | 2020-11-30T12:37:14Z | 2024-06-27T12:02:13Z | https://github.com/MonetDB/MonetDB/issues/3424 | 753,443,866 | 3,424 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-01-20 20:51:03 +0100
From: Klaudiusz <<klaudiusz223>>
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: @njnes
Last updated: 2014-02-20 15:02:26 +0100
## Comment 19474
Date: 2014-01-20 20:51:03 +0100
From: Klaudiusz <<klaudiusz223>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/32.0.1700.76 Safari/537.36
Build Identifier:
Using alias in "group by" clause with count(distinct x) causes that query doesn't work.
Probably introduced by
http://dev.monetdb.org/hg/MonetDB/rev/78eca92f6d68
Reproducible: Always
### Steps to Reproduce:
DROP table a;
CREATE table a (k int,b int);
INSERT into a values (1,2);
INSERT into a values (2,2);
INSERT into a values (3,3);
INSERT into a values (4,65);
INSERT into a values (5,21);
SELECT k as c,count(distinct b) from a group by c;
### Actual Results:
nothing happens
### Expected Results:
+------+------+
| c | L1 |
+======+======+
| 1 | 1 |
| 2 | 1 |
| 3 | 1 |
| 4 | 1 |
| 5 | 1 |
+------+------+
5 tuples (1.538ms)
MonetDB 5 server v11.17.1 "Jan2014" (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 252.2GiB available memory, 32 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: *@* (x86_64-redhat-linux-gnu)
Compilation: gcc -O3 -fomit-frame-pointer -pipe -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -m64 -mtune=generic -Wp,-D_FORTIFY_SOURCE=2
Linking : /usr/bin/ld -m elf_x86_64
## Comment 19482
Date: 2014-01-22 12:13:17 +0100
From: @njnes
fixed a bug in the relational optimizer. Rewriting of some special case of distinct with count incorrectly used the expressions of the lower relation. Now it correctly uses the alias names.
## Comment 19486
Date: 2014-01-22 13:08:59 +0100
From: @sjoerdmullender
Since this bug was reported against the RC, we need the NEXTRELEASE resolution (assuming the fix will appear in the correct branch).
## Comment 19488
Date: 2014-01-22 13:17:00 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [9d8906d9166d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9d8906d9166d) 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=9d8906d9166d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=9d8906d9166d)
Changeset description:
fixed bug #3423, improved the group by + distinct rewrite
also added a test for this bug
## Comment 19594
Date: 2014-02-20 15:02:26 +0100
From: @sjoerdmullender
Jan2014 has been released.
| Group by alias with distinct count doesn't work | https://api.github.com/repos/MonetDB/MonetDB/issues/3423/comments | 0 | 2020-11-30T12:37:11Z | 2024-06-27T12:02:12Z | https://github.com/MonetDB/MonetDB/issues/3423 | 753,443,845 | 3,423 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-01-20 12:44:10 +0100
From: Alex Shestakov <<shura.shestakov>>
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: @njnes, richard.monetdb, shura.shestakov, @yzchang
Last updated: 2014-10-31 14:13:42 +0100
## Comment 19462
Date: 2014-01-20 12:44:10 +0100
From: Alex Shestakov <<shura.shestakov>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/31.0.1650.63 Safari/537.36
Build Identifier:
mserver5 receives sigsegv after a big insert into a table.
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7f511e44c700 (LWP 14773)]
0x00007f620cfef1eb in gtr_update_table (tr=0x2d671e0, t=0x30f6920, tchanges=0x7f511e44bd5c) at bat_storage.c:1454
1454 for (n = t->columns.set->h; ok == LOG_OK && n; n = n->next) {
(gdb) where
0 0x00007f620cfef1eb in gtr_update_table (tr=0x2d671e0, t=0x30f6920, tchanges=0x7f511e44bd5c) at bat_storage.c:1454
1 0x00007f620cfeee0e in _gtr_update (tr=0x2d671e0, gtr_update_table_f=0x7f620cfef1d0 <gtr_update_table>) at bat_storage.c:1504
2 0x00007f620cfe91cc in store_manager () at store.c:1552
3 0x00007f620cf9b4c5 in mvc_logmanager () at sql_mvc.c:149
4 0x00007f6214a1178b in thread_starter (arg=0x2edb510) at gdk_system.c:505
5 0x0000003ee9a07851 in start_thread () from /lib64/libpthread.so.0
6 0x0000003ee92e890d in clone () from /lib64/libc.so.6
(gdb) print t->columns
$1 = {sa = 0x2da3b30, destroy = 0x7f620cfe01d0 <column_destroy>, set = 0x0, dset = 0x0, nelm = 0x0}
(gdb) print *t
$3 = {base = {wtime = 1, rtime = 0, allocated = 1, flag = 0, id = 8421, name = 0x30f6a50 "wave"}, type = 0, system = 0 '\000', persistence = SQL_PERSIST, commit_action = CA_COMMIT,
readonly = 0 '\000', query = 0x0, sz = 1024, pkey = 0x0, columns = {sa = 0x2da3b30, destroy = 0x7f620cfe01d0 <column_destroy>, set = 0x0, dset = 0x0, nelm = 0x0}, idxs = {
sa = 0x2da3b30, destroy = 0x7f620cfe14f0 <idx_destroy>, set = 0x0, dset = 0x0, nelm = 0x0}, keys = {sa = 0x2da3b30, destroy = 0x7f620cfe1e50 <key_destroy>, set = 0x0, dset = 0x0,
nelm = 0x0}, triggers = {sa = 0x2da3b30, destroy = 0x7f620cfe1250 <trigger_destroy>, set = 0x0, dset = 0x0, nelm = 0x0}, tables = {sa = 0x2da3b30,
destroy = 0x7f620cfeba70 <table_destroy>, set = 0x0, dset = 0x0, nelm = 0x0}, drop_action = 0, cleared = 0, data = 0x30e49a0, s = 0x2e80450, p = 0x0}
The problem seems to be caused not by the table being updated, but by a different table 'wave'.
Table wave was created as follows.
create table wave (
wave_id bigint auto_increment,
businessdate date not null,
name varchar(256),
unique (businessdate, name)
);
I've dropped table wave and re-created it again. The problem has gone.
Looks like the table had been in some broken state.
Reproducible: Sometimes
## Comment 19476
Date: 2014-01-21 11:22:21 +0100
From: @sjoerdmullender
What did you do to this table "wave" before you got the crash?
Also, if you can reproduce this, please try to do so while the server is running with the extra argument -d10. I would expect this to cause an assertion failure (i.e. abort) or warning message (depending on how the server was built). I'd like to see the messages produced.
## Comment 19477
Date: 2014-01-21 12:17:43 +0100
From: Alex Shestakov <<shura.shestakov>>
I've just created table wave as shown above and inserted data into it using
"insert into wave select from ..." statement.
After I've dropped and re-created the table the problem has gone. I can't reproduce it.
## Comment 19499
Date: 2014-01-24 10:28:29 +0100
From: Alex Shestakov <<shura.shestakov>>
It started happening again for no obvious reason. Now not only on inserts. I was doing a delete from a different table and it crashed.
[Switching to Thread 0x7f6ad8129700 (LWP 27642)]
0x00007f7bd414c1eb in gtr_update_table (tr=0x2d0b3b0, t=0x309b490, tchanges=0x7f6ad8128d5c) at bat_storage.c:1454
1454 for (n = t->columns.set->h; ok == LOG_OK && n; n = n->next) {
(gdb) print *t
$1 = {base = {wtime = 1, rtime = 0, allocated = 1, flag = 0, id = 8421, name = 0x309b5c0 "wave"}, type = 0, system = 0 '\000', persistence = SQL_PERSIST, commit_action = CA_COMMIT,
readonly = 0 '\000', query = 0x0, sz = 1024, pkey = 0x0, columns = {sa = 0x300acd0, destroy = 0x7f7bd413d1d0 <column_destroy>, set = 0x0, dset = 0x0, nelm = 0x0}, idxs = {
sa = 0x300acd0, destroy = 0x7f7bd413e4f0 <idx_destroy>, set = 0x0, dset = 0x0, nelm = 0x0}, keys = {sa = 0x300acd0, destroy = 0x7f7bd413ee50 <key_destroy>, set = 0x0, dset = 0x0,
nelm = 0x0}, triggers = {sa = 0x300acd0, destroy = 0x7f7bd413e250 <trigger_destroy>, set = 0x0, dset = 0x0, nelm = 0x0}, tables = {sa = 0x300acd0,
destroy = 0x7f7bd4148a70 <table_destroy>, set = 0x0, dset = 0x0, nelm = 0x0}, drop_action = 0, cleared = 0, data = 0x30c2b40, s = 0x2d970b0, p = 0x0}
(gdb) print t->columns
$2 = {sa = 0x300acd0, destroy = 0x7f7bd413d1d0 <column_destroy>, set = 0x0, dset = 0x0, nelm = 0x0}
(gdb) where
0 0x00007f7bd414c1eb in gtr_update_table (tr=0x2d0b3b0, t=0x309b490, tchanges=0x7f6ad8128d5c) at bat_storage.c:1454
1 0x00007f7bd414be0e in _gtr_update (tr=0x2d0b3b0, gtr_update_table_f=0x7f7bd414c1d0 <gtr_update_table>) at bat_storage.c:1504
2 0x00007f7bd41461cc in store_manager () at store.c:1552
3 0x00007f7bd40f84c5 in mvc_logmanager () at sql_mvc.c:149
4 0x00007f7bdbb6e78b in thread_starter (arg=0x2e18590) at gdk_system.c:505
5 0x0000003ee9a07851 in start_thread () from /lib64/libpthread.so.0
6 0x0000003ee92e890d in clone () from /lib64/libc.so.6
I was running mserver5 with -d10 option. It didn't complain on any settings.
The only thing it printed after the initial startup message was the following when I connected to it with SQuirrel SQL (jdbc).
