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: 2009-10-05 01:15:07 +0200
From: @drstmane
To: Fabian Groffen <<fabian>>
Version: -- development
Last updated: 2009-11-15 09:22:00 +0100
## Comment 12496
Date: 2009-10-05 13:15:07 +0200
From: @drstmane
When using the "-s" option, "mclient -t" does not show the "Timer" wall-clock time:
$ mclient -t -ls -s'select 1+2;'
% . table_name
% sql_add_single_value name
% tinyint type
% 1 length
[ 3 ]
$ mclient -t -lx -s'1+2'
3
Trans 4.703 msec
Shred 0.000 msec
Query 5.981 msec
Print 0.458 msec
while it does when reading the query from a file or from stdin (interactive or pipe).
## Comment 12497
Date: 2009-10-05 15:00:26 +0200
From: @grobian
this feature has been implemented in the Nov2009 branch
## Comment 12498
Date: 2009-10-08 16:30:14 +0200
From: @sjoerdmullender
Fabian implemented this feature in the Nov2009 branch.
Since this is a new feature, it will not be implemented in the Aug2009 branch, so I changed the Group accordingly.
## Comment 12499
Date: 2009-11-15 21:22:00 +0100
From: @drstmane
added tests
MonetDB4/src/tools/Tests/mclient-t-s.MIL.py
MonetDB5/src/tools/Tests/mclient-t-s.MAL.py
sql/src/test/Tests/mclient-t-s.SQL.py
pathfinder/src/tools/Tests/mclient-t-s.XQUERY.py
Closing.
## Comment 12500
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2872884 at http://sourceforge.net/support/tracker.php?aid=2872884
| mclient -t -s... does not show "Timer" info | https://api.github.com/repos/MonetDB/MonetDB/issues/2210/comments | 0 | 2020-11-30T10:27:59Z | 2024-06-27T11:42:00Z | https://github.com/MonetDB/MonetDB/issues/2210 | 753,354,016 | 2,210 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-10-05 01:07:43 +0200
From: @drstmane
To: Fabian Groffen <<fabian>>
Version: -- development
Last updated: 2009-10-05 02:54:51 +0200
## Comment 12493
Date: 2009-10-05 13:07:43 +0200
From: @drstmane
Until the Aug2009 release (branch), when using the command line option
-s stmt | --statement=stmt run single statement
mclient quits after execution of the given statement, i.e., it does not show an interactive prompt,
unless also the command line option
-i | --interactive read stdin after command line args
is used.
As of the Nov2009 candidate branch, this behavior has changed, i.e., with "-s" mclient always shows the interactive prompt after execution of the given statement, regardless of whether "-i" is used, too, or not.
## Comment 12494
Date: 2009-10-05 14:54:50 +0200
From: @grobian
fixed in Nov2009 branch
## Comment 12495
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2872880 at http://sourceforge.net/support/tracker.php?aid=2872880
| mclient shows prompt with "-s" option without "-i" option | https://api.github.com/repos/MonetDB/MonetDB/issues/2209/comments | 0 | 2020-11-30T10:27:56Z | 2024-06-27T11:41:59Z | https://github.com/MonetDB/MonetDB/issues/2209 | 753,353,973 | 2,209 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-10-04 03:49:03 +0200
From: Mitja Kleider <<mkleider>>
To: @sjoerdmullender
Version: -- development
Last updated: 2009-10-05 06:15:22 +0200
## Comment 12488
Date: 2009-10-04 15:49:03 +0200
From: Mitja Kleider <<mkleider>>
building sql with bison++ fails because of warnings/errors in bison++, see attached output of `make`
System information:
Ubuntu 9.04 Jaunty Jackalope
gcc 4.3.3-5ubuntu4
flex-old 2.5.4a-7build1
bison++ 1.21.11-3
autoconf 2.63-2ubuntu1
automake 1:1.10.2-0ubuntu3
libtool 2.2.6a-1ubuntu1
## Comment 12489
Date: 2009-10-05 14:07:41 +0200
From: @sjoerdmullender
Fixed in CVS by making sure the generated *.tab.c file starts with including our config.h file.
No test needed.
## Comment 12490
Date: 2009-10-05 14:33:46 +0200
From: @sjoerdmullender
I closed this bug prematurely.
The error I got while compiling MonetDB4 using bison++ was fixed, but not the one that caused sql to not compile.
## Comment 12491
Date: 2009-10-05 18:15:11 +0200
From: @sjoerdmullender
By adding some defines to sql_parser.mx I can get this through bison++, except for one issue. The issue is a comparison between int and size_t (the result of sizeof). This comparison is in the generated code (i.e. not in our code), so this is not something we can fix. Since we do not support bison++, I am not willing to override the normal treat warnings as errors option that we give to the compiler. If you want to use bison++ you can override that option for this file.
I am closing this with a Wont Fix since the remaining problem is not in our code.
## Comment 12492
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2872565 at http://sourceforge.net/support/tracker.php?aid=2872565
| building sql with bison++ 1.21.11 fails | https://api.github.com/repos/MonetDB/MonetDB/issues/2208/comments | 0 | 2020-11-30T10:27:52Z | 2024-06-28T13:22:47Z | https://github.com/MonetDB/MonetDB/issues/2208 | 753,353,923 | 2,208 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-10-04 02:50:19 +0200
From: @skinkie
To: @njnes
Version: -- development
Last updated: 2009-10-05 08:41:42 +0200
## Comment 12485
Date: 2009-10-04 02:50:19 +0200
From: @skinkie
gdb --args /opt/monetdb-head/bin/mserver5 --dbinit="include sql; sql.start();" --dbname=monitor --readonly --set mapi_port=50001
GNU gdb 6.8
Copyright (C) 2008 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-pc-linux-gnu"...
(gdb) r
Starting program: /opt/monetdb-head/bin/mserver5 --dbinit=include\ sql\;\ sql.start\(\)\; --dbname=monitor --readonly --set mapi_port=50001
[Thread debugging using libthread_db enabled]
[New Thread 0x7f352fe126f0 (LWP 26973)]
[New Thread 0x7f35271ab910 (LWP 26976)]
MonetDB server v5.17.0, based on kernel v1.35.0
Serving database 'monitor', using 4 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs dynamically linked
Detected 3.856 GiB (4140695552 bytes) main-memory.
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2009 MonetDB B.V., all rights reserved
Visit http://monetdb.cwi.nl/ for further information
warning: please don't forget to set your vault key!
(see /opt/monetdb-head/etc/monetdb5.conf)
[New Thread 0x7f351dd49910 (LWP 26977)]
Listening for connection requests on mapi:monetdb://127.0.0.1:50001/
MonetDB/SQL module v2.35.0 loaded
[New Thread 0x7f351d5b9910 (LWP 26981)]
>create function "epoch"(sec INT) returns TIMESTAMP external name timestamp."epoch";
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7f352fe126f0 (LWP 26973)]
0x0000000000000000 in ?? ()
(gdb) bt
0 0x0000000000000000 in ?? ()
1 0x00007f351d8b33fd in table_insert (tr=0x22441e8, t=0x22cf768) at ../../../../src/storage/restrict/restrict_table.mx:152
2 0x00007f351d8a6f4e in sql_trans_create_func (tr=0x22441e8, s=0x2293fd8, func=0x22bc8a8 "epoch", args=0x215c598, res=0x22bca10, sql=0 '\0', aggr=0 '\0',
mod=0x22bca58 "timestamp", impl=0x22bca68 "epoch", is_func=1) at ../../../src/storage/store.mx:3266
3 0x00007f351d863373 in mvc_create_func (sql=0x22ba738, s=0x2293fd8, name=0x22bc8a8 "epoch", args=0x215c598, res=0x22bca10, issql=0 '\0', aggr=0 '\0',
mod=0x22bca58 "timestamp", impl=0x22bca68 "epoch", is_func=1) at ../../../src/server/sql_mvc.mx:999
4 0x00007f351d869efa in create_func (sql=0x22ba738, qname=0x22bc8b8, params=0x22bc918, res=0x22bca38, ext_name=0x22bca78, body=0x0, is_func=1, is_aggr=0)
at ../../../src/server/sql_psm.mx:662
5 0x00007f351d86a843 in psm (sql=0x22ba738, s=0x22bcc08) at ../../../src/server/sql_psm.mx:814
6 0x00007f351d8453ca in semantic (sql=0x22ba738, s=0x22bcc08) at ../../../src/server/sql_semantic.mx:1062
7 0x00007f351d84544a in output_semantic (sql=0x22ba738, s=0x22bcc08) at ../../../src/server/sql_semantic.mx:1097
8 0x00007f351d7e30c2 in sql_symbol2stmt (c=0x22ba738, sym=0x22bcc08) at ../../../../src/backends/monet5/sql.mx:1264
9 0x00007f351d81e535 in SQLparser (c=0x605540) at ../../../../src/backends/monet5/sql_scenario.mx:1032
10 0x00007f352f93c668 in runPhase (c=0x605540, phase=1) at ../../../src/mal/mal_scenario.mx:602
11 0x00007f352f93c75f in runScenarioBody (c=0x605540) at ../../../src/mal/mal_scenario.mx:638
12 0x00007f352f93c989 in runScenario (c=0x605540) at ../../../src/mal/mal_scenario.mx:672
13 0x00007f352f8f773d in MSserveClient (dummy=0x605540) at ../../../src/mal/mal_session.mx:496
14 0x00000000004031f5 in main (argc=6, av=0x7fffdeb55c28) at ../../../src/tools/mserver5.mx:636
I would not expect a segmentation fault when the database is in readonly state when a write action takes place. Rather an informed response such operation is not possible.
## Comment 12486
Date: 2009-10-05 08:41:42 +0200
From: @njnes
fixed by disallowing create/drop functions in readonly mode
## Comment 12487
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2872421 at http://sourceforge.net/support/tracker.php?aid=2872421
| SQL: Crash when trying to create function to read only db | https://api.github.com/repos/MonetDB/MonetDB/issues/2207/comments | 0 | 2020-11-30T10:27:49Z | 2024-06-27T11:41:57Z | https://github.com/MonetDB/MonetDB/issues/2207 | 753,353,873 | 2,207 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-10-01 07:58:07 +0200
From: @drstmane
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2009-10-06 12:37:46 +0200
## Comment 12479
Date: 2009-10-01 19:58:07 +0200
From: @drstmane
Most probably since it addition, test sql/src/test/BugTracker-2009/Tests/mclient-lsql-D.SQL.sh fails on all platform except 64-bit Fedora.
On all non-Fedora platform, it produces an error like
"
MAPI = monetdb@build-x86-macosx:37561
ACTION= mapi_fetch_line
QUERY = SELECT * FROM "sys"."n8"
ERROR = Connection terminated
MAPI = monetdb@build-x86-macosx:37561
ACTION= mapi_query
ERROR = Connection lost
"
cf., e.g.,
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.32.32.d.1-Debian4.0/src_test_BugTracker-2009/mclient-lsql-D.err.00.html
Additionally, (also on 32-bit Fedora, where no error occurs), it produce output of 12MB to 308MB(!) instead of the expected 3kB output:
-rw-r--r-- 1 monet monet 3.0K 2009-10-01 02:07 .GNU.64.64.d.0-Fedora10..2009.10.01/mTests10/src/test/BugTracker-2009/mclient-lsql-D.test.out
-rw-r--r-- 1 monet monet 3.0K 2009-10-01 01:46 .Int.64.64.d.1-Fedora10..2009.10.01/mTests10/src/test/BugTracker-2009/mclient-lsql-D.test.out
-rw-r--r-- 1 monet monet 3.0K 2009-10-01 01:47 .Int.64.32.d.1-Fedora10..2009.10.01/mTests10/src/test/BugTracker-2009/mclient-lsql-D.test.out
-rw-r--r-- 1 monet monet 3.0K 2009-10-01 02:10 .GNU.64.64.d.1-Fedora10..2009.10.01/mTests10/src/test/BugTracker-2009/mclient-lsql-D.test.out
-rw-r--r-- 1 monet monet 3.0K 2009-10-01 02:10 .GNU.64.32.d.1-Fedora10..2009.10.01/mTests10/src/test/BugTracker-2009/mclient-lsql-D.test.out
-rw-r--r-- 1 monet monet 12M 2009-10-01 04:38 .GNU.64.64.d.1-SunOS5.10..2009.10.01/mTests10/src/test/BugTracker-2009/mclient-lsql-D.test.out
-rw-r--r-- 1 monet monet 12M 2009-10-01 04:38 .GNU.64.32.d.1-SunOS5.10..2009.10.01/mTests10/src/test/BugTracker-2009/mclient-lsql-D.test.out
-rw-r--r-- 1 monet monet 13M 2009-10-01 04:47 .GNU.32.32.d.1-SunOS5.10..2009.10.01/mTests10/src/test/BugTracker-2009/mclient-lsql-D.test.out
-rw-r--r-- 1 monet monet 22M 2009-10-01 02:53 .GNU.64.64.d.1-SunOS5.11..2009.10.01/mTests10/src/test/BugTracker-2009/mclient-lsql-D.test.out
-rwxr-xr-x 1 monet monet 69M 2009-10-01 01:59 .Int.32.32.d.1-Windows5.1..2009.10.01/mTests10/src/test/BugTracker-2009/mclient-lsql-D.test.out*
-rw-r--r-- 1 monet monet 82M 2009-10-01 02:18 .GNU.32.32.d.1-Debian4.0..2009.10.01/mTests10/src/test/BugTracker-2009/mclient-lsql-D.test.out
-rwxr-xr-x 1 monet monet 145M 2009-10-01 02:40 .Int.64.32.d.1-Windows5.2..2009.10.01/mTests10/src/test/BugTracker-2009/mclient-lsql-D.test.out*
-rw-r--r-- 1 monet monet 151M 2009-10-01 02:45 .GNU.32.32.d.1-Fedora10..2009.10.01/mTests10/src/test/BugTracker-2009/mclient-lsql-D.test.out
-rw-r--r-- 1 monet monet 164M 2009-10-01 02:08 .GNU.64.64.d.1-Gentoo2.0.1..2009.10.01/mTests10/src/test/BugTracker-2009/mclient-lsql-D.test.out
-rw-r--r-- 1 monet monet 267M 2009-10-01 02:12 .GNU.32.32.d.1-Darwin9.8.0..2009.10.01/mTests10/src/test/BugTracker-2009/mclient-lsql-D.test.out
-rw-r--r-- 1 monet monet 308M 2009-10-01 05:14 .GNU.64.64.d.1-Darwin9.8.0..2009.10.01/mTests10/src/test/BugTracker-2009/mclient-lsql-D.test.out
Instead of simply looking like
'
START TRANSACTION;
SET SCHEMA "sys";
CREATE TABLE "sys"."table_a" (
"table_a_id" int NOT NULL,
"value_a" int NOT NULL,
CONSTRAINT "pk_table_a_id" PRIMARY KEY ("table_a_id")
);
CREATE TABLE "sys"."table_b" (
"table_b_id" int NOT NULL,
"table_a_id" int NOT NULL,
CONSTRAINT "pk_table_b_id" PRIMARY KEY ("table_b_id")
);
CREATE TABLE "sys"."table_d" (
"table_d_id" int NOT NULL,
"table_a_id" int NOT NULL,
CONSTRAINT "pk_table_d_id" PRIMARY KEY ("table_d_id")
);
CREATE TABLE "sys"."table_c" (
"table_c_id" int NOT NULL,
"table_a_id" int NOT NULL,
"table_b_id" int NOT NULL,
"table_d_id" int NOT NULL,
"value_c" int NOT NULL,
CONSTRAINT "pk_table_c_id" PRIMARY KEY ("table_c_id")
);
SET SCHEMA "sys";
ALTER TABLE "sys"."table_b" ADD CONSTRAINT "fk_table_a_id" FOREIGN KEY ("table_a_id") REFERENCES "sys"."table_a" ("table_a_id");
ALTER TABLE "sys"."table_c" ADD CONSTRAINT "fk_table_c_table_a_id" FOREIGN KEY ("table_a_id") REFERENCES "sys"."table_a" ("table_a_id");
ALTER TABLE "sys"."table_c" ADD CONSTRAINT "fk_table_c_table_d_id" FOREIGN KEY ("table_d_id") REFERENCES "sys"."table_b" ("table_b_id");
ALTER TABLE "sys"."table_d" ADD CONSTRAINT "fk_table_d_table_a_id" FOREIGN KEY ("table_a_id") REFERENCES "sys"."table_a" ("table_a_id");
COMMIT;
'
the output starts with (e.g.)
'
START TRANSACTION;
SET SCHEMA "sys";
CREATE TABLE "sys"."n8" (
"a" decimal(14,2)
);
COPY 100000000 RECORDS INTO "sys"."n8" FROM stdin USING DELIMITERS '\t','\n','"';
1.21
1.21
1.21
[...]
'
## Comment 12480
Date: 2009-10-01 21:46:41 +0200
From: @sjoerdmullender
This test depends on what has gone before, and in particular, whether other tests have cleaned up their mess.
Unfortunately, other tests don't always clean up their mess, especially not if they fail.
I think that is the problem why this test gives incorrect results.
## Comment 12481
Date: 2009-10-01 22:35:30 +0200
From: @drstmane
Thanks for the analysis and explanation, Sjoerd!
I moved this test and its "companion" 'alter_table_drop_constraint.SF-2803477' to the from of the list to prevent them from being corrupted by remains of other tests in case those do not work / clean-up properly.
Tonight's testing will show whether this helps ...
## Comment 12482
Date: 2009-10-05 22:48:27 +0200
From: @njnes
test now run?
## Comment 12483
Date: 2009-10-06 00:37:37 +0200
From: @drstmane
thanks for the reminder.
the tests run fine, now.
closing.
## Comment 12484
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2871426 at http://sourceforge.net/support/tracker.php?aid=2871426
| SQL: mclient-lsql-D fails everywhere but on 64-bit Fedora | https://api.github.com/repos/MonetDB/MonetDB/issues/2206/comments | 0 | 2020-11-30T10:27:45Z | 2024-06-27T11:41:56Z | https://github.com/MonetDB/MonetDB/issues/2206 | 753,353,828 | 2,206 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-29 04:37:10 +0200
From: @skinkie
To: @njnes
Version: -- development
Last updated: 2009-10-05 10:41:56 +0200
## Comment 12476
Date: 2009-09-29 16:37:10 +0200
From: @skinkie
We had a super friendly user again that managed to create 11GB bats on disk, therefore crashing the kernel and vmx drivers. So I decided to upgrade MonetDB to Head again. Surprisingly this time no mega bats, but just a very friendly error message :)
sql>select * from kvk,concernrelaties where upper(bedrijfsnaam) like 'VANAD%' and (kvk.kvk=concernrelaties.kvk or kvk.kvk=ouderkvk) limit 10;
TypeException:user.s1_1[60]:'algebra.leftjoin' undefined in: _84:any := algebra.leftjoin(_83:bat[:oid,:void], _41:bat[:oid,:lng], _77:lng)
TypeException:user.s1_1[65]:'algebra.slice' undefined in: _92:any := algebra.slice(_84:any, _85:int, _90:int)
TypeException:user.s1_1[67]:'algebra.leftjoin' undefined in: _94:any := algebra.leftjoin(_83:bat[:oid,:void], _93:bat[:oid,:str], _77:lng)
TypeException:user.s1_1[68]:'algebra.semijoin' undefined in: _95:any := algebra.semijoin(_94:any, _92:any)
TypeException:user.s1_1[79]:'algebra.leftjoin' undefined in: _111:any := algebra.leftjoin(_83:bat[:oid,:void], _109:bat[:oid,:str], _110:lng)
TypeException:user.s1_1[80]:'algebra.semijoin' undefined in: _112:any := algebra.semijoin(_111:any, _92:any)
TypeException:user.s1_1[91]:'algebra.leftjoin' undefined in: _124:any := algebra.leftjoin(_83:bat[:oid,:void], _123:bat[:oid,:str], _110:lng)
TypeException:user.s1_1[92]:'algebra.semijoin' undefined in: _125:any := algebra.semijoin(_124:any, _92:any)
TypeException:user.s1_1[103]:'algebra.leftjoin' undefined in: _141:any := algebra.leftjoin(_83:bat[:oid,:void], _140:bat[:oid,:str], _110:lng)
TypeException:user.s1_1[104]:'algebra.semijoin' undefined in: _142:any := algebra.semijoin(_141:any, _92:any)
TypeException:user.s1_1[115]:'algebra.leftjoin' undefined in: _156:any := algebra.leftjoin(_83:bat[:oid,:void], _154:bat[:oid,:str], _155:lng)
TypeException:user.s1_1[116]:'algebra.semijoin' undefined in: _157:any := algebra.semijoin(_156:any, _92:any)
TypeException:user.s1_1[127]:'algebra.leftjoin' undefined in: _172:any := algebra.leftjoin(_83:bat[:oid,:void], _171:bat[:oid,:int], _155:lng)
TypeException:user.s1_1[128]:'algebra.semijoin' undefined in: _173:any := algebra.semijoin(_172:any, _92:any)
TypeException:user.s1_1[139]:'algebra.leftjoin' undefined in: _186:any := algebra.leftjoin(_83:bat[:oid,:void], _185:bat[:oid,:int], _155:lng)
TypeException:user.s1_1[140]:'algebra.semijoin' undefined in: _187:any := algebra.semijoin(_186:any, _92:any)
TypeException:user.s1_1[145]:'algebra.semijoin' undefined in: _194:any := algebra.semijoin(_192:bat[:oid,:lng], _92:any)
TypeException:user.s1_1[156]:'algebra.semijoin' undefined in: _210:any := algebra.semijoin(_209:bat[:oid,:str], _92:any)
TypeException:user.s1_1[167]:'algebra.semijoin' undefined in: _223:any := algebra.semijoin(_222:bat[:oid,:str], _92:any)
TypeException:user.s1_1[178]:'algebra.semijoin' undefined in: _238:any := algebra.semijoin(_236:bat[:oid,:str], _92:any)
TypeException:user.s1_1[180]:'algebra.semijoin' undefined in: _240:any := algebra.semijoin(_239:bat[:oid,:int], _92:any)
SQLException:SQLengine:Program contains errors
plan select * from kvk,concernrelaties where upper(bedrijfsnaam) like 'VANAD%' and (kvk.kvk=concernrelaties.kvk or kvk.kvk=ouderkvk) limit 10;
top N (
| project (
| | join (
| | | table(sys.concernrelaties) [ concernrelaties.kvk, concernrelaties.ouder, concernrelaties.postcode, concernrelaties.plaats, concernrelaties.ouderkvk, concernrelaties.%TID% NOT NULL ],
| | | select (
| | | | table(sys.kvk) [ kvk.kvk, kvk.bedrijfsnaam, kvk.adres, kvk.postcode, kvk.plaats, kvk.type, kvk.kvks, kvk.sub, kvk.%TID% NOT NULL ]
| | | ) [ convert(upper(kvk.bedrijfsnaam)) like convert('VANAD%') ]
| | ) [ (kvk.kvk = concernrelaties.kvk) or (kvk.kvk = convert(concernrelaties.ouderkvk)) ]
| ) [ kvk.kvk, kvk.bedrijfsnaam, kvk.adres, kvk.postcode, kvk.plaats, kvk.type, kvk.kvks, kvk.sub, concernrelaties.kvk, concernrelaties.ouder, concernrelaties.postcode, concernrelaties.plaats, concernrelaties.ouderkvk ]
) [ 10 ]
CREATE TABLE "sys"."kvk" (
"kvk" bigint,
"bedrijfsnaam" varchar(255),
"adres" varchar(64),
"postcode" varchar(6),
"plaats" varchar(32),
"type" varchar(16),
"kvks" int,
"sub" int
);
CREATE TABLE "sys"."concernrelaties" (
"kvk" bigint,
"ouder" varchar(30),
"postcode" varchar(6),
"plaats" varchar(32),
"ouderkvk" int
);
## Comment 12477
Date: 2009-10-05 22:41:50 +0200
From: @njnes
added test to BugTracker-2009 semijoin_bug.SF-2869818.sql
fixed by only rewriting IN (values) into semijoins not (x.c0==y.c1 or x.c1==y.c2)
## Comment 12478
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2869818 at http://sourceforge.net/support/tracker.php?aid=2869818
| SQL: Program contains errors | https://api.github.com/repos/MonetDB/MonetDB/issues/2205/comments | 0 | 2020-11-30T10:27:41Z | 2024-06-27T11:41:55Z | https://github.com/MonetDB/MonetDB/issues/2205 | 753,353,784 | 2,205 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-28 09:08:18 +0200
From: Wouter Alink <<alink>>
To: Wouter Alink <<alink>>
Version: -- development
Last updated: 2009-10-08 05:59:46 +0200
## Comment 12469
Date: 2009-09-28 21:08:18 +0200
From: Wouter Alink <<alink>>
Using Aug2009 branch (compiled about a month ago, but seems to persist on recently compiled Aug2009 code):
The bug was actually found by Eugen. The following query:
(for $query in fn:doc("x")/kaas
let $coll := collection("aap")
let $qid := tijah:query-id($coll, $query)
for $res in tijah:nodes($qid)
return <beer><aap>{$res/text()}</aap></beer>)/aap
fails to compile. The error is:
$ pf test.xq
looking for col: item19, ty: 0x1 in function do_reduce (line 9708)
environment looks like:
col: iter18, type: 0x1, mvar->name: a0016
col: item19, type: 0x200000, mvar->name: a0018
col: item19, type: 0x100000, mvar->name: a0017
fatal error: column 'item19' with type '0x1' not found in environment
halted in /home/wouter/workspace_monetdb/MonetDB/pathfinder/compiler/mil/milgen.brg (env_mvar_), line 695
It seems to be related to PF/TIJAH, but I'm not completely sure.