Bulk operator required for str.stringleft
Bulk operator required for str.stringleft
## Comment 19929
Date: 2014-07-25 12:55:39 +0200
From: Richard Hughes <<richard.monetdb>>
I've had a SIGSEGV that looks sufficiently similar to this that I'm going to put it here rather than creating a new bug:
v11.17.17 "Jan2014-SP2"
(gdb) bt
0 gtr_update_delta (tr=tr@entry=0x3333e10, cbat=0x7fb538058840,
changes=changes@entry=0x7fd52ff7fe5c) at bat_storage.c:1429
1 0x00007fd5308a3025 in gtr_update_table (tchanges=<synthetic pointer>,
t=0x31bdbf0, tr=0x3333e10) at bat_storage.c:1468
2 _gtr_update (gtr_update_table_f=<optimized out>, tr=0x3333e10)
at bat_storage.c:1512
3 gtr_update (tr=0x3333e10) at bat_storage.c:1529
4 0x00007fd530893672 in store_manager () at store.c:1591
5 0x00007fd530839605 in mvc_logmanager () at sql_mvc.c:149
6 0x00007fd53714497b in thread_starter (arg=0x3110fd0) at gdk_system.c:505
7 0x00007fd534c770a4 in start_thread (arg=0x7fd52ff80700)
at pthread_create.c:309
8 0x00007fd5349ac04d in clone ()
at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111
(gdb) p cur
$1 = (BAT *) 0x0
(gdb) p *cbat
$2 = {name = 0x7fb538090420 "sys_staging_data_id", bid = 0, ibase = 0,
ibid = 28806, ubid = 2252, cnt = 10273, ucnt = 0, cached = 0x0,
wtime = 6398, next = 0x7fb6645e6070}
This happened once only, with a load pattern that hasn't changed before or since. Our system runs with exactly one process (having one connection) doing all writes (mostly bulk inserts, with some updates sprinkled in), and a few processes (2 or 3) doing SELECTs. My speculation (based on pretty much zero evidence) is that one of the SELECTs required an index rebuild and raced with a concurrent COPY or UPDATE. Maybe. Good luck!
I'll keep the core file for a few weeks, if anybody wants me to ask it any more questions.
## Comment 19937
Date: 2014-07-30 15:07:14 +0200
From: Richard Hughes <<richard.monetdb>>
Another SIGSEGV that looks similar, v11.17.17 "Jan2014-SP2" (same binary as my last comment)
(gdb) bt
0 BATcleanProps (b=0x0) at bat_storage.c:1407
1 tr_update_delta (tr=tr@entry=0x7f0558007f30, obat=0x7f0559844bc0,
cbat=0x7f0559a229d0, unique=0) at bat_storage.c:1635
2 0x00007f0566ea9b05 in update_table (tr=0x7f0558007f30, ft=0x7f055803fa60,
tt=0x1df86b0) at bat_storage.c:1778
3 0x00007f0566e98558 in rollforward_update_table (tr=0x7f0558007f30,
ft=0x7f055803fa60, tt=0x1df86b0, mode=3) at store.c:2744
4 0x00007f0566ea082a in rollforward_changeset_updates (mode=3,
fd=<optimized out>,
rollforward_deletes=0x7f0566e94d30 <rollforward_drop_table>,
rollforward_creates=0x7f0566e94850 <rollforward_create_table>,
rollforward_updates=0x7f0566e97e50 <rollforward_update_table>,
b=0x1b6cbe0, ts=0x1b6cc08, fs=0x7f0558008258, tr=0x7f0558007f30)
at store.c:2357
5 rollforward_update_schema (tr=0x7f0558007f30, fs=0x7f0558008230,
ts=0x1b6cbe0, mode=3) at store.c:2820
6 0x00007f0566ea1145 in rollforward_changeset_updates (
mode=<optimized out>, fd=<optimized out>,
rollforward_deletes=<optimized out>, rollforward_creates=<optimized out>,
rollforward_updates=<optimized out>, b=<optimized out>,
ts=<optimized out>, fs=<optimized out>, tr=<optimized out>)
at store.c:2357
7 rollforward_trans (mode=<optimized out>, tr=<optimized out>)
at store.c:2842
8 sql_trans_commit (tr=0x7f0558007f30) at store.c:3258
9 0x00007f0566e3fb28 in mvc_commit (m=0x7f05580053f0, chain=chain@entry=0,
name=name@entry=0x0) at sql_mvc.c:274
10 0x00007f0566db5e6d in SQLtransaction (cntxt=0x815968, mb=<optimized out>,
stk=<optimized out>, pci=0x7f0559ca0050) at sql.c:290
11 0x00007f056f673158 in runMALsequence (cntxt=0x0, mb=0x7f0560107590,
startpc=1, stoppc=15, stk=0x7f0559c59e50, env=0x68ec, pcicaller=0x0)
at mal_interpreter.c:648
12 0x00007f056f6749bf in runMAL (cntxt=0x0, cntxt@entry=0x815968,
mb=0x7f0559a6e1f0, mbcaller=0x1, mbcaller@entry=0x0, env=0xf,
env@entry=0x0) at mal_interpreter.c:374
13 0x00007f0566dcb95b in SQLengineIntern (c=0x815968, be=0x7f05580c96b0)
at sql_scenario.c:2301
14 0x00007f056f68f4b7 in runPhase (phase=4, c=0x815968) at mal_scenario.c:528
15 runScenarioBody (c=c@entry=0x815968) at mal_scenario.c:572
16 0x00007f056f69001d in runScenario (c=c@entry=0x815968)
at mal_scenario.c:592
17 0x00007f056f6900e8 in MSserveClient (dummy=0x815968) at mal_session.c:456
18 0x00007f056f1dc97b in thread_starter (arg=0x7f0560000a40)
at gdk_system.c:505
19 0x00007f056cd0f0a4 in start_thread (arg=0x7f0566385700)
at pthread_create.c:309
20 0x00007f056ca4404d in clone ()
at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111
(gdb) f 1
1 tr_update_delta (tr=tr@entry=0x7f0558007f30, obat=0x7f0559844bc0,
cbat=0x7f0559a229d0, unique=0) at bat_storage.c:1635
1635 BATcleanProps(cur);
(gdb) p *obat
$1 = {name = 0x7f0559a28260 "sys_staging_data_id", bid = 0, ibase = 0,
ibid = 52498, ubid = 2252, cnt = 2044, ucnt = 0, cached = 0x0,
wtime = 26859, next = 0x7f0558d04760}
The statement being run at the time was
copy into staging_data from stdin delimiters '\t' null as '\\N';
Can somebody please advise me what debugging code I can add in order to help track this down?
## Comment 19940
Date: 2014-08-01 11:26:14 +0200
From: Richard Hughes <<richard.monetdb>>
Freshly upgraded to Jan2014SP3 (11.17.21), same backtrace as comment 4.
(gdb) bt
0 gtr_update_delta (tr=tr@entry=0x1e9a760, cbat=0x7f001c1011e0,
changes=changes@entry=0x7f0032f4ce5c) at bat_storage.c:1429
1 0x00007f00356c07b5 in gtr_update_table (tchanges=<synthetic pointer>,
t=0x2516790, tr=0x1e9a760) at bat_storage.c:1468
2 _gtr_update (gtr_update_table_f=<optimized out>, tr=0x1e9a760)
at bat_storage.c:1512
3 gtr_update (tr=0x1e9a760) at bat_storage.c:1529
4 0x00007f00356b0e02 in store_manager () at store.c:1591
5 0x00007f0035656c35 in mvc_logmanager () at sql_mvc.c:149
6 0x00007f003d9f4bfb in thread_starter (arg=0x2230940) at gdk_system.c:505
7 0x00007f003b5270a4 in start_thread (arg=0x7f0032f4d700)
at pthread_create.c:309
8 0x00007f003b25c04d in clone ()
at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111
(gdb) p *cur
Cannot access memory at address 0x0
(gdb) p *cbat
$1 = {name = 0x7f001c101100 "sys_timerangecache_t", bid = 0, ibase = 0,
ibid = 19882, ubid = 1166, cnt = 9730, ucnt = 0, cached = 0x0,
wtime = 9788, next = 0x251cea0}
The statements that had recently executed related to sys.timerangecache were:
delete from timerangecache;
insert into timerangecache (t,n) select t,count(*) from (select (start-(select timestamp '1970-1
-1'))/1800000 from fulldata) as t(t,n) group by t;
commit;
## Comment 20163
Date: 2014-09-17 16:45:21 +0200
From: @yzchang
Possibly fixed by Sjoerd's changeset: [fe0b3b84c297](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=fe0b3b84c297)
Based on a patch of Richard Huges, thanks!
Please test
## Comment 20334
Date: 2014-10-31 14:13:42 +0100
From: @sjoerdmullender
Oct2014 has been released.
| Segmentation fault after large table insert | https://api.github.com/repos/MonetDB/MonetDB/issues/3422/comments | 0 | 2020-11-30T12:37:09Z | 2024-06-27T12:02:11Z | https://github.com/MonetDB/MonetDB/issues/3422 | 753,443,824 | 3,422 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-01-17 19:21:50 +0100
From: Alex Shestakov <<shura.shestakov>>
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.17.9 (Jan2014)
CC: @mlkersten, @njnes, shura.shestakov, @drstmane
Last updated: 2014-09-16 22:02:50 +0200
## Comment 19457
Date: 2014-01-17 19:21:50 +0100
From: Alex Shestakov <<shura.shestakov>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/31.0.1650.63 Safari/537.36
Build Identifier:
An SQL query with aggregation returns multiple rows per group:
sql>create table t2 as select * from orderconnectormetrics where toplevelorderid = 5508 with data;
operation successful (128.450ms)
sql>select * from t2;
+-----------------+--------------+-----------+-----------------+-----------------+---------------+-----------------+
| toplevelorderid | businessdate | connector | filledvalue | filledqty | numberoffills | chargeusd |
+=================+==============+===========+=================+=================+===============+=================+
| 5508 | 2013-02-01 | C1 | 3650790 | 3500000 | 4 | null |
| 5508 | 2013-02-01 | C2 | 1564640 | 1500000 | 2 | null |
+-----------------+--------------+-----------+-----------------+-----------------+---------------+-----------------+
2 tuples (0.639ms)
sql>select toplevelorderid, count(*), sum(numberoffills) from t2 group by toplevelorderid;
+-----------------+------+------+
| toplevelorderid | L1 | L2 |
+=================+======+======+
| 5508 | 1 | 4 |
| 5508 | 1 | 2 |
+-----------------+------+------+
2 tuples (0.623ms)
sql>select toplevelorderid, count(*), sum(numberoffills) from t2 where toplevelorderid = 5508 group by toplevelorderid;
+-----------------+------+------+
| toplevelorderid | L1 | L2 |
+=================+======+======+
| 5508 | 1 | 4 |
| 5508 | 1 | 2 |
+-----------------+------+------+
2 tuples (0.284ms)
sql>
After the server restart the problem with t2 table went away:
sql>select toplevelorderid, count(*) from t2 group by toplevelorderid;
+-----------------+------+
| toplevelorderid | L1 |
+=================+======+
| 5508 | 2 |
+-----------------+------+
1 tuple (1.889ms)
But the problem with the original bigger table remained:
sql>select toplevelorderid, count(*), sum(numberoffills) from orderconnectormetrics where toplevelorderid = 5508 group by toplevelorderid;
+-----------------+------+------+
| toplevelorderid | L1 | L2 |
+=================+======+======+
| 5508 | 1 | 4 |
| 5508 | 1 | 2 |
+-----------------+------+------+
If i re-populate t2 from the big table like follows, the problem with t2 comes back.
sql>delete from t2;
2 affected rows (142.768ms)
sql>insert into t2 select * from orderconnectormetrics where toplevelorderid = 5508;
2 affected rows (44.737ms)
If I populate t2 manually, there is no problem:
sql>insert into t2 values (5508, '2013-02-01', 'C1', 3000000, 3000000, 4, null);
1 affected row (150.950ms)
sql>insert into t2 values (5508, '2013-02-01', 'C2', 2000000, 2000000, 2, null);
1 affected row (81.379ms)
sql>select toplevelorderid, count(*), sum(numberoffills) from t2 where toplevelorderid = 5508 group by toplevelorderid;
+-----------------+------+------+
| toplevelorderid | L1 | L2 |
+=================+======+======+
| 5508 | 2 | 6 |
+-----------------+------+------+
1 tuple (0.331ms)
Here is the query plan:
>explain select toplevelorderid, count(*), sum(numberoffills) from t2 where toplevelorderid = 5508 group by toplevelorderid;
+---------------------------------------------------------------------------------------------------------------------------------------------------------+
| mal |
+=========================================================================================================================================================+
| function user.s2_1{autoCommit=true}(A0:lng):void; |
| X_3 := sql.mvc(); |
| X_7 := sql.bind(X_3,"sys","t2","toplevelorderid",0); |
| X_4:bat[:oid,:oid] := sql.tid(X_3,"sys","t2"); |
| X_63 := algebra.subselect(X_7,X_4,A0,A0,true,true,false); |
| (X_10,r1_10) := sql.bind(X_3,"sys","t2","toplevelorderid",2); |
| X_64 := algebra.subselect(r1_10,A0,A0,true,true,false); |
| X_13 := sql.bind(X_3,"sys","t2","toplevelorderid",1); |
| X_65 := algebra.subselect(X_13,X_4,A0,A0,true,true,false); |
| X_15 := sql.subdelta(X_63,X_4,X_10,X_64,X_65); |
| X_17 := sql.projectdelta(X_15,X_7,X_10,r1_10,X_13); |
| (X_18,r1_21,r2_21) := group.subgroupdone(X_17); |
| X_21 := algebra.leftfetchjoin(r1_21,X_17); |
| X_22:bat[:oid,:wrd] := aggr.subcount(X_18,X_18,r1_21,false); |
| X_23 := sql.bind(X_3,"sys","t2","numberoffills",0); |
| (X_25,r1_28) := sql.bind(X_3,"sys","t2","numberoffills",2); |
| X_27 := sql.bind(X_3,"sys","t2","numberoffills",1); |
| X_28 := sql.projectdelta(X_15,X_23,X_25,r1_28,X_27); |
| X_29:bat[:oid,:lng] := aggr.subsum(X_28,X_18,r1_21,true,true); |
| X_31 := sql.resultSet(3,1,X_21); |
| sql.rsColumn(X_31,"sys.t2","toplevelorderid","bigint",64,0,X_21); |
| sql.rsColumn(X_31,"sys.L1","L1","wrd",64,0,X_22); |
| sql.rsColumn(X_31,"sys.L2","L2","bigint",32,0,X_29); |
| X_46 := io.stdout(); |
| sql.exportResult(X_46,X_31); |
| end s2_1; |
| querylog.define("select toplevelorderid, count(*), sum(numberoffills) from t2 where toplevelorderid = 5508 group by toplevelorderid;","default_pipe") |
| optimizer.mitosis() |
| optimizer.dataflow() |
+---------------------------------------------------------------------------------------------------------------------------------------------------------+
29 tuples (0.591ms)
Reproducible: Sometimes
## Comment 19458
Date: 2014-01-17 19:25:30 +0100
From: Alex Shestakov <<shura.shestakov>>
The original table was created as follows:
create table OrderConnectorMetrics (
toplevelorderid bigint not null,
businessdate date,
connector varchar(256),
filledValue real,
filledQty real,
numberOfFills int,
chargeUsd real,
unique (toplevelorderid, connector)
);
sql>select * from storage() where "table" = 't2';
+--------+-------+-----------------+---------+-------------+-------+-----------+------------+----------+---------+--------+
| schema | table | column | type | location | count | typewidth | columnsize | heapsize | indices | sorted |
+========+=======+=================+=========+=============+=======+===========+============+==========+=========+========+
| sys | t2 | toplevelorderid | bigint | 01/23/12354 | 2 | 8 | 16 | 0 | 0 | true |
| sys | t2 | businessdate | date | 01/24/12456 | 2 | 4 | 8 | 0 | 0 | true |
| sys | t2 | connector | varchar | 01/25/12556 | 2 | 3 | 2 | 10240 | 0 | true |
| sys | t2 | filledvalue | real | 01/26/12656 | 2 | 4 | 8 | 0 | 0 | false |
| sys | t2 | filledqty | real | 01/27/12756 | 2 | 4 | 8 | 0 | 0 | false |
| sys | t2 | numberoffills | int | 01/30/13056 | 2 | 4 | 8 | 0 | 0 | false |
| sys | t2 | chargeusd | real | 01/31/13156 | 2 | 4 | 8 | 0 | 0 | true |
+--------+-------+-----------------+---------+-------------+-------+-----------+------------+----------+---------+--------+
## Comment 19459
Date: 2014-01-17 19:35:12 +0100
From: Alex Shestakov <<shura.shestakov>>
The problem exists both in Feb2013-SP6 and Jan2014 Release Candidate
## Comment 19460
Date: 2014-01-17 20:05:46 +0100
From: Alex Shestakov <<shura.shestakov>>
The big table seems to have an index on toplevelorderid:
sql>select * from storage() where "table"='orderconnectormetrics';
+--------+-----------------------+--------------------------------------------------------+---------+--------------+---------+-----------+------------+----------+----------+--------+
| schema | table | column | type | location | count | typewidth | columnsize | heapsize | indices | sorted |
+========+=======================+========================================================+=========+==============+=========+===========+============+==========+==========+========+
| sys | orderconnectormetrics | toplevelorderid | bigint | 02/53/25362 | 7757706 | 8 | 62061648 | 0 | 64618496 | false |
| sys | orderconnectormetrics | businessdate | date | 01/11/11146 | 7757706 | 4 | 31030824 | 0 | 0 | true |
| sys | orderconnectormetrics | connector | varchar | 10/50/105017 | 7757706 | 3 | 15515412 | 10240 | 0 | false |
| sys | orderconnectormetrics | filledvalue | real | 01/56/15660 | 7757706 | 4 | 31030824 | 0 | 0 | false |
| sys | orderconnectormetrics | filledqty | real | 07/56/75654 | 7757706 | 4 | 31030824 | 0 | 0 | false |
| sys | orderconnectormetrics | numberoffills | int | 02/61/26174 | 7757706 | 4 | 31030824 | 0 | 0 | false |
| sys | orderconnectormetrics | chargeusd | real | 02/61/26163 | 7757706 | 4 | 31030824 | 0 | 0 | false |
| sys | orderconnectormetrics | orderconnectormetrics_toplevelorderid_connector_unique | oid | 03/30/33025 | 7757706 | 8 | 62061648 | 0 | 0 | false |
+--------+-----------------------+--------------------------------------------------------+---------+--------------+---------+-----------+------------+----------+----------+--------+
8 tuples (4.588ms)
## Comment 19461
Date: 2014-01-18 10:38:39 +0100
From: @mlkersten
Hi,
Thanks for reporting this issue. It is certainly not a simple one, because trying to reproduce the errors on the Jan2014 release so far failed. It behaves correctly.
## Comment 19463
Date: 2014-01-20 14:32:27 +0100
From: Alex Shestakov <<shura.shestakov>>
I've dropped and re-created the table. The problem has gone. It may be related to previous server segfault crashes.
## Comment 19464
Date: 2014-01-20 15:06:21 +0100
From: @sjoerdmullender
Given comment 5, I think we can close this.
If you don't agree, please reopen.
## Comment 19466
Date: 2014-01-20 15:17:09 +0100
From: Alex Shestakov <<shura.shestakov>>
In my view it is still broken.
Somehow the aggregation doesn't work when a table is in some weird state.
I don't know what causes that state but it can be reached by just using the database.
The more interesting fact is that the "weirdness" is inherited if I merely copy the data into another table as illustrated in this bug report.
I'd assume that the aggregation logic (as far as I understand the part that splits the dataset into subgroups) makes some assumptions about the underlying data structures that may not be true.
I don't know if this logic can somehow be made more robust by checking this assumptions, adding more defensive code, etc.
As of now it looks like the database may be in inconsistent state after a segfault which doesn't seem to be such a rare occasion in my experience.
## Comment 19468
Date: 2014-01-20 15:22:47 +0100
From: @sjoerdmullender
Reopened per comment 7.
## Comment 19469
Date: 2014-01-20 15:24:30 +0100
From: @mlkersten
Hi Alex,
Just for clarification, you have not added any UDFs for yourself do you?