The query does not fail when rewriting it to an attribute:
(for $query in fn:doc("x")/kaas
let $coll := collection("aap")
let $qid := tijah:query-id($coll, $query)
for $res in tijah:nodes($qid)
return <beer><aap noot="{$res/text()}"/></beer>)/aap/@noot
## Comment 12470
Date: 2009-09-30 09:19:44 +0200
From: Jan Rittinger <<jan.rittinger>>
Hi Wouter, I think I fixed this bug on Monday. Can you please check with the most-recent stable branch.
## Comment 12471
Date: 2009-09-30 09:33:25 +0200
From: Wouter Alink <<alink>>
Yep, seems to work. Thanks! Shall I add a test, or is it covered by other tests?
## Comment 12472
Date: 2009-09-30 11:10:44 +0200
From: @drstmane
I added tests for
ID: 2860574 "PF: fatal error: column not found / BATappend access denied"
https://sourceforge.net/tracker/?func=detail&aid=2860574&group_id=56967&atid=482468
in
pathfinder/tests/BugTracker/Tests/compilation_error.SF-2860574*
Please check whether these cover your problem.
If so, mention it here and close;
otherwise, add an extra test.
## Comment 12473
Date: 2009-10-08 16:31:28 +0200
From: @sjoerdmullender
Wouter, what is the status?
## Comment 12474
Date: 2009-10-08 17:59:45 +0200
From: Wouter Alink <<alink>>
I can't really judge whether the given test covers the bug. According to JanR's comment it is related. If the bug was not PF/TIJAH related, then I guess the example provided should not become a testcase.
Closing this bug.
## Comment 12475
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2869117 at http://sourceforge.net/support/tracker.php?aid=2869117
| PF or PF/TIJAH: querying constructed elements | https://api.github.com/repos/MonetDB/MonetDB/issues/2204/comments | 0 | 2020-11-30T10:27:37Z | 2024-06-27T11:41:54Z | https://github.com/MonetDB/MonetDB/issues/2204 | 753,353,738 | 2,204 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-25 02:56:46 +0200
From: @yzchang
To: MonetDB4 devs <<bugs-monetdb4>>
Version: -- development
Last updated: 2011-04-28 16:38:53 +0200
## Comment 12466
Date: 2009-09-25 14:56:46 +0200
From: @yzchang
(stable branch, source code 'cvs up'-ed this morning. compiled with configure option --enable-debug, --enable-assert, --enable-strict, --disable-optimise, on CWI desktop)
First load 'module(mapi);' and then load 'module(pathfinder);' seems to cause assertion failure in Mserver:
-----------------------
Assertion failure 1:
-----------------------
In terminal 1:
$ Mserver --dbinit="module(mapi); mapi_register(mil_frontend()); module(pathfinder); "
MonetDB Server v4.32.2
based on GDK v1.32.1
Copyright (c) 1993-July 2008, CWI. All rights reserved.
Copyright (c) August 2008-2009, MonetDB B.V.. All rights reserved.
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs; dynamically linked.
Visit http://monetdb.cwi.nl/ for further information.
PF/Tijah module v0.13.0 loaded. http://dbappl.cs.utwente.nl/pftijah
MonetDB/XQuery module v0.28.3 loaded (default back-end is 'algebra')
XRPC administrative console at http://127.0.0.1:50001/admin
MonetDB>Mserver: /ufs/zhang/monet/stable/MonetDB4/src/monet/monet_tbl.mx:1747: TBL_M_decref: Assertion `p != ((BUN) 9223372036854775807LL)' failed.
Aborted
This was caused by '\q' in terminal 2:
$ mclient -lmil
mil>"blah".print();
+-----------------------------------------------------------------------------+
| blah |
+-----------------------------------------------------------------------------+
1 tuple
mil>\q
Using 'exit();' or 'quit();', instead of '\q' didn't cause the assertion failure.
-----------------------
Assertion failure 2:
-----------------------
In terminal 1:
$ Mserver --dbinit="module(mapi); mapi_register(mil_frontend()); module(pathfinder); "
MonetDB Server v4.32.2
based on GDK v1.32.1
Copyright (c) 1993-July 2008, CWI. All rights reserved.
Copyright (c) August 2008-2009, MonetDB B.V.. All rights reserved.
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs; dynamically linked.
Visit http://monetdb.cwi.nl/ for further information.
PF/Tijah module v0.13.0 loaded. http://dbappl.cs.utwente.nl/pftijah
MonetDB/XQuery module v0.28.3 loaded (default back-end is 'algebra')
XRPC administrative console at http://127.0.0.1:50001/admin
MonetDB>Mserver: /ufs/zhang/monet/stable/MonetDB4/src/monet/monet_tbl.mx:1747: TBL_M_decref: Assertion `p != ((BUN) 9223372036854775807LL)' failed.
Aborted
This was caused by the command in terminal 2:
$ echo '"blah".print();' | mclient -lmil
[ "blah" ]
Correct output has been given before the assertion.
The assertion was not triggered by any particular MIL statement, as it happens by (seemly) any MIL statement.
The assertion won't happen if Mserver was started without 'module(pathfinder);':
Mserver --dbinit="module(mapi); mapi_register(mil_frontend()); "
or, if 'module(pathfinder);' is loaded before 'module(mapi);':
Mserver --dbinit="module(pathfinder); module(mapi); mapi_register(mil_frontend()); "
## Comment 12467
Date: 2009-09-25 15:39:12 +0200
From: @drstmane
While the assertion indeed should not happen (I can reproduce it with a simple `Ctrl-D` on the mclient console when module mapi is loaded before module pathfinder), the work around is (obviously) to not explicitly load module mapi before module pathfinder, since loading module pathfinder already triggers the load of module mapi, i.e., the following works fine for me:
Mserver --dbinit='module(pathfinder); mil_start();'
## Comment 12468
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2866544 at http://sourceforge.net/support/tracker.php?aid=2866544
The original assignee of this bug does not have
an account here. Reassigning to the default assignee
for the component, bugs-monetdb4@monetdb.org.
Previous assignee was nobody@users.sourceforge.net.
## Comment 15758
Date: 2011-04-28 16:38:53 +0200
From: @sjoerdmullender
Development on MonetDB4 and MonetDB/XQuery has ceased. It is unlikely that this bug will be fixed.
The last release of MonetDB/XQuery and MonetDB4 is Mar2011.
| Mserver assertion failed at monet_tbl.mx:1747: TBL_M_decref | https://api.github.com/repos/MonetDB/MonetDB/issues/2203/comments | 0 | 2020-11-30T10:27:34Z | 2024-06-28T13:22:46Z | https://github.com/MonetDB/MonetDB/issues/2203 | 753,353,697 | 2,203 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-24 11:44:45 +0200
From: @swingbit
To: @njnes
Version: -- development
Last updated: 2009-10-06 11:59:10 +0200
## Comment 12460
Date: 2009-09-24 11:44:45 +0200
From: @swingbit
The following query fails on Aug2009 updated on 23/09/2009, with an assertion on incref.
It did work before the update, I can't be sure but I think the previous update was about 2-3 weeks ago.
I can provide the data locally if necessary, it is quite large.