## Comment 19470
Date: 2014-01-20 15:28:50 +0100
From: Alex Shestakov <<shura.shestakov>>
Not native code ones.
I do have some SQL functions defined.
## Comment 19471
Date: 2014-01-20 15:49:27 +0100
From: @sjoerdmullender
From comment 7, it looks like some properties may be set incorrectly. Can you run the server with (additional) argument -d10 ? If properties are set incorrectly, this should tell you that.
## Comment 19472
Date: 2014-01-20 16:09:36 +0100
From: Alex Shestakov <<shura.shestakov>>
I've just used the command line that results from using "monetdb start" and added -d10 to it.
The server doesn't seem to complain on any properties:
builtin opt gdk_dbpath = <INSTALL PATH>/var/monetdb5/dbfarm/demo
builtin opt gdk_debug = 0
builtin opt gdk_vmtrim = yes
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_dbpath = <DATABASE PATH>
cmdline opt merovingian_uri = mapi:monetdb://<HOST>:50000/<DBNAME>
cmdline opt mapi_open = false
cmdline opt mapi_port = 0
cmdline opt mapi_usock = <PATH>/.mapi.sock
cmdline opt monet_vault_key = <PATH>/.vaultkey
cmdline opt gdk_nr_threads = 16
cmdline opt max_clients = 200
cmdline opt sql_optimizer = default_pipe
cmdline opt monet_daemon = yes
MonetDB 5 server v11.17.1 "Jan2014"
Serving database '<NAME>', using 16 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs dynamically linked
Found 35.293 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 UNIX domain connection requests on mapi:monetdb://<PATH>/.mapi.sock
MonetDB/JAQL module loaded
MonetDB/SQL module loaded
## Comment 19473
Date: 2014-01-20 17:16:54 +0100
From: @drstmane
You also need to run your queries.
## Comment 19481
Date: 2014-01-21 15:20:55 +0100
From: Alex Shestakov <<shura.shestakov>>
No complains on the query either.
But again the problem is gone after I've re-created the table.
## Comment 19986
Date: 2014-08-06 19:11:51 +0200
From: @mlkersten
Alex
Does the bug appear in the latest bug-fix release?
Or did it disappear?
## Comment 20162
Date: 2014-09-16 22:02:50 +0200
From: @njnes
without the set of statemens which caused the initial 'property' problem we cannot debug this problem. Unless the problem returns we close the bug with a needsinfo.
| aggregation is broken | https://api.github.com/repos/MonetDB/MonetDB/issues/3421/comments | 0 | 2020-11-30T12:37:05Z | 2024-06-27T12:02:10Z | https://github.com/MonetDB/MonetDB/issues/3421 | 753,443,775 | 3,421 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-01-16 11:37:10 +0100
From: Christian Braun <<hcb>>
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
CC: @mlkersten, @njnes
Last updated: 2014-02-20 15:02:48 +0100
## Comment 19456
Date: 2014-01-16 11:37:10 +0100
From: Christian Braun <<hcb>>
User-Agent: Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:26.0) Gecko/20100101 Firefox/26.0
Build Identifier:
After upgrading from 'MonetDB 5 server v11.15.19 "Feb2013-SP6"' to the release candidate, the database does not start. I can create and start a new database. But my existing database does not start. In the log i have:
2014-01-15 17:36:45 MSG merovingian[19494]: starting database 'db', up min/avg/max: 0s/3d/1w, crash average: 0.00 0.00 0.20 (77-67=10)
2014-01-15 17:36:45 MSG db[20059]: arguments: /usr/bin/mserver5 --dbpath=/var/monetdb/dbfarm/db --set merovingian_uri=mapi:monetdb://etna:50000/db --set mapi_open=false --set mapi_port=0 --set mapi_usock=/var/monetdb/dbfarm/db/.mapi.sock --set monet_vault_key=/var/monetdb/dbfarm/db/.vaultkey --set gdk_nr_threads=8 --set max_clients=64 --set sql_optimizer=default_pipe --set monet_daemon=yes
The mserver just hangs and uses 100% cpu on one core with no disk activity. It did not complete the start after over 10 hours. I then terminated the MonetDB processes and reinstalled Feb2013-SP6. With the previous version the database starts fine.
Is there anything else i can try to start my database?
Reproducible: Always
## Comment 19475
Date: 2014-01-21 09:31:42 +0100
From: @sjoerdmullender
We have little to go on here, and I haven't been able to reproduce this problem. Can you provide a stack trace (e.g. run the program ptrace on the server, or attach gdb and execute "thread apply all bt")?
Also, can you attach strace (a few hundred system calls is probably enough) for a bit so that we can get an idea what the server is doing?
Please attach the outputs to this bug report.
## Comment 19478
Date: 2014-01-21 14:52:12 +0100
From: Christian Braun <<hcb>>
Created attachment 254
(gdb) thread apply all bt
> Attached file: [gdb.log](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3420_gdb.log_254) (text/x-log, 8479 bytes)
> Description: (gdb) thread apply all bt
## Comment 19479
Date: 2014-01-21 14:53:18 +0100
From: Christian Braun <<hcb>>
Created attachment 255
strace -tt -p 20487 -o strace.log
> Attached file: [strace.log.gz](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3420_strace.log.gz_255) (application/x-gzip, 126017 bytes)
> Description: strace -tt -p 20487 -o strace.log
## Comment 19480
Date: 2014-01-21 14:56:53 +0100
From: Christian Braun <<hcb>>
Thank you Sjoerd. Please let me know if i can help with anything else.
## Comment 19483
Date: 2014-01-22 12:44:32 +0100
From: Christian Braun <<hcb>>
Created attachment 256
20 hour startup
> Attached file: [merovingian.log](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3420_merovingian.log_256) (text/x-log, 24953 bytes)
> Description: 20 hour startup
## Comment 19484
Date: 2014-01-22 12:50:57 +0100
From: Christian Braun <<hcb>>
After 20 hours the startup of MonetDB completed.
My database has 5k tables with 100k columns and a size of 300GB. Though 90% of the tables/columns are empty. Maybe it is slow because of the number of columns.
## Comment 19485
Date: 2014-01-22 13:03:00 +0100
From: @njnes
5K tables with each (on average) 20 columns?
## Comment 19487
Date: 2014-01-22 13:14:44 +0100
From: Christian Braun <<hcb>>
yes, on average 20 columns.
## Comment 19490
Date: 2014-01-22 17:07:05 +0100
From: @sjoerdmullender
Since the server does start up, I'd say we can drop the "critical" designation. The bug seems to be about slow startup, not about no startup.
## Comment 19491
Date: 2014-01-22 17:32:23 +0100
From: Christian Braun <<hcb>>
A 20 hour startup time is still critical to me. It is not much better then a not starting at all. Makes the new release unusable for me.
## Comment 19492
Date: 2014-01-22 17:39:10 +0100
From: @sjoerdmullender
The question is, does the 20 hours only happen the for the first restart or for all subsequent restarts?
## Comment 19493
Date: 2014-01-22 20:05:47 +0100
From: Christian Braun <<hcb>>
I did a restart 8 hours ago and it is still not finished. So unfortunately not faster on restart.
## Comment 19494
Date: 2014-01-22 20:14:52 +0100
From: @mlkersten
Could you take a sample of an strace call,
it may show what the system is doing.
## Comment 19495
Date: 2014-01-22 20:26:54 +0100
From: Christian Braun <<hcb>>
Created attachment 257
strace -p 6851 -tt -o strace2.log
> Attached file: [strace2.log.gz](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3420_strace2.log.gz_257) (application/x-gzip, 964568 bytes)
> Description: strace -p 6851 -tt -o strace2.log
## Comment 19496
Date: 2014-01-22 20:27:29 +0100
From: Christian Braun <<hcb>>
Comment on attachment 257
strace -p 6851 -tt -o strace2.log
strace -p 6851 -tt -o strace2.log
## Comment 19497
Date: 2014-01-22 20:42:25 +0100
From: @mlkersten
Thank you, that trace is very informative and may help to chase the bug quickly.
It looks like during recovery files are opened/closed way too many times.
As if individual updates leads to fsync requests.
To be analysed by the experts.
## Comment 19498
Date: 2014-01-24 09:14:38 +0100
From: @sjoerdmullender
Can you build your own version of the server from the sources? Or aternatively, can you tell me the distribution (and version) you're using so that I can build a set of packages for you?
We can't get a handle on what the server is doing, so we've extended the debug flags a little, and we would like you to start the server with those flags set. You would need to look in the merovingian.log file to see the exact command used to start the server, and copy that to a shell command line. Then add the flags
--set sql_debug=3
to the command line option and save the error output in a file (2> file).
If you can build yourself, do so with configure flags --enable-debug --disable-optimize --enable-assert so that the debugger can make sense of the binary.
## Comment 19500
Date: 2014-01-24 13:55:38 +0100
From: Christian Braun <<hcb>>
I am using Debian wheezy and the packages from:
deb http://dev.monetdb.org/downloads/testing/deb/ wheezy monetdb
I downloaded your source package, changed the 3 options in the debian/rules file and rebuild with dpkg-buildpackage. But i am not getting any extra log message on the console or in merovingian.log.
$ /usr/bin/mserver5 --dbpath=/var/monetdb/dbfarm/db --set merovingian_uri=mapi:monetdb://etna:50000/db --set mapi_open=false --set mapi_port=0 --set mapi_usock=/var/monetdb/dbfarm/db/.mapi.sock --set monet_vault_key=/var/monetdb/dbfarm/db/.vaultkey --set gdk_nr_threads=8 --set max_clients=64 --set sql_optimizer=default_pipe --set monet_daemon=yes --set sql_debug=3
MonetDB 5 server v11.17.1 "Jan2014"
Serving database 'db', using 8 threads
Compiled for x86_64-pc-linux-gnu/64bit with 64bit OIDs dynamically linked
Found 47.263 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 UNIX domain connection requests on mapi:monetdb:///var/monetdb/dbfarm/db/.mapi.sock
$ mserver5 --version --dbname=db
MonetDB 5 server v11.17.1 "Jan2014" (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 47.3GiB available memory, 8 available cpu cores
Libraries:
libpcre: 8.30 2012-02-04 (compiled with 8.30)
openssl: OpenSSL 1.0.1e 11 Feb 2013 (compiled with )
libxml2: 2.8.0 (compiled with 2.8.0)
Compiled by: root@wheezy (x86_64-pc-linux-gnu)
Compilation: gcc -g
Linking : /usr/bin/ld -m elf_x86_64
## Comment 19501
Date: 2014-01-24 14:01:16 +0100
From: @sjoerdmullender
The source package doesn't have the debug flags yet. You need to get the sources from our Mercurial repository. Use
hg clone -u Jan2014 http://dev.monetdb.org/hg/MonetDB/
Then compile using
./bootstrap
./configure --enable-debug --enable-assert --disable-optimize --prefix=...
make
make install
Or else I can build a package for you.