CREATE TABLE docdict_docdict AS
SELECT c.docid as c_docid, s.docid as s_docid
FROM docdict() as c, docDict as s
WHERE c.doc = s.doc
WITH DATA;
where docdict() is a mal function:
create function docdict()
returns table (docID int, doc varchar(1000))
external name sql.docdict;
function docdict{inline}():bat[:str,:bat];
b := bat.new(:str,:bat,2);
col1:bat[:void,:int] := bbp.bind("LHM_docDict_docID");
col2:bat[:void,:str] := bbp.bind("LHM_docDict_doc");
bat.insert(b, "docid", col1);
bat.insert(b, "doc", col2);
return docdict := b;
end docdict;
And the gdb backtrace:
(gdb) run --config=/ufs/roberto/lhm_code/HyServer/console/monetdb5.conf --dbname=ALPA-406_100000 --dbinit="include sql;" --set monet_daemon=yes --set mapi_open=false --set mapi_port=50152 --set monet_vault_key=/export/scratch0/roberto/MonetDB/dbfarm.stable/ALPA-406_100000/.vaultkey --set gdk_nr_threads=4Starting program: /ufs/roberto/lhm_code/HyServer/MonetDB/MonetDB.stable/MonetDB/bin/mserver5 --config=/ufs/roberto/lhm_code/HyServer/console/monetdb5.conf --dbname=ALPA-406_100000 --dbinit="include sql;" --set monet_daemon=yes --set mapi_open=false --set mapi_port=50152 --set monet_vault_key=/export/scratch0/roberto/MonetDB/dbfarm.stable/ALPA-406_100000/.vaultkey --set gdk_nr_threads=4
[Thread debugging using libthread_db enabled]
[New Thread 0x7ffff67cd7b0 (LWP 5881)]
[New Thread 0x7fffec2e7950 (LWP 5890)]
MonetDB server v5.14.1, based on kernel v1.32.1
Serving database 'ALPA-406_100000', using 4 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs dynamically linked
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2009 MonetDB B.V., all rights reserved
Visit http://monetdb.cwi.nl/ for further information
[New Thread 0x7fffe3450950 (LWP 5891)]
Listening for connection requests on mapi:monetdb://127.0.0.1:50152/
MonetDB/SQL module v2.32.1 loaded
[New Thread 0x7fffe2ac2950 (LWP 5892)]
[New Thread 0x7fffe28c1950 (LWP 5980)]
[New Thread 0x7fffe26c0950 (LWP 5982)]
[New Thread 0x7fffe24bf950 (LWP 5983)]
[New Thread 0x7fffe22be950 (LWP 5984)]
[New Thread 0x7fffe20bd950 (LWP 5985)]
mserver5: /ufs/roberto/lhm_code/HyServer/MonetDB/MonetDB.stable/src/MonetDB/src/gdk/gdk_bbp.mx:1758: incref: Assertion `BBP[(((int)(i))>0?((int)(i)):-((int)(i)))].refs+BBP[(((int)(i))>0?((int)(i)):-((int)(i)))].lrefs || BBP[(((int)(i))>0?((int)(i)):-((int)(i)))].status&16' failed.
Program received signal SIGABRT, Aborted.
[Switching to Thread 0x7fffe26c0950 (LWP 5982)]
0x0000003174432f05 in raise () from /lib64/libc.so.6
Missing separate debuginfos, use: debuginfo-install bzip2-libs-1.0.5-3.fc10.x86_64 cyrus-sasl-lib-2.1.22-19.fc10.x86_64 e2fsprogs-libs-1.41.4-6.fc10.x86_64 glibc-2.9-3.x86_64 keyutils-libs-1.2-3.fc9.x86_64 krb5-libs-1.6.3-18.fc10.x86_64 libcurl-7.19.6-1.fc10.x86_64 libidn-0.6.14-8.x86_64 libselinux-2.0.78-1.fc10.x86_64 libssh2-0.18-7.fc9.x86_64 libxml2-2.7.3-2.fc10.x86_64 libxslt-1.1.24-2.fc10.x86_64 ncurses-libs-5.6-20.20080927.fc10.x86_64 nspr-4.7.4-1.fc10.x86_64 nss-3.12.3.99.3-2.10.4.fc10.x86_64 openldap-2.4.12-1.fc10.x86_64 openssl-0.9.8g-14.fc10.x86_64 pcre-7.8-1.fc10.x86_64 raptor-1.4.18-1.fc10.x86_64 readline-5.2-13.fc9.x86_64 zlib-1.2.3-18.fc9.x86_64
(gdb) bt
0 0x0000003174432f05 in raise () from /lib64/libc.so.6
1 0x0000003174434a73 in abort () from /lib64/libc.so.6
2 0x000000317442bef9 in __assert_fail () from /lib64/libc.so.6
3 0x00007ffff7392dc4 in incref (i=2878, logical=1, lock=0) at /ufs/roberto/lhm_code/HyServer/MonetDB/MonetDB.stable/src/MonetDB/src/gdk/gdk_bbp.mx:1758
4 0x00007ffff7393965 in decref (i=2878, logical=1, releaseShare=0)
at /ufs/roberto/lhm_code/HyServer/MonetDB/MonetDB.stable/src/MonetDB/src/gdk/gdk_bbp.mx:1836
5 0x00007ffff7393166 in BBPdecref (i=2878, logical=1) at /ufs/roberto/lhm_code/HyServer/MonetDB/MonetDB.stable/src/MonetDB/src/gdk/gdk_bbp.mx:1872
6 0x00007ffff7d06aa9 in DFLOWstep (t=0x12ef608, fs=0x7fffe28bdcf0)
at /ufs/roberto/lhm_code/HyServer/MonetDB/MonetDB.stable/src/MonetDB5/src/mal/mal_interpreter.mx:1590
7 0x00007ffff7d0e502 in runDFLOWworker (t=0x12ef608) at /ufs/roberto/lhm_code/HyServer/MonetDB/MonetDB.stable/src/MonetDB5/src/mal/mal_interpreter.mx:845
8 0x00000031750073da in start_thread () from /lib64/libpthread.so.0
9 0x00000031744e62bd in clone () from /lib64/libc.so.6
(gdb)
## Comment 12461
Date: 2009-09-24 11:49:12 +0200
From: @swingbit
I forgot to describe the second table in input for this query:
sql>\d docdict
CREATE TABLE "lhm"."docdict" (
"docid" int NOT NULL,
"doc" varchar(1000),
"type" varchar(255),
"prob" double DEFAULT 1.000000,
CONSTRAINT "docdict_docid_pkey" PRIMARY KEY ("docid")
);
## Comment 12462
Date: 2009-09-24 13:45:50 +0200
From: @swingbit
Update: the problem seems related to using the function docdict(), which binds two bats, in the join.
By first materialising this virtual table in a standard sql table, the problem disappears:
CREATE TABLE c_docdict AS
SELECT *
FROM docdict()
WITH DATA;
CREATE TABLE docdict_docdict AS
SELECT c.docid as c_docid, s.docid as s_docid
FROM c_docdict as c, docDict as s
WHERE c.doc = s.doc
WITH DATA;
## Comment 12463
Date: 2009-09-28 21:00:00 +0200
From: @njnes
I cannot reproduce the problem.Maybe related to the missing data. ie could we get the data of these tables?
## Comment 12464
Date: 2009-10-06 11:59:00 +0200
From: @swingbit
After an update (shortly after my bug report) I cannot reproduce the bug anymore.
Also, another similar problem I had and for which I had found a workaround,
does not seem to happen anymore.
Maybe this below was the fix?
I close this bug report as fixed.
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> 2009/09/22 - nielsnes: src/storage/bat/bat_storage.mx,1.42.2.4
>
> fix for bug ID 2864313
> SQL: Update changes all columns with same type and default
>
> A single bat was used for multiple columns.
> We now make sure we copy the bat when its already used.
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
## Comment 12465
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2865631 at http://sourceforge.net/support/tracker.php?aid=2865631
| SQL: incref assertion | https://api.github.com/repos/MonetDB/MonetDB/issues/2202/comments | 0 | 2020-11-30T10:27:29Z | 2024-06-27T11:41:52Z | https://github.com/MonetDB/MonetDB/issues/2202 | 753,353,637 | 2,202 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-23 02:57:07 +0200
From: Axel Belinfante <<axelbel>>
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2011-09-07 11:21:15 +0200
## Comment 12455
Date: 2009-09-23 14:57:07 +0200
From: Axel Belinfante <<axelbel>>
MonetDB craashes after the following sequence of commands:
echo 'import module namespace ecv = "http://www.cs.utwente.nl/~belinfan/ecv" at "/local/dbguest/ECV/xrpc/export/ecv.xq"; ecv:get-active-views("view-defs.xml")' | mclient -lx -p51012
echo 'import module namespace ecv = "http://www.cs.utwente.nl/~belinfan/ecv" at "/local/dbguest/ECV/xrpc/export/ecv.xq"; ecv:save-view("view-defs.xml", "feenstra", "axel", <view now="1253707155.50922" name="feenstra" prevname="untitled" prevseq="0" user="axel"><select allany="" inex="exclude"><cond cmp="contains" field="creators_family">feenstra</cond></select></view>)' | mclient -lx -p 51012
echo 'import module namespace ecv = "http://www.cs.utwente.nl/~belinfan/ecv" at "/local/dbguest/ECV/xrpc/export/ecv.xq"; ecv:get-active-views("view-defs.xml")' | mclient -lx -p51012
where the database is started as
$MONET_PREFIX/bin/Mserver --config=$demodir/MonetDB.conf --set monet_daemon=yes \
--set mapi_port=51012 --set xrpc_port=51013 \
--dbinit="module(mapi);module(pathfinder);module(pftijah); xrpc_trusted.delete(); mil_start();" \
--dbname $FARM start-mxq.mil
and start-mxq.mil contains:
shred_doc("/local/dbguest/ECV/preload/updatable/view-defs.xml","view-defs.xml","ECVsave",lng(10));
I attach view-defs.xml and ecv.xq
this is on linux: Linux ewi865 2.6.16.21-0.8-smp 1 SMP Mon Jul 3 18:25:39 UTC 2006 x86_64 x86_64 x86_64 GNU/Linux
a gdb stack trace follows below.
(gdb) run --config=/local/dbguest/ECV/MonetDB.conf --set monet_daemon=yes --set mapi_port=51012 --set xrpc_port=51013 --dbinit="module(mapi);module(pathfinder);module(pftijah); xrpc_trusted.delete(); mil_start();" --dbname demo start-mxq.mil
Starting program: /local/dbguest/Install/MonetDB-XQuery_Aug2009/bin/Mserver --config=/local/dbguest/ECV/MonetDB.conf --set monet_daemon=yes --set mapi_port=51012 --set xrpc_port=51013 --dbinit="module(mapi);module(pathfinder);module(pftijah); xrpc_trusted.delete(); mil_start();" --dbname demo start-mxq.mil
[Thread debugging using libthread_db enabled]
[New Thread 47689001507792 (LWP 8600)]
[New Thread 1075841344 (LWP 8607)]
MonetDB Server v4.32.1
based on GDK v1.32.1
Copyright (c) 1993-July 2008, CWI. All rights reserved.
Copyright (c) August 2008-2009, MonetDB B.V.. All rights reserved.
Compiled for x86_64-suse-linux/64bit with 32bit OIDs; dynamically linked.
Visit http://monetdb.cwi.nl/ for further information.
[New Thread 1077942592 (LWP 8608)]
PF/Tijah module v0.13.0 loaded. http://dbappl.cs.utwente.nl/pftijah
MonetDB/XQuery module v0.28.3 loaded (default back-end is 'algebra')
XRPC administrative console at http://127.0.0.1:51013/admin
[New Thread 1080043840 (LWP 8609)]
[New Thread 1082145088 (LWP 8610)]
DB group ECV demo server (Aug2009) started.
!ERROR: shred_doc: document eprints-dummy.xml already exists in collection ECV (1 such errors)!
[New Thread 1084246336 (LWP 8612)]
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 1084246336 (LWP 8612)]
0x00002b5f744fddae in interpret_assignment (stk=2287, lt=0x17e8cf8, res=0x409fecf0) at monet_interpreter.mx:1842
1842 ATOMunfix(oldvalue.vtype, VALptr(&oldvalue));
(gdb) bt
0 0x00002b5f744fddae in interpret_assignment (stk=2287, lt=0x17e8cf8, res=0x409fecf0) at monet_interpreter.mx:1842
1 0x00002b5f744f0fac in interpret (stk=2287, lt=0x17e8cf8, res=0x409fecf0) at monet_interpreter.mx:725
2 0x00002b5f744fdf4e in interpret_seqblock (stk=2287, lt=0x17f1cc8, res=0x409fecf0, scope=0) at monet_interpreter.mx:1877
3 0x00002b5f744f103d in interpret (stk=2287, lt=0x17f1cc8, res=0x409fecf0) at monet_interpreter.mx:732
4 0x00002b5f744fdf4e in interpret_seqblock (stk=2287, lt=0x17f1c90, res=0x409fecf0, scope=1) at monet_interpreter.mx:1877
5 0x00002b5f744f103d in interpret (stk=1050, lt=0x17f1c90, res=0x409fecf0) at monet_interpreter.mx:732
6 0x00002b5f744fdf4e in interpret_seqblock (stk=1050, lt=0x1802dd8, res=0x409fecf0, scope=0) at monet_interpreter.mx:1877
7 0x00002b5f744f103d in interpret (stk=1050, lt=0x1802dd8, res=0x409fecf0) at monet_interpreter.mx:732
8 0x00002aaaadb886a5 in xquery_tree_exec (ctx=0x10357d8, t=0x1802dd8, repeat=1) at pathfinder.mx:4525
9 0x00002aaaadb8bbba in xquery_function_call (ctx=0x10357d8, usec=28412951, ns=0x13a93bc "ecv", module=0x40a037d0 "http://www.cs.utwente.nl/~belinfan/ecv",
method=0x13a93c0 "get-active-views", seqnr=0x0, qid=0x2aaaadd6504b "", caller=0x2aaaadd6504b "", timeout=0, mode=0x2aaaadd6504b "", argc=1, itercnt=1,
argcnt=0x40a033c8, argtpe=0x40a03330, argval=0x40a03130, shredBAT=0x0) at pathfinder.mx:5147
10 0x00002aaaadb8f540 in xquery_prepare (ctx=0x10357d8, usec=28412951,
query=0x13a9349 "import module namespace ecv = \"http://www.cs.utwente.nl/~belinfan/ecv\" at \"/local/dbguest/ECV/xrpc/export/ecv.xq\"; ecv")
at pathfinder.mx:6032
11 0x00002aaaadb8fc89 in xquery_client_engine (mc=0x2aaaac501f88) at pathfinder.mx:6220
12 0x00002aaaac3fe291 in mapi_client_engine (FC=0x2aaaac501f88) at mapi.mx:497
13 0x00002b5f75770193 in start_thread () from /lib64/libpthread.so.0
14 0x00002b5f75e7745d in clone () from /lib64/libc.so.6
15 0x0000000000000000 in ?? ()
## Comment 12456
Date: 2009-09-24 15:33:40 +0200
From: @drstmane
I can reproduce the problem as follows:
in shell 1:
$ gdb --args Mserver --dbinit="module(pathfinder); mil_start();" --dbname ECV /tmp/start-mxq.mil
GNU gdb Fedora (6.8-32.fc10)
Copyright (C) 2008 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-redhat-linux-gnu"...
(gdb) r
Starting program: /net/rig.ins.cwi.nl/export/scratch0/manegold/Monet/Testing/Stable/prefix.--enable-strict_--disable-optimize_--enable-debug_--enable-assert/bin/Mserver --dbinit=module\(pathfinder\)\;\ mil_start\(\)\; --dbname ECV /tmp/start-mxq.mil
[Thread debugging using libthread_db enabled]
[New Thread 0x7ffff56df7b0 (LWP 22629)]
[New Thread 0x7fffeb1f9950 (LWP 22682)]
MonetDB Server v4.32.1
based on GDK v1.32.1
Copyright (c) 1993-July 2008, CWI. All rights reserved.
Copyright (c) August 2008-2009, MonetDB B.V.. All rights reserved.
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs; dynamically linked.
Visit http://monetdb.cwi.nl/ for further information.
PF/Tijah module v0.13.0 loaded. http://dbappl.cs.utwente.nl/pftijah
[New Thread 0x7fffe6377950 (LWP 22686)]
MonetDB/XQuery module v0.28.3 loaded (default back-end is 'algebra')
XRPC administrative console at http://127.0.0.1:50001/admin
[New Thread 0x7fffe6176950 (LWP 22687)]
[New Thread 0x7fffe5f75950 (LWP 22688)]
DB group ECV demo server (Aug2009) started.
shredding with holes: pagesize=16384, fillsize=14746
Elapsed time = 431us [043us/node]
Shredded 1 XML document (view-defs.xml), total time after commit=0.011s
MonetDB>[New Thread 0x7fffe5d74950 (LWP 22706)]
[New Thread 0x7fffe5b73950 (LWP 22770)]
[Thread 0x7fffe5b73950 (LWP 22770) exited]
[New Thread 0x7fffe5972950 (LWP 22776)]
[Thread 0x7fffe5972950 (LWP 22776) exited]
in shell 2:
$ mclient -lx -s'doc("view-defs.xml")'
<?xml version="1.0" encoding="utf-8"?>
<views>
<active>
</active>
<archive>
</archive>
</views>
$ echo 'import module namespace ecv = "http://www.cs.utwente.nl/~belinfan/ecv" at "/tmp/ecv.xq"; ecv:get-active-views("view-defs.xml")' | mclient -lx
<views/>
$ echo 'import module namespace ecv = "http://www.cs.utwente.nl/~belinfan/ecv" at "/tmp/ecv.xq"; ecv:save-view("view-defs.xml", "feenstra", "axel", <view now="1253707155.50922" name="feenstra" prevname="untitled" prevseq="0" user="axel"><select allany="" inex="exclude"><cond cmp="contains" field="creators_family">feenstra</cond></select></view>)' | mclient -lx
$ mclient -lx -s'doc("view-defs.xml")'
<?xml version="1.0" encoding="utf-8"?>
<views>
<active>
<view seq="0" now="1253707155.50922" name="feenstra" prevname="untitled" prevseq="0" user="axel"><select inex="exclude" allany=""><cond field="creators_family" cmp="contains">feenstra</cond></select></view></active>
<archive>
</archive>
</views>
$ mclient -lx -s'doc("view-defs.xml")/views'
<views>
<active>
<view seq="0" now="1253707155.50922" name="feenstra" prevname="untitled" prevseq="0" user="axel"><select inex="exclude" allany=""><cond field="creators_family" cmp="contains">feenstra</cond></select></view></active>
<archive>
</archive>
</views>
$ mclient -lx -s'doc("view-defs.xml")/views/active'
<active>
<view seq="0" now="1253707155.50922" name="feenstra" prevname="untitled" prevseq="0" user="axel"><select inex="exclude" allany=""><cond field="creators_family" cmp="contains">feenstra</cond></select></view></active>
$ mclient -lx -s'doc("view-defs.xml")/views/active/view'
<view seq="0" now="1253707155.50922" name="feenstra" prevname="untitled" prevseq="0" user="axel"><select inex="exclude" allany=""><cond field="creators_family" cmp="contains">feenstra</cond></select></view>
$ mclient -lx -s'for $v in doc("view-defs.xml")/views/active/view return $v'
<view seq="0" now="1253707155.50922" name="feenstra" prevname="untitled" prevseq="0" user="axel"><select inex="exclude" allany=""><cond field="creators_family" cmp="contains">feenstra</cond></select></view>
$ mclient -lx -s'<view> { for $v in doc("view-defs.xml")/views/active/view return $v } </view>'
<view><view seq="0" now="1253707155.50922" name="feenstra" prevname="untitled" prevseq="0" user="axel"><select inex="exclude" allany=""><cond field="creators_family" cmp="contains">feenstra</cond></select></view></view>
$ echo 'import module namespace ecv = "http://www.cs.utwente.nl/~belinfan/ecv" at "/tmp/ecv.xq"; ecv:get-active-views("view-defs.xml")' | pf | mclient -lmil
<?xml version="1.0" encoding="utf-8"?>
<XQueryResult><views><view seq="0" now="1253707155.50922" name="feenstra" prevname="untitled" prevseq="0" user="axel"><select inex="exclude" allany=""><cond field="creators_family" cmp="contains">feenstra</cond></select></view></views></XQueryResult>
$ echo 'import module namespace ecv = "http://www.cs.utwente.nl/~belinfan/ecv" at "/tmp/ecv.xq"; ecv:get-active-views("view-defs.xml")' | mclient -lx
[
Note that the second call of ecv:get-active-views("view-defs.xml") (i.e., after the update query) works fine when piped through `pf | mclient -lmil`, but crashes the server as reported when piped through `mclient -lx` ...