## Comment 19502
Date: 2014-01-24 14:02:44 +0100
From: Christian Braun <<hcb>>
Created attachment 258
strace -p 3039 -tt -o strace3.log
> Attached file: [strace3.log.gz](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3420_strace3.log.gz_258) (application/x-gzip, 310521 bytes)
> Description: strace -p 3039 -tt -o strace3.log
## Comment 19503
Date: 2014-01-24 15:21:05 +0100
From: Christian Braun <<hcb>>
Created attachment 259
mserver5 --set sql_debug #3
> Attached file: [debug1.txt.bz2](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3420_debug1.txt.bz2_259) (application/x-redhat-package-manager, 889852 bytes)
> Description: mserver5 --set sql_debug #3
## Comment 19504
Date: 2014-01-24 15:25:15 +0100
From: Christian Braun <<hcb>>
I followed your instructions. Attached is what i got from mserver5. I add a timestamp to the output. Table and column names i had to replace with a placeholder. The process is still running and loading a column every ~2 seconds.
## Comment 19505
Date: 2014-01-24 15:57:19 +0100
From: @sjoerdmullender
Thanks to the debug output we now have a very plausible explanation of why things are so slow, and also why the traces look like they do (the traces show a BAT being created with an OID head column that is quickly destroyed again).
The cause is that you have probably at some time during the life of the database dropped a column or table. We should deal better with this situation.
Hopefully we'll now be able to fix this problem, and then I will create a new release candidate next week.
## Comment 19506
Date: 2014-01-25 12:05:34 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [fa8c4a05a9f6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=fa8c4a05a9f6) 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=fa8c4a05a9f6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=fa8c4a05a9f6)
Changeset description:
cache bat access during catalog loading, solves bug #3420
## Comment 19507
Date: 2014-01-28 10:06:24 +0100
From: Christian Braun <<hcb>>
Thank you. With v11.17.3 "Jan2014" my database starts up in 60 seconds.
## Comment 19508
Date: 2014-01-28 10:14:34 +0100
From: @sjoerdmullender
Thanks for the feedback, and also thanks for the help nailing this problem.
I think we can now consider this bug fixed.
## Comment 19607
Date: 2014-02-20 15:02:48 +0100
From: @sjoerdmullender
Jan2014 has been released.
| Database does not start after upgrade | https://api.github.com/repos/MonetDB/MonetDB/issues/3420/comments | 0 | 2020-11-30T12:37:02Z | 2024-06-27T12:02:09Z | https://github.com/MonetDB/MonetDB/issues/3420 | 753,443,742 | 3,420 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-01-15 15:18:35 +0100
From: Christian Braun <<hcb>>
To: SQL devs <<bugs-sql>>
Version: 11.17.9 (Jan2014)
Last updated: 2014-02-20 15:02:34 +0100
## Comment 19451
Date: 2014-01-15 15:18:35 +0100
From: Christian Braun <<hcb>>
User-Agent: Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:26.0) Gecko/20100101 Firefox/26.0
Build Identifier:
After upgrading from 'MonetDB 5 server v11.15.19 "Feb2013-SP6"' to the release candidate, the database does not start anymore:
2014-01-15 15:13:44 MSG control[14903]: (local): served status list
2014-01-15 15:13:44 MSG merovingian[14903]: starting database 'db', up min/avg/max: 0s/3d/1w, crash average: 0.00 0.10 0.27 (70-60=10)
2014-01-15 15:13:45 MSG db[16499]: arguments: /usr/bin/mserver5 --dbpath=/var/monetdb/dbfarm/db --set merovingian_uri=mapi:monetdb://etna:50000/db --set mapi_open=false --set mapi_port=0 --set mapi_usock=/var/monetdb/dbfarm/db/.mapi.sock --set monet_vault_key=/var/monetdb/dbfarm/db/.vaultkey --set gdk_nr_threads=8 --set max_clients=64 --set sql_optimizer=default_pipe --set monet_daemon=yes
2014-01-15 15:13:46 MSG merovingian[14903]: sending process 16499 (database 'db') the TERM signal
2014-01-15 15:13:46 MSG db[16499]: MonetDB 5 server v11.17.1 "Jan2014"
2014-01-15 15:13:46 MSG db[16499]: Serving database 'db', using 8 threads
2014-01-15 15:13:46 MSG db[16499]: Compiled for x86_64-pc-linux-gnu/64bit with 64bit OIDs dynamically linked
2014-01-15 15:13:46 MSG db[16499]: Found 47.263 GiB available main-memory.
2014-01-15 15:13:46 MSG db[16499]: Copyright (c) 1993-July 2008 CWI.
2014-01-15 15:13:46 MSG db[16499]: Copyright (c) August 2008-2014 MonetDB B.V., all rights reserved
2014-01-15 15:13:46 MSG db[16499]: Visit http://www.monetdb.org/ for further information
2014-01-15 15:13:46 MSG db[16499]: !SyntaxException:parseError:include sysmon;
2014-01-15 15:13:46 MSG db[16499]: !SyntaxException:parseError:^MALException:malInclude:could not open file: querylog
2014-01-15 15:13:46 MSG db[16499]: !SyntaxException:parseError:include sysmon;
2014-01-15 15:13:46 MSG db[16499]: !SyntaxException:parseError: ^';' expected
2014-01-15 15:13:46 MSG merovingian[14903]: database 'db' (16499) has exited with exit status 0
2014-01-15 15:13:46 MSG merovingian[14903]: database 'db' has shut down
2014-01-15 15:13:46 ERR control[14903]: (local): failed to fork mserver: database 'db' started up, but failed to open up a communication channel
Reproducible: Always
## Comment 19452
Date: 2014-01-15 15:42:36 +0100
From: @sjoerdmullender
It's not just an upgraded database that doesn't work. Also a new one doesn't work.
The problem turns out to be a missing file from the Debian/Ubuntu packages. The file /usr/lib/monetdb5/querylog.mal is missing.
You could try copying the file from the sources (e.g., http://dev.monetdb.org/hg/MonetDB/file/0193a77a33c5/monetdb5/modules/mal/querylog.mal then click on "raw" in the left margin) and install it in the location mentioned.
## Comment 19453
Date: 2014-01-15 15:44:39 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [4d42c99421c6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=4d42c99421c6) 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=4d42c99421c6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=4d42c99421c6)
Changeset description:
Add missing file in Debian/Ubuntu packages.
This fixes bug #3419.
## Comment 19454
Date: 2014-01-15 15:45:18 +0100
From: @sjoerdmullender
File is now packaged.
## Comment 19455
Date: 2014-01-15 18:07:10 +0100
From: Christian Braun <<hcb>>
Thanks. The error messages are gone. I can create and start a new database. But my existing database does not start. In the log i have:
2014-01-15 17:36:45 MSG merovingian[19494]: starting database 'db', up min/avg/max: 0s/3d/1w, crash average: 0.00 0.00 0.20 (77-67=10)
2014-01-15 17:36:45 MSG db[20059]: arguments: /usr/bin/mserver5 --dbpath=/var/monetdb/dbfarm/db --set merovingian_uri=mapi:monetdb://etna:50000/db --set mapi_open=false --set mapi_port=0 --set mapi_usock=/var/monetdb/dbfarm/db/.mapi.sock --set monet_vault_key=/var/monetdb/dbfarm/db/.vaultkey --set gdk_nr_threads=8 --set max_clients=64 --set sql_optimizer=default_pipe --set monet_daemon=yes
The mserver just hangs and uses 100% cpu on one core with no disk activity. It is doing that for 30 minutes now. Is there anything else i can try to start the database?
## Comment 19599
Date: 2014-02-20 15:02:34 +0100
From: @sjoerdmullender
Jan2014 has been released.
| Database does not start after upgrade | https://api.github.com/repos/MonetDB/MonetDB/issues/3419/comments | 0 | 2020-11-30T12:36:59Z | 2024-06-27T12:02:08Z | https://github.com/MonetDB/MonetDB/issues/3419 | 753,443,716 | 3,419 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-01-14 15:31:17 +0100
From: Alex Shestakov <<shura.shestakov>>
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.15.19 (Feb2013-SP6)
CC: @mlkersten, @drstmane
Last updated: 2014-02-20 15:02:45 +0100
## Comment 19446
Date: 2014-01-14 15:31:17 +0100
From: Alex Shestakov <<shura.shestakov>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/31.0.1650.63 Safari/537.36
Build Identifier:
I have a table DailyData defined as follows.
create table DailyData (
sym varchar(32),
businessdate date,
closePrice float,
adjustedClosePrice float,
openingPrice float,
accVol float,
unique(businessdate, sym)
);
I make the following query.