]
in shell 1:
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7fffe5d74950 (LWP 22706)]
0x00007ffff6bf38aa in interpret_assignment (stk=1057, lt=0x19869d8, res=0x7fffe5d6ebf0) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/monet/monet_interpreter.mx:1842
1842 ATOMunfix(oldvalue.vtype, VALptr(&oldvalue));
Missing separate debuginfos, use: debuginfo-install bzip2-libs-1.0.5-3.fc10.x86_64 e2fsprogs-libs-1.41.4-6.fc10.x86_64 glibc-2.9-3.x86_64 keyutils-libs-1.2-3.fc9.x86_64 krb5-libs-1.6.3-18.fc10.x86_64 libselinux-2.0.78-1.fc10.x86_64 libxml2-2.7.3-2.fc10.x86_64 openssl-0.9.8g-14.fc10.x86_64 pcre-7.8-1.fc10.x86_64 zlib-1.2.3-18.fc9.x86_64
(gdb) bt
0 0x00007ffff6bf38aa in interpret_assignment (stk=1057, lt=0x19869d8, res=0x7fffe5d6ebf0) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/monet/monet_interpreter.mx:1842
1 0x00007ffff6be696e in interpret (stk=1057, lt=0x19869d8, res=0x7fffe5d6ebf0) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/monet/monet_interpreter.mx:725
2 0x00007ffff6bf3a53 in interpret_seqblock (stk=1057, lt=0x1986988, res=0x7fffe5d6ebf0, scope=0) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/monet/monet_interpreter.mx:1877
3 0x00007ffff6be69ff in interpret (stk=1057, lt=0x1986988, res=0x7fffe5d6ebf0) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/monet/monet_interpreter.mx:732
4 0x00007ffff6bf3a53 in interpret_seqblock (stk=1057, lt=0x1986950, res=0x7fffe5d6ebf0, scope=1) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/monet/monet_interpreter.mx:1877
5 0x00007ffff6be69ff in interpret (stk=1056, lt=0x1986950, res=0x7fffe5d6ebf0) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/monet/monet_interpreter.mx:732
6 0x00007ffff6bf3a53 in interpret_seqblock (stk=1056, lt=0x1986438, res=0x7fffe5d6ebf0, scope=0) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/monet/monet_interpreter.mx:1877
7 0x00007ffff6be69ff in interpret (stk=1056, lt=0x1986438, res=0x7fffe5d6ebf0) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/monet/monet_interpreter.mx:732
8 0x00007fffe70f8c5b in xquery_tree_exec (ctx=0x13c0e98, t=0x1986438, repeat=1) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/pathfinder/runtime/pathfinder.mx:4525
9 0x00007fffe70fbd82 in xquery_function_call (ctx=0x13c0e98, usec=111580345, ns=0x163da32 "ecv", module=0x7fffe5d73670 "http://www.cs.utwente.nl/~belinfan/ecv", method=0x163da36 "get-active-views", seqnr=0x0, qid=0x7fffe72caf2b "", caller=0x7fffe72caf2b "", timeout=0, mode=0x7fffe72caf2b "", argc=1, itercnt=1,
argcnt=0x7fffe5d73268, argtpe=0x7fffe5d731d0, argval=0x7fffe5d72fd0, shredBAT=0x0) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/pathfinder/runtime/pathfinder.mx:5147
10 0x00007fffe70ff665 in xquery_prepare (ctx=0x13c0e98, usec=111580345, query=0x163d9d9 "import module namespace ecv = \"http://www.cs.utwente.nl/~belinfan/ecv\" at \"/tmp/ecv.xq\"; ecv") at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/pathfinder/runtime/pathfinder.mx:6032
11 0x00007fffe70ffdb0 in xquery_client_engine (mc=0x7fffe707af88) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/pathfinder/runtime/pathfinder.mx:6223
12 0x00007fffe6e770a6 in mapi_client_engine (FC=0x7fffe707af88) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/mapi/mapi.mx:497
13 0x0000003675e073da in start_thread () from /lib64/libpthread.so.0
14 0x00000036752e62bd in clone () from /lib64/libc.so.6
(gdb) print oldvalue.vtype
$1 = -606348325
(gdb) print oldvalue
$2 = {val = {ival = 28879488, oval = 28879488, shval = -21888, btval = -128 '\200', wval = 28879488, cval = "\200��\001", fval = 6.78355807e-38, pval = 0x1b8aa80, Bval = 0x1b8aa80, bval = 28879488, br = {id = 28879488, stamp = 0}, sval = 0x1b8aa80 " ", dval = 1.4268362890284529e-316, lval = 28879488},
len = -606348325, vtype = -606348325}
(gdb) print value
$3 = (ValRecord *) 0x198aa38
(gdb) print *value
$4 = {val = {ival = 62, oval = 140733193388094, shval = 62, btval = 62 '>', wval = 140733193388094, cval = ">\000\000", fval = 8.68805048e-44, pval = 0x7fff0000003e, Bval = 0x7fff0000003e, bval = 62, br = {id = 62, stamp = 32767}, sval = 0x7fff0000003e <Address 0x7fff0000003e out of bounds>,
dval = 6.953143608258971e-310, lval = 140733193388094}, len = 0, vtype = 5}
(gdb)
Apparently, "oldvalue" contains invalid/corrupted data --- I have no idea, yet, why ...
## Comment 12457
Date: 2009-09-28 18:21:58 +0200
From: @drstmane
Thsi one might be related to ID: 2859986 "xrpc use messes up mclient use" (https://sourceforge.net/tracker/?func=detail&aid=2859986&group_id=56967&atid=482468), and hence to MPS + query plan caching, afterall ... needs to be analysed in more detail ...
## Comment 12458
Date: 2009-10-12 19:45:32 +0200
From: Axel Belinfante <<axelbel>>
I'm trying to understand the consequences of your observation:
| Note that the second call of ecv:get-active-views("view-defs.xml") (i.e.,
| after the update query) works fine when piped through `pf | mclient -lmil`,
| but crashes the server as reported when piped through `mclient -lx` ...
to my understanding, this might be a workaround.
The scenario that triggers the bug described here, comes from a
ruby-on-rails application that offers a gui to construct (abstract) queries.
One part of the application more or less alternates between
reading a document and updating it.
It uses xquery.
so far I have tried to use xrpc and mapi to access the MonetDB server
(mapi access via the ruby code supplied in the monetdb distribution).
When I use a ruby wrapper that first calls pf and then passes the
result to mclient -lmil it seems to work, and even more, it seems to
work around the issue I posted as ID: 2859986 "xrpc use messes up mclient use".
So far the good news.
The document that is alternately read and updated contains abstract queries.
From those abstract queries I generate xquery code, and execute those
both from the webclient (to allow the user to preview the result) as well as via cgi.
At the moment I generate from each query a module containing several functions,
with one main function that does the work (using the other ones).
To execute a query from a module, I construct a new query that imports the
module and invokes its main function.
The bad news is that some queries are more complex.
At least, pf takes forever to compile them.
The interesting thing is that, nevertheless, mclient -lx can handle them rather fast
(order of seconds or less than a second).
I don't know whether this is because of caching - but even then, I don't understand
that also the first invocation via mclient -lx is rather fast (uncomparable to the pf that
takes forever).
The other bad news is that I failed to use the ruby mapi code
to pass mil code to monetdb.
(after the same set-up that I used to pass xquery code
(except for using 'mil' instead of 'xquery' in the connection set-up))
I just use the send function of MonetDBConnection.rb to pass the mil code
(that I obtained from stdout of pf)
as a single long string (with embedded newlines).
Unfortunately, I then get an error message:
!ERROR: "Xmodule ("pathfinder");"
! ^
! new MIL syntax (>=4.4) requires parentheses for all operators.
!.
## Comment 12459
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2864991 at http://sourceforge.net/support/tracker.php?aid=2864991
## Comment 16190
Date: 2011-09-07 11:21:15 +0200
From: @sjoerdmullender
Development on MonetDB4 and MonetDB/XQuery has ceased. It is unlikely that
this bug will be fixed.
The last release of MonetDB/XQuery and MonetDB4 is Mar2011.
| monetbd crash | https://api.github.com/repos/MonetDB/MonetDB/issues/2201/comments | 0 | 2020-11-30T10:27:25Z | 2024-06-28T13:22:45Z | https://github.com/MonetDB/MonetDB/issues/2201 | 753,353,574 | 2,201 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-22 03:26:52 +0200
From: @drstmane
To: @njnes
Version: -- development
Last updated: 2009-09-23 09:28:26 +0200
## Comment 12449
Date: 2009-09-22 15:26:52 +0200
From: @drstmane
When updating a not yet filled column, all columns of the same type and with the same default get modified:
sql>CREATE TABLE "t0" (
more> "d0" double
more>);
Operation successful
sql>CREATE TABLE "t1" (
more> "d1" double,
more> "v1" varchar(255),
more> "v2" varchar(255)
more>);
Operation successful
sql>COPY 1025 RECORDS INTO t0 FROM '/tmp/err2.data' USING DELIMITERS '\t';
Rows affected 1025
sql>select count(*) from t0;
+-------+
| L1 |
+=======+
| 1025 |
+-------+
1 tuple
sql>INSERT INTO t1 (d1) SELECT d0 FROM t0;
Rows affected 1025
sql>select count(*) from t1;
+-------+
| L2 |
+=======+
| 1025 |
+-------+
1 tuple
sql>select d1, v2, v1 from t1 limit 1;
+------------------------+---------------------------------------------------------------------------------------------------------+---------------------------------------------------------------------------------------------------------+
| d1 | v2 | v1 |
+========================+=========================================================================================================+=========================================================================================================+
| 0 | null | null |
+------------------------+---------------------------------------------------------------------------------------------------------+---------------------------------------------------------------------------------------------------------+
1 tuple
sql>Update t1 Set v2 = 'AA';
Rows affected 1025
sql>select d1, v2, v1 from t1 limit 1;
+------------------------+-------+-------+
| d1 | v2 | v1 |
+========================+=======+=======+
| 0 | AA | AA |
+------------------------+-------+-------+
1 tuple
sql>Update t1 Set d1 = -1.2;
Rows affected 1025
sql>select d1, v2, v1 from t1 limit 1;
+------------------------+-------+-------+
| d1 | v2 | v1 |
+========================+=======+=======+
| -1.2 | AA | AA |
+------------------------+-------+-------+
1 tuple
sql>Update t1 Set v1 = 'BB';
Rows affected 1025
sql>select d1, v2, v1 from t1 limit 1;
+------------------------+-------+-------+
| d1 | v2 | v1 |
+========================+=======+=======+
| -1.2 | BB | BB |
+------------------------+-------+-------+
1 tuple
sql>drop table t1;
Operation successful
sql>drop table t0;
Operation successful
It works fine when bulkloading <= 1024 tuples, or when using different defaults for all non-initialized columns, e.g.,
"v1" varchar(255) default 'x',
"v2" varchar(255) default 'y'
I will add a test to CVS.
## Comment 12450
Date: 2009-09-22 16:44:58 +0200
From: @sjoerdmullender
This needs to be fixed before the Aug2009-SP1 bug fix release.
## Comment 12451
Date: 2009-09-22 22:24:22 +0200
From: @sjoerdmullender
I think the problem is in tr_update_delta in bat_storage.mx. There is a line there
if (!BATcount(cur) && BATcount(ins) > snapshot_minsize){
which makes the difference between the 1024 and 1025 cases (snapshot_minsize==1024).
The original BAT (cur) is empty in both the v1 and v2 columns, and the inserted BAT is longer than 1024, so the then branch is taken where the BAT representing the column is replaced by the inserted BAT.
But because the inserted BATs for the v1 and v2 columns are identical (look at the explain output for the INSERT where the same BAT is appended to both columns), the replaced BAT in the code in tr_udpate_delta is the same BAT for both columns.
The subsequent UPDATE then updates the bat for the one column, but since it is now shared, it affects the other columns as well.
One solution would be to remove the then case and always do the else case which will append the values to the cur BAT (which initially are different empty BATs).
Another solution wouls be to generate code to append different BATs to the two columsn, thereby avoiding the problem.
## Comment 12452
Date: 2009-09-23 09:36:30 +0200
From: @drstmane
I added tests in sql/src/test/BugTracker-2009/Tests/updating_non-loaded_columns.SF-2864313*
and Niels' yesterday's checkin seem to fix the problem.
Not closing, yet, as I first want to analyze / discuss Sjoerd comment (which I see only now) with Sjoerd and/or Niels ...
## Comment 12453
Date: 2009-09-23 21:28:26 +0200
From: @njnes
we avoid the problem described by sjoerd allready in the append_delta code. This is done using the BBPlrefs counter
## Comment 12454
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2864313 at http://sourceforge.net/support/tracker.php?aid=2864313
| SQL: Update changes all columns with same type and default | https://api.github.com/repos/MonetDB/MonetDB/issues/2200/comments | 0 | 2020-11-30T10:27:22Z | 2024-06-27T11:41:50Z | https://github.com/MonetDB/MonetDB/issues/2200 | 753,353,532 | 2,200 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-22 12:30:26 +0200
From: @arjenderijke
To: @njnes
Version: -- development
Last updated: 2009-11-11 07:43:00 +0100
## Comment 12442
Date: 2009-09-22 12:30:26 +0200
From: @arjenderijke
When the result of a function call within a function is assigned to a variable and that variable is not used in the calling function, the called function is never executed. The attached file demonstrates this behaviour. It also explaines the failing of sql test src/test/Via-m/Tests/file_s18.
## Comment 12443
Date: 2009-09-22 12:49:25 +0200
From: @drstmane
Dead code elimination at work?
Or: dead code elimination does (can?) not recognize that your function has a side effect, and hence skips the function call, since its result is not used?
## Comment 12444
Date: 2009-09-23 21:45:59 +0200
From: @njnes
fixed on current by martin
## Comment 12445
Date: 2009-09-25 10:10:18 +0200
From: @arjenderijke
According to http://monetdb.cwi.nl/testing/projects/monetdb/Current/sql/.mTests103/GNU.64.64.d.0-Fedora10/src_test_Via-m/file_s18.out.00.html of today the problem still exists. But maybe that test is more complicated than the script i attached to this bugreport. To be precise, the problem occurs in test file_s17, that contains the call to a function. And because that function does not insert records into a table, the test file_s18 failes, because the table does not contain the records that are expected.
## Comment 12446
Date: 2009-09-30 14:14:54 +0200
From: @arjenderijke
I added the script as a test on current, function_call_not_handled.SF-2864179. Last testing on current the test failed, so the bug is not fixed.
## Comment 12447
Date: 2009-11-11 19:43:00 +0100
From: @njnes
fixed by improving the check for side effect statements in a sql generated function.
## Comment 12448
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2864179 at http://sourceforge.net/support/tracker.php?aid=2864179
| function call not handled | https://api.github.com/repos/MonetDB/MonetDB/issues/2199/comments | 0 | 2020-11-30T10:27:18Z | 2024-06-27T11:41:49Z | https://github.com/MonetDB/MonetDB/issues/2199 | 753,353,493 | 2,199 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-21 10:56:23 +0200
From: Wouter Alink <<alink>>
To: @njnes
Version: -- development
Last updated: 2009-09-22 11:36:21 +0200
## Comment 12438
Date: 2009-09-21 22:56:23 +0200
From: Wouter Alink <<alink>>
In the following scenario erroneous data is sometimes returned:
- start transaction
- create a view (with a where-clause containing literals)
- select some data from the view
- abort the transaction
- start a new transaction
- create a view with the same name as the previous view (but with another literal in the where-clause)
- select from the view
The view will in some (not all) cases still contain the result of the old view definition.
Apparantly Niels already fixed this today, but to be able to add a test-script, I need a bug-number.
## Comment 12439
Date: 2009-09-21 23:13:36 +0200
From: Wouter Alink <<alink>>
added test in Aug2009 branch. awaiting a green testweb now.
## Comment 12440
Date: 2009-09-22 11:36:21 +0200
From: Wouter Alink <<alink>>
testweb is red for this test, but the test-output is ok. I modified the stable-output. closing this bug.
## Comment 12441
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2863804 at http://sourceforge.net/support/tracker.php?aid=2863804
| SQL: old views incorrectly reused | https://api.github.com/repos/MonetDB/MonetDB/issues/2198/comments | 0 | 2020-11-30T10:27:15Z | 2024-06-27T11:41:48Z | https://github.com/MonetDB/MonetDB/issues/2198 | 753,353,444 | 2,198 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-21 02:36:00 +0200
From: Wouter Alink <<alink>>
To: Jan Flokstra <<jflokstra>>
Version: -- development
Last updated: 2009-11-12 06:31:26 +0100
## Comment 12434
Date: 2009-09-21 14:36:00 +0200
From: Wouter Alink <<alink>>
Eugen detected the following: of the following 3 queries, the latter 2 seem to fail (with the algebra frontend only).
Note that the latter 2 queries produce compile-time warnings such as "warning: pin count for variable a0002 below zero". It seems to be influenced by the use of tijah:tokenize(), as I couldn't reproduce the error without using that function.
$ cat tmp.xq
(: works :)
for $n in (1,2)
let $query := tijah:tokenize(fn:string("aap beer noot"))
let $len := fn:string-length(concat("vuur",$query))
return ($len)
<>
(: fails :)
for $n in (1,2)
let $query := tijah:tokenize(fn:string("aap beer noot"))
let $len := fn:string-length($query)
return ($len)
<>
(: fails :)
for $n in (1,2)
let $query := tijah:tokenize(fn:string("aap beer noot"))
let $len := fn:string-length(concat("vuur",$query))
return ($len,$query)
$ pf --help
Pathfinder XQuery Compiler ($Revision: 1.115 $, $Date: 2009/05/07 14:21:53 $)
(c) Database Systems Group, Eberhard Karls Universitaet Tuebingen
$ mclient -g -lxq -p50333 < tmp.xq
17,
17
13,
13
17,
"aap beer noot",
17,
"aap beer noot"
$ mclient -G -lxq -p50333 < tmp.xq
17,
17
warning: pin count for variable a0002 below zero
warning: pin count for variable a0001 below zero
MAPI = monetdb@localhost:50333
QUERY = (: fails :)
ERROR = !ERROR: interpret: no matching MIL operator to 'leftjoin(void, BAT[oid,oid])'.
!MAYBE YOU MEAN:
! leftjoin(BAT[any::1,any::2], BAT[any::2,any::3]) : BAT[any::1,any::3]
! leftjoin(any::1, BAT[any::1,any::2]) : any::2
! leftjoin(oid, any::1) : any::1
! leftjoin(BAT[void,oid], any::1) : any::1
! leftjoin(BAT[oid,oid], any::1) : BAT[oid,any::1]
warning: pin count for variable a0002 below zero
warning: pin count for variable a0001 below zero
MAPI = monetdb@localhost:50333
QUERY = (: fails :)
ERROR = !ERROR: interpret: no matching MIL operator to 'append(BAT[oid,oid], BAT[oid,lng])'.
!MAYBE YOU MEAN:
! append(BAT[any::1,BAT], any) : BAT[any::1,BAT]
! append(BAT[void,any::1], any::1, bit) : BAT[void,any::1]
! append(BAT[void,any::1], any::1) : BAT[void,any::1]
! append(BAT[oid,any::1], any::1, bit) : BAT[oid,any::1]
! append(BAT[oid,any::1], any::1) : BAT[oid,any::1]
! append(BAT[void,any::1], BAT[oid,any::1], bit) : BAT[void,any::1]
! append(BAT[void,any::1], BAT[oid,any::1]) : BAT[void,any::1]
! append(BAT[void,any::1], BAT[void,any::1], bit) : BAT[void,any::1]
! append(BAT[void,any::1], BAT[void,any::1]) : BAT[void,any::1]
! append(BAT[oid,any::1], BAT[oid,any::1], bit) : BAT[oid,any::1]
! append(BAT[oid,any::1], BAT[oid,any::1]) : BAT[oid,any::1]
!ERROR: interpret_params: access(param 1): evaluation error.
## Comment 12435
Date: 2009-09-21 14:38:04 +0200
From: Wouter Alink <<alink>>
Henning told me that I should assign it to jan flokstra... (hope that's ok :)
## Comment 12436
Date: 2009-09-22 12:28:07 +0200
From: Jan Flokstra <<jflokstra>>
I fixed the bug. The pin count of the tokenize operation result bats was always set to 1 but
should have been set to the number of times the operand is used by other
operations. The exampoles work on my machine now.
I added a testset in pftijah/Tests/test_comp_tokenize.
## Comment 12437
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2863458 at http://sourceforge.net/support/tracker.php?aid=2863458
| PF/TIJAH: tokenize in a for-loop | https://api.github.com/repos/MonetDB/MonetDB/issues/2197/comments | 0 | 2020-11-30T10:27:11Z | 2024-06-27T11:41:47Z | https://github.com/MonetDB/MonetDB/issues/2197 | 753,353,401 | 2,197 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-19 02:36:48 +0200
From: Alex Bo <<alexbod>>
To: @njnes
Version: -- development
Last updated: 2009-09-19 08:54:50 +0200
## Comment 12431
Date: 2009-09-19 14:36:48 +0200
From: Alex Bo <<alexbod>>
This query doesn't work:
SELECT COUNT (*) AS c1, COUNT (*) AS c2 FROM t2 limit 100
It throws the following exception:
!TypeException:user.s5_3[21]:'algebra.semijoin' undefined in: _31:any := algebra.semijoin(_30:wrd, _28:wrd)
!SQLException:SQLengine:Program contains errors
The following queries do work:
1. Same but with only 1 aggregation field: SELECT COUNT (*) AS c1 FROM t2 limit 100
2. Same but without LIMIT: SELECT COUNT (*) AS c1, COUNT (*) AS c2 FROM t2
## Comment 12432
Date: 2009-09-19 20:54:50 +0200
From: @njnes
fixed the problem, we correctly handle single values with limits now.
added test to src/test/BugTracker-2009/Tests/double_count_limit_bug.SF-2862146.sql
## Comment 12433
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2862146 at http://sourceforge.net/support/tracker.php?aid=2862146
| Two COUNT(*) with LIMIT doesn't execute | https://api.github.com/repos/MonetDB/MonetDB/issues/2196/comments | 0 | 2020-11-30T10:27:08Z | 2024-06-27T11:41:46Z | https://github.com/MonetDB/MonetDB/issues/2196 | 753,353,373 | 2,196 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-17 12:13:01 +0200
From: Axel Belinfante <<axelbel>>
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2009-09-29 05:12:37 +0200
## Comment 12425
Date: 2009-09-17 12:13:01 +0200
From: Axel Belinfante <<axelbel>>
when I paste the forst following (lower down) xquery fragment at the interactive prompt of mclient -lx -p portnumber I get an error:
MAPI = monetdb@localhost:51012
QUERY = declare function add-int-attr($element as element()
ERROR = !fatal error: column 'item72' with type '0x1' not found in environment
! halted in milgen.brg (env_mvar_), line 695
when I simplify the stuff that I paste (as shown in second block, even lower down)
it does work as expected.
this is on linux Linux ewi865 2.6.16.21-0.8-smp 1 SMP Mon Jul 3 18:25:39 UTC 2006 x86_64 x86_64 x86_64 GNU/Linux
with the Aug-2009 version.
This used to work in the May 2009 version (same machine).
Sidenote: when I access the same functionality via xrpc I get a different error:
Error occurred during execution.
!ERROR: BATappend: access denied to tmp_375, aborting.
!ERROR: BATappend_wrap: operation failed.
====== first block start ===================
declare function add-int-attr($element as element()
,$val as xs:integer?)
as element()?
{
element {$element/name()} {
attribute { 'seq' } { $val },
$element/@*,
$element/node()
}
};
declare updating function save-view($savedoc as xs:string
,$name as xs:string
,$user as xs:string
,$view as element())
{
let $doc := doc($savedoc)
let $save := $doc/views/active/view[@name eq $name and @user eq $user]
return
if (empty($save))
then
let $v := add-int-attr($view, 0)
return do insert $v as last into $doc/views/active
else (
let $n := (abs($save/@seq) +1 ) cast as xs:integer?
let $v := add-int-attr($view, $n)
return (
do replace $save with $v,
do insert $save as last into $doc/views/archive
)
)
};
save-view("view-defs.xml", "zambon", "axel", <view now='1253179900.00036' name='zambon' prevseq='0' user='axel'><select allany='' inex='include' ><cond cmp='eq' field='creators_family'>zambon</cond></select></view>)
====== first block end ===================
====== second block start ================
declare function add-int-attr($element as element()
,$val as xs:integer?)
as element()?