with dd as (select * from DailyData where businessdate > '2013-06-01'),
r as (select t.businessdate, t.sym, 1 - t.adjustedcloseprice / y.adjustedcloseprice as ret from dd t
left outer join dd y on t.businessdate = y.businessdate + (interval '1' day) and t.sym = y.sym),
v as (select sym, stddev_samp(ret) as volatility from r group by sym)
select * from v where v.sym like '%.L' limit 100;
It results in a segmentation fault:
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7f5f17023700 (LWP 25178)]
0x0000003ee9327a9a in __strcmp_sse42 () from /lib64/libc.so.6
(gdb) where
0 0x0000003ee9327a9a in __strcmp_sse42 () from /lib64/libc.so.6
1 0x00007f6f2d4ef6f8 in rel_use_index (changes=<value optimized out>, sql=0x7f5f08005470, rel=0x7f5f0850aa70) at rel_optimizer.c:4829
2 0x00007f6f2d4ea225 in rewrite (sql=0x7f5f08005470, rel=0x7f5f0850aa70, rewriter=0x7f6f2d4ef610 <rel_use_index>, has_changes=0x7f5f17022cac) at rel_optimizer.c:6106
3 0x00007f6f2d4ea2bc in rewrite (sql=0x7f5f08005470, rel=0x7f5f0850cb40, rewriter=0x7f6f2d4ef610 <rel_use_index>, has_changes=0x7f5f17022cac) at rel_optimizer.c:6093
4 0x00007f6f2d4ea2bc in rewrite (sql=0x7f5f08005470, rel=0x7f5f0850ec30, rewriter=0x7f6f2d4ef610 <rel_use_index>, has_changes=0x7f5f17022cac) at rel_optimizer.c:6093
5 0x00007f6f2d4ea2bc in rewrite (sql=0x7f5f08005470, rel=0x7f5f0850ef50, rewriter=0x7f6f2d4ef610 <rel_use_index>, has_changes=0x7f5f17022cac) at rel_optimizer.c:6093
6 0x00007f6f2d4ea2bc in rewrite (sql=0x7f5f08005470, rel=0x7f5f08512a90, rewriter=0x7f6f2d4ef610 <rel_use_index>, has_changes=0x7f5f17022cac) at rel_optimizer.c:6093
7 0x00007f6f2d4ea2bc in rewrite (sql=0x7f5f08005470, rel=0x7f5f0850feb0, rewriter=0x7f6f2d4ef610 <rel_use_index>, has_changes=0x7f5f17022cac) at rel_optimizer.c:6093
8 0x00007f6f2d4ea2bc in rewrite (sql=0x7f5f08005470, rel=0x7f5f085101c0, rewriter=0x7f6f2d4ef610 <rel_use_index>, has_changes=0x7f5f17022cac) at rel_optimizer.c:6093
9 0x00007f6f2d4ecf58 in _rel_optimizer (sql=0x7f5f08005470, rel=<value optimized out>, level=0) at rel_optimizer.c:6278
10 0x00007f6f2d484369 in sql_symbol2relation (c=0x7f5f08005470, sym=<value optimized out>) at sql.mx:1755
11 0x00007f6f2d454b2f in SQLparser (c=0x7f6f2df49308) at sql_scenario.c:1708
12 0x00007f6f352ecdd3 in runPhase (c=0x7f6f2df49308) at mal_scenario.c:522
13 runScenarioBody (c=0x7f6f2df49308) at mal_scenario.c:558
14 0x00007f6f352ecf1f in runScenario (c=0x7f6f2df49308) at mal_scenario.c:586
15 0x00007f6f352ed778 in MSserveClient (dummy=0x7f6f2df49308) at mal_session.c:446
16 0x0000003ee9a07851 in start_thread () from /lib64/libpthread.so.0
17 0x0000003ee92e890d in clone () from /lib64/libc.so.6
Apparently at rel_optimizer.c:4829 strcmp(exp_relname(nre) returns NULL which is not checked.
If I rewrite the query in the following way it works if not very fast.
with r as (select t.businessdate, t.sym, 1 - t.adjustedcloseprice / y.adjustedcloseprice as ret from DailyData t
left outer join DailyData y on t.businessdate = y.businessdate + (interval '1' day) and t.sym = y.sym
where t.businessdate > '2013-06-01'),
v as (select sym, stddev_samp(ret) as volatility from r group by sym)
select * from v where v.sym like '%.L' limit 100;
Reproducible: Always
## Comment 19447
Date: 2014-01-14 15:58:19 +0100
From: @mlkersten
Not confirmed on the main trunk.
## Comment 19448
Date: 2014-01-14 16:54:00 +0100
From: @drstmane
Confirmed (i.e., segfault does occur) with Feb2013-SP6.
## Comment 19449
Date: 2014-01-14 16:55:30 +0100
From: @drstmane
Query appears to work fine (on empty table) with the Jan2014 release candidate branch.
## Comment 19450
Date: 2014-01-15 14:50:06 +0100
From: Alex Shestakov <<shura.shestakov>>
I can confirm that the bug is fixed in Jan2014 release candidate.
## Comment 19465
Date: 2014-01-20 15:16:47 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [1bcf7aae3e03](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=1bcf7aae3e03) 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=1bcf7aae3e03](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=1bcf7aae3e03)
Changeset description:
Added test for bug #3418.
## Comment 19467
Date: 2014-01-20 15:20:21 +0100
From: @sjoerdmullender
Bug seems to be fixed in the upcoming release.
## Comment 19606
Date: 2014-02-20 15:02:45 +0100
From: @sjoerdmullender
Jan2014 has been released.
| Segmentation fault on a query from table expression | https://api.github.com/repos/MonetDB/MonetDB/issues/3418/comments | 0 | 2020-11-30T12:36:56Z | 2024-06-27T12:02:07Z | https://github.com/MonetDB/MonetDB/issues/3418 | 753,443,680 | 3,418 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-01-09 12:51:02 +0100
From: @hannesmuehleisen
To: SQL devs <<bugs-sql>>
Version: 11.15.11 (Feb2013-SP3)
CC: @njnes
Last updated: 2014-02-20 15:02:52 +0100
## Comment 19442
Date: 2014-01-09 12:51:02 +0100
From: @hannesmuehleisen
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/32.0.1700.68 Safari/537.36
Build Identifier:
I have a two-layers deep common table expression query which does not work. See the attached synthetic example which provokes the bug.
Reproducible: Always
### Steps to Reproduce:
CREATE TABLE wtf(a int, b int, c int);
INSERT INTO wtf VALUES(1, 2, 3);
INSERT INTO wtf VALUES(4, 5, 6);
WITH a AS
(SELECT a,
b
FROM wtf
WHERE a IN (1,
4)),
b AS
(SELECT a,
b,
count(*) AS d
FROM a
GROUP BY a,
b),
c AS
(SELECT a,
count(*) AS e
FROM b
GROUP BY b)
SELECT *
FROM c;
### Actual Results:
Crash or no reaction from server
### Expected Results:
A result set or an error.
PLAN works, EXPLAIN also leads to crash.
PLAN output:
+---------------------------------------------------------------+
| rel |
+===============================================================+
| project ( |
| | group by ( |
| | | project ( |
| | | | group by ( |
| | | | | project ( |
| | | | | | select ( |
| | | | | | | table(sys.wtf) [ wtf.a, wtf.b ] COUNT |
| | | | | | ) [ wtf.a in (int[tinyint "1"], int[tinyint "4"]) ] |
| | | | | ) [ wtf.a as a.a, wtf.b as a.b ] |
| | | | ) [ a.a, a.b ] [ a.a, a.b ] |
| | | ) [ a.a as b.a, a.b as b.b ] |
| | ) [ b.b ] [ b.b, sys.count() NOT NULL as L2.L2 ] |
| ) [ b.a as c.a, L2 NOT NULL as c.e ] |
+---------------------------------------------------------------+
## Comment 19443
Date: 2014-01-10 17:00:20 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [6e845d6414e1](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=6e845d6414e1) 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=6e845d6414e1](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=6e845d6414e1)
Changeset description:
fix Bug #3417 - Nested Common Table Expressions Crash (edit)
properly check if a column result is a group by expression (or an aggregate)
## Comment 19444
Date: 2014-01-10 17:00:23 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [11ae2f3582c6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=11ae2f3582c6) 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=11ae2f3582c6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=11ae2f3582c6)
Changeset description:
added test for Bug #3417
## Comment 19445
Date: 2014-01-10 17:00:58 +0100
From: @njnes
fixed by better checking if a projection column really is a group by or
aggregate expression.
## Comment 19610
Date: 2014-02-20 15:02:52 +0100
From: @sjoerdmullender
Jan2014 has been released.
| Nested Common Table Expressions Crash | https://api.github.com/repos/MonetDB/MonetDB/issues/3417/comments | 0 | 2020-11-30T12:36:54Z | 2024-06-27T12:02:06Z | https://github.com/MonetDB/MonetDB/issues/3417 | 753,443,656 | 3,417 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-01-07 17:01:49 +0100
From: @hannesmuehleisen
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @kutsurak
Last updated: 2019-04-30 12:36:04 +0200
## Comment 19441
Date: 2014-01-07 17:01:49 +0100
From: @hannesmuehleisen
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/32.0.1700.68 Safari/537.36
Build Identifier:
While it is great to have a natural logarithm function in MonetDB, sometimes other bases are required. For example, MySQL and SQL Server support a second parameter to LOG(), which defines the base.
Reproducible: Always
### Steps to Reproduce:
1. SELECT LOG(42,2);
### Actual Results:
SELECT: no such binary operator 'log(tinyint,tinyint)'
### Expected Results:
5.392317
## Comment 21238
Date: 2015-08-31 15:51:24 +0200
From: @kutsurak
Created attachment 348
Proposed patch for the enhancement
Hi,
this patch adds the functionality described in this issue. Hopefully you'll find it useful. I have only tried this on Linux, but I don't expect it to create problems on other platforms.
Best regards,
Panos.
> Attached file: [log_base.patch](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3416_log_base.patch_348) (text/plain, 5077 bytes)
> Description: Proposed patch for the enhancement
## Comment 21239
Date: 2015-08-31 16:06:28 +0200
From: @kutsurak
Created attachment 349
Corrected patch
I think I made a mistake with the previous patch (I used hg log -r <rev>), and it did not keep the original commit message. This one was created using hg export, so hopefully it should be OK.
Sorry about this, I am new with Mercurial :)
Best regards,
Panos.
> Attached file: [log_base.patch](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3416_log_base.patch_349) (text/plain, 5214 bytes)
> Description: Corrected patch
## Comment 26837
Date: 2019-01-22 15:07:27 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [a116ae0c1407](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=a116ae0c1407) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=a116ae0c1407](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=a116ae0c1407)
Changeset description:
Implemented a two argument log function.
This fixed bug #3416.
| log() SQL function only supports natural logarithm | https://api.github.com/repos/MonetDB/MonetDB/issues/3416/comments | 0 | 2020-11-30T12:36:52Z | 2024-06-27T12:02:05Z | https://github.com/MonetDB/MonetDB/issues/3416 | 753,443,638 | 3,416 |
[
"MonetDB",
"MonetDB"
] | Date: 2014-01-02 06:27:52 +0100
From: lsf <<shuifeng.lsf>>
To: clients devs <<bugs-clients>>
Version: 11.21.19 (Jul2015-SP4)
CC: arkr17997, martin.van.dinther
Last updated: 2019-02-12 14:01:45 +0100
## Comment 19437
Date: 2014-01-02 06:27:52 +0100
From: lsf <<shuifeng.lsf>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/31.0.1650.63 Safari/537.36
Build Identifier:
Exception in thread "main" java.lang.IndexOutOfBoundsException
at java.nio.StringCharBuffer.subSequence(StringCharBuffer.java:92)
at nl.cwi.monetdb.mcl.parser.StartOfHeaderParser.getNextAsString(StartOfHeaderParser.java:146)
at nl.cwi.monetdb.jdbc.MonetConnection$ResponseList.executeQuery(MonetConnection.java:2394)
at nl.cwi.monetdb.jdbc.MonetConnection$ResponseList.processQuery(MonetConnection.java:2284)
at nl.cwi.monetdb.jdbc.MonetStatement.internalExecute(MonetStatement.java:508)
at nl.cwi.monetdb.jdbc.MonetStatement.execute(MonetStatement.java:349)
at nl.cwi.monetdb.jdbc.MonetStatement.executeUpdate(MonetStatement.java:544)
at MJDBCTest.main(MJDBCTest.java:79)
Reproducible: Always
### Steps to Reproduce:
1. getConnection
2. createStatement
3. Statement. executeUpdate(DML)
RD 1388402803226: read final block: 69 bytes
RX 1388402803227: QxD2lyxtdlzL3PygE:merovingian:9:RIPEMD160,SHA256,SHA1,MD5:LIT:SHA512:
RD 1388402803227: inserting prompt
TD 1388402803240: write final block: 102 bytes
TX 1388402803240: BIG:merovingian:{SHA256}44513d504b6cc87b499c918c6749847d0365afa07d104ee0b555b221b1f2ab16:sql:cdo_xetl:
RD 1388402803240: read final block: 44 bytes
RX 1388402803240: ^mapi:merovingian://proxy?database=cdo_xetl
RD 1388402803240: inserting prompt
RD 1388402803241: read final block: 57 bytes
RX 1388402803241: shsTmw0VM:mserver:9:RIPEMD160,SHA256,SHA1,MD5:LIT:SHA512:
RD 1388402803241: inserting prompt
TD 1388402803242: write final block: 98 bytes
TX 1388402803242: BIG:monetdb:{SHA256}6f9654f9583ea53cdc0c5da878aba901247e13ba8c2c98e1dc8d05d38da8760c:sql:cdo_xetl:
RD 1388402803243: read final block: 0 bytes
RX 1388402803243:
RD 1388402803243: inserting prompt
TD 1388402803250: write final block: 48 bytes
TX 1388402803250: sSET TIME ZONE INTERVAL '+08:00' HOUR TO MINUTE;
RD 1388402803251: read final block: 3 bytes
RX 1388402803251: &3
RD 1388402803251: inserting prompt
TD 1388402803256: write final block: 15 bytes
TX 1388402803256: Xreply_size 250
RD 1388402803256: read final block: 0 bytes
RX 1388402803256:
RD 1388402803256: inserting prompt
TD 1388402803256: write final block: 429 bytes
TX 1388402803256: screate table fdi_order (order_id varchar(200), buyer_nid varchar(200), buyer_nick varchar(200) , seller_nid varchar(200) , seller_nick varchar(200) , category_level1_id varchar(200) , category_level2_id varchar(200) , category_level1_name varchar(200) , category_level2_name varchar(200) , pay_flag varchar(200) , gmv_flag varchar(200) , order_bctype_id varchar(200) , pay_amt double , gmv_amt double, order_auction_qty bigint);
RD 1388402803262: read final block: 3 bytes
RX 1388402803262: &3
RD 1388402803262: inserting prompt
TD 1388402803264: write final block: 320 bytes
TX 1388402803264: sinsert into fdi_order values ('order_id_value', 'buyer_nid_value', 'buyer_nick_value', 'seller_nid_value', 'seller_nick_value', 'category_level1_id_value', 'category_level2_id_value', 'category_level1_name_value', 'category_level2_name_value', 'pay_flag_value', 'gmv_flag_value', 'order_bctype_id_value', 1.6, 1.0, 1);;
RD 1388402803266: read final block: 8 bytes
RX 1388402803266: &2 1 -1
RD 1388402803266: inserting prompt
## Comment 19438
Date: 2014-01-02 06:30:43 +0100
From: lsf <<shuifeng.lsf>>
jdbc version is monetdb-jdbc-2.10.jar, this works on my windows but not on my linux.
## Comment 19439
Date: 2014-01-02 08:41:40 +0100
From: @grobian
which Java version is this? (java -version will do)
## Comment 19440
Date: 2014-01-02 09:15:31 +0100
From: lsf <<shuifeng.lsf>>
(In reply to comment 2)
> which Java version is this? (java -version will do)
java version "1.6.0_16"
Java(TM) SE Runtime Environment (build 1.6.0_16-b01)
Java HotSpot(TM) 64-Bit Server VM (build 14.2-b01, mixed mode)
## Comment 24512
Date: 2016-10-13 19:53:33 +0200
From: Martin van Dinther <<martin.van.dinther>>
Can you reproduce this problem with the latest MonetDB JDBC driver: monetdb-jdbc-2.23.jar
It is downloadable from: https://www.monetdb.org/downloads/Java/Latest/
If you can reproduce it, please share the Java program code, so we can analyse what is the root cause.
Else we will close this old issue.
## Comment 25147
Date: 2017-03-09 18:26:29 +0100
From: Martin van Dinther <<martin.van.dinther>>
We cannot reproduce this old issue anymore.
| An IndexOutOfBoundsException was thrown when statement.executeUpdate is invoked | https://api.github.com/repos/MonetDB/MonetDB/issues/3415/comments | 0 | 2020-11-30T12:36:48Z | 2024-06-28T07:19:42Z | https://github.com/MonetDB/MonetDB/issues/3415 | 753,443,599 | 3,415 |
[
"MonetDB",
"MonetDB"
] | Date: 2013-12-27 23:44:39 +0100
From: Gennadiy <<gennadiy1g>>
To: Merovingian devs <<bugs-merovingian>>
Version: 11.21.19 (Jul2015-SP4)
Last updated: 2017-05-19 10:59:19 +0200
## Comment 19434
Date: 2013-12-27 23:44:39 +0100
From: Gennadiy <<gennadiy1g>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:26.0) Gecko/20100101 Firefox/26.0
Build Identifier:
The following statement causes mserver5.exe to crash:
CREATE VIEW v_models_that_satisfy_criteria_min_set_dup AS
SELECT a.*
FROM v_models_that_satisfy_criteria_dup AS a
WHERE a.set_number = (
SELECT MIN(b.set_number)
FROM v_models_that_satisfy_criteria_dup AS b
WHERE b.model_number = a.model_number
);
Reproducible: Always
### Steps to Reproduce:
1. Create table v_models_that_satisfy_criteria_dup.
2. Populate it with data.
3. Execute CREATE VIEW statement above.
MonetDB 5 server v11.15.19 "Feb2013-SP6"
Serving database 'soft10_20131227_152326482', using 4 threads
Compiled for x86_64-pc-winnt/64bit with 64bit OIDs dynamically linked
Found 3.799 GiB available main-memory.
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2013 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/JAQL module loaded
MonetDB/SQL module loaded
>
## Comment 19435
Date: 2013-12-27 23:45:36 +0100
From: Gennadiy <<gennadiy1g>>
Created attachment 252
SQL script to create and populate a table
> Attached file: [v_models_that_satisfy_criteria_dup_201312271710.sql](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3414_v_models_that_satisfy_criteria_dup_201312271710.sql_252) (text/plain, 27446 bytes)
> Description: SQL script to create and populate a table
## Comment 19436
Date: 2013-12-27 23:47:26 +0100
From: Gennadiy <<gennadiy1g>>
Created attachment 253
Screenshot
> Attached file: [screenshot.4.jpg](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3414_screenshot.4.jpg_253) (image/jpeg, 729478 bytes)
> Description: Screenshot
## Comment 19521
Date: 2014-02-04 10:20:03 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [246635fc47aa](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=246635fc47aa) 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=246635fc47aa](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=246635fc47aa)
Changeset description:
Added test for bug #3414.
## Comment 19550
Date: 2014-02-05 20:12:08 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [a7880f7ebea3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=a7880f7ebea3) 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=a7880f7ebea3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=a7880f7ebea3)
Changeset description:
fixed bug #3414, ie crash because apply wasn't rewriten
## Comment 25344
Date: 2017-05-19 10:59:19 +0200
From: @sjoerdmullender
Fixed a long time ago.
| Server crash on CREATE VIEW with a correlated subquery | https://api.github.com/repos/MonetDB/MonetDB/issues/3414/comments | 0 | 2020-11-30T12:36:46Z | 2024-06-27T12:02:03Z | https://github.com/MonetDB/MonetDB/issues/3414 | 753,443,579 | 3,414 |
[
"MonetDB",
"MonetDB"
] | Date: 2013-12-26 14:13:19 +0100
From: Jeet <<jeet.kaushik>>
To: GDK devs <<bugs-common>>
Version: 11.21.19 (Jul2015-SP4)
CC: jeet.kaushik, @mlkersten
Last updated: 2016-04-11 11:22:54 +0200
## Comment 19429
Date: 2013-12-26 14:13:19 +0100
From: Jeet <<jeet.kaushik>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/31.0.1650.63 Safari/537.36
Build Identifier:
Data is imported using COPY command and database is running for few days and at some point after restart database corrupts. all the column values are reset to 0 or negative values.
Reproducible: Couldn't Reproduce
### Steps to Reproduce:
1. Use copy command to insert data.
2. restart DB after sometime.
3. At some point DB corrupts.
### Actual Results:
All column values are reset to 0 and datetime column value rest to "-1-01-01 00:00:00.000000" value.
### Expected Results:
Data should not corrupts. column values should be retained as actual.
## Comment 19430
Date: 2013-12-26 14:24:16 +0100
From: @mlkersten
This is a hard one to debug. Can you report on what the merovingian.log says before. It may contain a hint on the status at/before crash.
Is your disk full? Or was your disk almost full before you observed the crash?
## Comment 19431
Date: 2013-12-27 05:43:09 +0100
From: Jeet <<jeet.kaushik>>
Created attachment 251
monetdb merovingian log file
> Attached file: [merovingian.log](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_3413_merovingian.log_251) (application/octet-stream, 130564 bytes)
> Description: monetdb merovingian log file
## Comment 19432
Date: 2013-12-27 08:07:52 +0100
From: @mlkersten
Comment on attachment 251
monetdb merovingian log file
Log does not show anything exceptional, except for the last line, which indicates that you had either a network problem, or there was already a mserver running. You might have accidentally connected to that one. That one, could have been a test version with empty db?
You should check also the free diskspace available? and be aware that a COPY into may require up to 3 x the source size: 1 for original csv, 1 for column store, and 1 for the recovery process (unless you used LOCKED)
## Comment 19433
Date: 2013-12-27 08:39:57 +0100
From: @grobian
The tail of the log shows merovingian "disappeared", at least it didn't stop correctly, then it was started. The order of events which lead to that "gap" may explain why the corruption happened.
| Database corrupted after few days on database restart | https://api.github.com/repos/MonetDB/MonetDB/issues/3413/comments | 0 | 2020-11-30T12:36:44Z | 2024-06-28T07:19:42Z | https://github.com/MonetDB/MonetDB/issues/3413 | 753,443,560 | 3,413 |
[
"MonetDB",
"MonetDB"
] | Date: 2013-12-12 21:30:29 +0100
From: ittaibalaban
To: SQL devs <<bugs-sql>>
Version: 11.15.19 (Feb2013-SP6)
CC: @njnes
Last updated: 2014-02-20 15:02:35 +0100
## Comment 19419
Date: 2013-12-12 21:30:29 +0100
From: ittaibalaban
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/31.0.1650.63 Safari/537.36
Build Identifier:
Given a boolean column X, using a WHERE clause of the form "X and <other-expr>" seems to cause <other-expr> to not be evaluated, whereas "X = true and <other-expr>" is evaluated correctly.
Reproducible: Always
### Steps to Reproduce:
1. create table boolean_test (b boolean not null, i int not null);
2. insert into boolean_test (b, i) values (true, 1), (true, 2), (true, 3), (true, 4);
3. select * from boolean_test where b and (i < 3)
### Actual Results:
All 4 rows are returned.
### Expected Results:
Only 2 rows should be returned, with i in (1,2)
The rewritten query "select * from boolean_test where b = true and (i < 3)" is evaluated correctly.
## Comment 19423
Date: 2013-12-13 15:24:28 +0100
From: @njnes
handle single 'boolean' column expressions properly
## Comment 19427
Date: 2013-12-13 16:08:45 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [482b6561151a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=482b6561151a) 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=482b6561151a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=482b6561151a)
Changeset description:
added tests for bugs 3411 and 3412
## Comment 19600
Date: 2014-02-20 15:02:35 +0100
From: @sjoerdmullender
Jan2014 has been released.
| Boolean expressions in WHERE clause, result in incorrect resulsts | https://api.github.com/repos/MonetDB/MonetDB/issues/3412/comments | 0 | 2020-11-30T12:36:42Z | 2024-06-27T12:02:00Z | https://github.com/MonetDB/MonetDB/issues/3412 | 753,443,540 | 3,412 |
[
"MonetDB",
"MonetDB"
] | Date: 2013-12-12 18:52:00 +0100
From: @swingbit
To: SQL devs <<bugs-sql>>
Version: 11.15.19 (Feb2013-SP6)
CC: @njnes
Last updated: 2014-02-20 15:02:34 +0100
## Comment 19418
Date: 2013-12-12 18:52:00 +0100
From: @swingbit
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/31.0.1650.57 Safari/537.36
Build Identifier:
set optimizer='sequential_pipe';
START TRANSACTION;
CREATE TABLE t (
"a" INT,
"b" INT
);
-- The following 3 queries are EQUIVALENT
explain
SELECT t2.a
FROM t t1, t t2
WHERE t2.a between t1.a and t1.b;
explain
SELECT t2.a
FROM t t1, t t2
WHERE t2.a >= t1.a
AND t2.a <= t1.b;
explain
SELECT t2.a
FROM t t1, t t2
WHERE t1.a <= t2.a
AND t1.b >= t2.a;
The last query, however, does not produce a (range)join.
The effect, on large data, is terrible.
This is somewhat related to bug #3259.
Here below the MAL plans for the three queries.
The first two are fine.
The LAST plan is correct but much slower than the other two.
sql>set optimizer='sequential_pipe';
operation successful (0.394ms)
sql>
sql>START TRANSACTION;
auto commit mode: off
sql>
sql>CREATE TABLE t (
more>"a" INT,
more>"b" INT
more>);
operation successful (0.847ms)
sql>
sql>explain
more>SELECT t2.a
more>FROM t t1, t t2
more>WHERE t2.a between t1.a and t1.b;
+-----------------------------------------------------------+
| mal |
+===========================================================+
| function user.s4_1():void; |
| X_1 := sql.mvc(); |
| X_2:bat[:oid,:oid] := sql.tid(X_1,"spinque","t"); |
| X_5 := sql.bind(X_1,"spinque","t","a",0); |
| X_8 := algebra.leftfetchjoin(X_2,X_5); |
| X_9 := sql.bind(X_1,"spinque","t","b",0); |
| X_11 := algebra.leftfetchjoin(X_2,X_9); |
| (X_12,r1_15) := algebra.join(X_8,X_8,X_11,true,true); |
| X_15 := algebra.leftfetchjoin(X_12,X_8); |
| X_16 := sql.resultSet(1,1,X_15); |
| sql.rsColumn(X_16,"spinque.t2","a","int",32,0,X_15); |
| X_22 := io.stdout(); |
| sql.exportResult(X_22,X_16); |
| end s4_1; |
+-----------------------------------------------------------+
14 tuples (1.291ms)
sql>
sql>explain
more>SELECT t2.a
more>FROM t t1, t t2
more>WHERE t2.a >= t1.a
more>AND t2.a <= t1.b;
+-----------------------------------------------------------+
| mal |
+===========================================================+
| function user.s5_1():void; |
| X_1 := sql.mvc(); |
| X_2:bat[:oid,:oid] := sql.tid(X_1,"spinque","t"); |
| X_5 := sql.bind(X_1,"spinque","t","a",0); |
| X_8 := algebra.leftfetchjoin(X_2,X_5); |
| X_9 := sql.bind(X_1,"spinque","t","b",0); |
| X_11 := algebra.leftfetchjoin(X_2,X_9); |
| (X_12,r1_15) := algebra.join(X_8,X_8,X_11,true,true); |
| X_15 := algebra.leftfetchjoin(X_12,X_8); |
| X_16 := sql.resultSet(1,1,X_15); |
| sql.rsColumn(X_16,"spinque.t2","a","int",32,0,X_15); |
| X_22 := io.stdout(); |
| sql.exportResult(X_22,X_16); |
| end s5_1; |
+-----------------------------------------------------------+
14 tuples (1.254ms)
sql>
sql>explain
more>SELECT t2.a
more>FROM t t1, t t2
more>WHERE t1.a <= t2.a
more>AND t1.b >= t2.a;
+------------------------------------------------------------------------+
| mal |
+========================================================================+
| function user.s6_1():void; |
| X_1 := sql.mvc(); |
| X_2:bat[:oid,:oid] := sql.tid(X_1,"spinque","t"); |
| X_5 := sql.bind(X_1,"spinque","t","a",0); |
| X_8 := algebra.leftfetchjoin(X_2,X_5); |
| (X_9,r1_12) := algebra.thetajoin(X_8,X_8,-2); |
| X_12 := sql.bind(X_1,"spinque","t","b",0); |
| X_14:bat[:oid,:int] := algebra.leftfetchjoinPath(X_9,X_2,X_12); |
| X_15 := algebra.leftfetchjoin(r1_12,X_8); |
| X_16:bat[:oid,:bit] := batcalc.>=(X_14,X_15); |
| X_17 := algebra.subselect(X_16,true,true,true,true,false); |
| X_20:bat[:oid,:int] := algebra.leftfetchjoinPath(X_17,r1_12,X_8); |
| X_21 := sql.resultSet(1,1,X_20); |
| sql.rsColumn(X_21,"spinque.t2","a","int",32,0,X_20); |
| X_27 := io.stdout(); |
| sql.exportResult(X_27,X_21); |
| end s6_1; |
+------------------------------------------------------------------------+
17 tuples (1.774ms)
sql>
Reproducible: Always
### Expected Results:
The pattern in the last query should be recognised as a BETWEEN clause and produce a rangejoin.
$ mserver5 --version
MonetDB 5 server v11.15.20 (64-bit, 64-bit oids)
This is an unreleased version
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2013 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: 8.32 2012-11-30 (compiled with 8.32)
openssl: OpenSSL 1.0.1e 11 Feb 2013 (compiled with OpenSSL 1.0.1e-fips 11 Feb 2013)
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 19424
Date: 2013-12-13 15:25:02 +0100
From: @njnes
added missing case in exp_merge_range
## Comment 19426
Date: 2013-12-13 16:08:44 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [482b6561151a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=482b6561151a) 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=482b6561151a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=482b6561151a)
Changeset description:
added tests for bugs 3411 and 3412
## Comment 19598
Date: 2014-02-20 15:02:34 +0100
From: @sjoerdmullender
Jan2014 has been released.
| (disguised) BETWEEN clause not recognised. Hence no rangejoin. | https://api.github.com/repos/MonetDB/MonetDB/issues/3411/comments | 0 | 2020-11-30T12:36:39Z | 2024-06-27T12:01:59Z | https://github.com/MonetDB/MonetDB/issues/3411 | 753,443,513 | 3,411 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.