{
element {$element/name()} {
attribute { 'seq' } { $val },
$element/@*,
$element/node()
}
};
declare updating function save-view($savedoc as xs:string
,$name as xs:string
,$user as xs:string
,$view as element())
{
let $doc := doc($savedoc)
let $save := $doc/views/active/view[@name eq $name and @user eq $user]
return
if (empty($save))
then
let $v := $view
return do insert $v as last into $doc/views/active
else (
let $v := $view
return (
do replace $save with $v,
do insert $save as last into $doc/views/archive
)
)
};
save-view("view-defs.xml", "zambon", "axel", <view now='1253179900.00036' name='zambon' prevseq='0' user='axel'><select allany='' inex='include' ><cond cmp='eq' field='creators_family'>zambon</cond></select></view>)
====== second block end ===================
## Comment 12426
Date: 2009-09-24 15:39:38 +0200
From: @drstmane
With the two XQuery blocks saved in /tmp/2860574-1.xq & /tmp/2860574-2.xq, respectively, I get
$ pf /tmp/2860574-1.xq > /tmp/2860574-1.mil
looking for col: item72, ty: 0x1 in function do_reduce (line 9708)
environment looks like:
col: item72, type: 0x100000, mvar->name: a0037
col: item72, type: 0x200000, mvar->name: a0039
col: iter73, type: 0x1, mvar->name: a0036
fatal error: column 'item72' with type '0x1' not found in environment
halted in /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/pathfinder/compiler/mil/milgen.brg (env_mvar_), line 695
$ wc /tmp/2860574-1.mil
0 0 0 /tmp/2860574-1.mil
$ pf /tmp/2860574-2.xq > /tmp/2860574-2.mil
$ wc /tmp/2860574-2.mil
2405 9605 92523 /tmp/2860574-2.mil
Looks like a compiler / optimizer problem to me ...
Jan,
could you plase have a look?
Thanks!
## Comment 12427
Date: 2009-09-28 15:46:05 +0200
From: Jan Rittinger <<jan.rittinger>>
Added fix in stable version (compiler/algebra planner.c, Aug2009, 1.86, 1.86.2.1).
## Comment 12428
Date: 2009-09-28 18:23:12 +0200
From: @drstmane
Thanks for the fix, Jan --- reopened to remind us that we need to add a test as well ...
## Comment 12429
Date: 2009-09-29 17:12:33 +0200
From: @drstmane
Added test in pathfinder/tests/BugTracker/Tests/compilation_error.SF-2860574*
## Comment 12430
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2860574 at http://sourceforge.net/support/tracker.php?aid=2860574
| PF: fatal error: column not found / BATappend access denied | https://api.github.com/repos/MonetDB/MonetDB/issues/2195/comments | 0 | 2020-11-30T10:27:05Z | 2024-06-27T11:41:45Z | https://github.com/MonetDB/MonetDB/issues/2195 | 753,353,334 | 2,195 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-16 04:33:22 +0200
From: Axel Belinfante <<axelbel>>
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2009-09-29 05:12:39 +0200
## Comment 12418
Date: 2009-09-16 16:33:22 +0200
From: Axel Belinfante <<axelbel>>
attempting to run the attached xquery results in fatal error: aborted too deep recursion.
replacing the call to processISBN(...) by () is sufficient to 'make it work'.
to reproduce this, it is sufficient to add a document named eprints-export.xml
with contents
<repository/>
mclient -lx -p 51012 ./convexport-dummy.xq
MAPI = monetdb@localhost:51012
QUERY = declare function processISBN($fields as element()*) as element()*
ERROR = !fatal error: aborted too deep recursion
! halted in mil_dce.c (mil_dce_worker), line 146
this is with the aug-2009 monetdb (same with the may-2009) release,
on linux:
Linux ewi865 2.6.16.21-0.8-smp 1 SMP Mon Jul 3 18:25:39 UTC 2006 x86_64 x86_64 x86_64 GNU/Linux
## Comment 12419
Date: 2009-09-24 15:58:57 +0200
From: @drstmane
This looks like when embedded in Mserver, the pf compiler (or its optimizer) requires more stack space (due to deep recursion) than available to compile/optimize your query:
$ diff -du /tmp/convexport.xq /tmp/convexport-dummy.xq
--- /tmp/convexport.xq 2009-09-24 15:40:01.000000000 +0200
+++ /tmp/convexport-dummy.xq 2009-09-24 15:41:47.000000000 +0200
@@ -54,7 +54,7 @@
let $ed := processNames("editor",$flds[@name eq "editors"])
let $sup := processNames("supervisor",$flds[@name eq "supervisors"])
let $asup := processNames("assistant_supervisor",$flds[@name eq "assistant_supervisors"])
- let $isbn := processISBN($flds[@name eq "isbn" or @name eq "isbn_13"])
+ let $isbn := ($flds[@name eq "isbn" or @name eq "isbn_13"])
let $ep :=
<eprint>{
$cr,$ed,$sup,$asup,$isbn,
$ pf /tmp/convexport.xq > /tmp/convexport.mil
$ wc /tmp/convexport.mil
15179 60656 537972 /tmp/convexport.mil
$ pf /tmp/convexport-dummy.xq > /tmp/convexport-dummy.mil
$ wc /tmp/convexport-dummy.mil
13515 53969 478507 /tmp/convexport-dummy.mil
$ cat /tmp/eprints-export.xml
<repository/>
$ mclient -lx -s'pf:add-doc("/tmp/eprints-export.xml","eprints-export.xml")'
$ mclient -lx -s'doc("eprints-export.xml")'
<?xml version="1.0" encoding="utf-8"?>
<repository/>
$ pf /tmp/convexport-dummy.xq | mclient -lmil
<?xml version="1.0" encoding="utf-8"?>
<XQueryResult><repository/></XQueryResult>
$ pf /tmp/convexport.xq | mclient -lmil
<?xml version="1.0" encoding="utf-8"?>
<XQueryResult><repository/></XQueryResult>
$ mclient -lx /tmp/convexport-dummy.xq
MAPI = monetdb@localhost:50000
QUERY = declare function processISBN($fields as element()*) as element()*
ERROR = !fatal error: aborted too deep recursion
! halted in /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/pathfinder/compiler/mil/mil_dce.c (mil_dce_worker), line 146
$ mclient -lx /tmp/convexport.xq
MAPI = monetdb@localhost:50000
QUERY = declare function processISBN($fields as element()*) as element()*
ERROR = !fatal error: aborted too deep recursion
! halted in /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/pathfinder/compiler/mil/mil_dce.c (mil_dce_worker), line 146
Maybe Jan has some idea or comment?
## Comment 12420
Date: 2009-09-28 14:55:57 +0200
From: Jan Rittinger <<jan.rittinger>>
A MIL program is internally represented as a tree. Assignments are separated by a seq operator (that builds a sequence). As the code generation builds programs in a left-deep manner the nesting depth in mil_dce.c depends on the number of seq operators (and thus also indirectly on the number of MIL code lines). The conclusion is: the bigger the program, the deeper the nesting becomes.
In the attached example the DCE sees a nesting depth of ~12.400.
It seems awkward to me to produce 'bushy' MIL plans to stay alive longer. A better solution would be to adjust the tree walker to handle operator seq separately and avoid the recursion fence alltogether.
Sjoerd, do you have any favourite solution in mind?
## Comment 12421
Date: 2009-09-28 17:45:47 +0200
From: Jan Rittinger <<jan.rittinger>>
I just checked in a fix (compiler/mil mil_dce.c, Aug2009, 1.29.2.1, 1.29.2.2) that delays the stack space problem (by hopefully a factor of 10). If we hit the problem again we need to adjust the MIL code generation.
The query now compiles fine in client setting.
## Comment 12422
Date: 2009-09-28 18:23:27 +0200
From: @drstmane
Thanks for the fix, Jan --- reopened to remind us that we need to add a test as well ...
## Comment 12423
Date: 2009-09-29 17:12:39 +0200
From: @drstmane
Added test in pathfinder/tests/BugTracker/Tests/compilation_error.SF-2860037*
## Comment 12424
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2860037 at http://sourceforge.net/support/tracker.php?aid=2860037
| PF: fatal error: aborted too deep recursion on linux | https://api.github.com/repos/MonetDB/MonetDB/issues/2194/comments | 0 | 2020-11-30T10:27:02Z | 2024-06-27T11:41:44Z | https://github.com/MonetDB/MonetDB/issues/2194 | 753,353,286 | 2,194 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-16 03:24:29 +0200
From: Axel Belinfante <<axelbel>>
To: @yzchang
Version: -- development
Last updated: 2011-07-18 14:24:08 +0200
## Comment 12411
Date: 2009-09-16 15:24:29 +0200
From: Axel Belinfante <<axelbel>>
doing an xquery via mclient works immediately after starting Monetdb.
doing then a couple of xrpc requests somehow messes up some state in the server,
such that then doing the same xquery request via mclient once more yields an ERROR result,
or an empty result.
I have seen cases where at some point mclient continuously returns ERROR for all xquery queries,
even for simple ones like <p>aap</p> .
in other cases I just get an empty result.
it is easy to reproduce on our system. but turned out to be harder to isolate such
that it still is reproducable. I hope I succeeded in that.
server: Aug2009 Monetdb release, running on (uname -a output):
Linux ewi865 2.6.16.21-0.8-smp 1 SMP Mon Jul 3 18:25:39 UTC 2006 x86_64 x86_64 x86_64 GNU/Linux
## Comment 12412
Date: 2009-09-16 15:42:34 +0200
From: Axel Belinfante <<axelbel>>
I atdded the files that I used to set up the database at
http://library.cs.utwente.nl/ecv/files-for-bugreport-2859986.tgz
(size: 22861872 bytes)
the file start-mxq-test.mil in the .tgz shows how I initialized the database.
I invoked the xrpc commands by using telnet from within an xterm,
and pasting the contents of the attached files.
I invoked mclient as: mclient -lx -p myportnumber mclient-query06
so, when I cleanup the database 'var' directory, and restart it the mclient query works fine
after I have done the first xrpc query, the mclient query still works fine
after I have done the second xrpc query, the mclient query still works fine
after I have done the third xrpc query, the mclient query returns an 'empty' result
( <div><div class="items"/></div> )
however, if I then do the same request via xrpc, by repeating the third xrpc request,
I do get a correct result...
## Comment 12413
Date: 2009-09-17 10:11:38 +0200
From: Axel Belinfante <<axelbel>>
I wrote: "however, if I then do the same request via xrpc, by repeating the third xrpc request, I do get a correct result..."
this is not always the case.
sometimes then such xrpc request returns an empty result ( <div><div class="items"/></div> ).
Sometimes when the mclient request fails, it returns 'ERROR', together with almost the correct result:
the result is correct, but the first couple of characters are missing (overwritten by the ERROR?).
Finally, I don't know whether it is related to this issue, it happens that the server gets into a state where a pf:add-doc that is done via mclient just hangs, forever, until I kill the server to restart it.
This issue was already present in the May-2009 release, and is still present in the Aug-2009 release.
## Comment 12414
Date: 2009-09-17 18:10:36 +0200
From: @yzchang
Hello Axel,
Thank you very much for reporting this problem. In the very beginning when xrpc was implemented, we have notices that there are situations, in which xrpc conflicts with mclient. We have tried to solve the cases we have noticed, but we could have missed some cases. I will look into your problem soon and post more information here, as soon as I know more.
Regards,
Jennie
## Comment 12415
Date: 2009-09-25 14:20:33 +0200
From: @yzchang
Hello Axel,
I have run your queries, and I can reproduce it (the xrpc request messages were first translated into xquery queries with xrpc calls). I think it is a bug in the function cache in MPS.
Function cache in pathfinder: in the (old) MPS front-end, the whole module file is compiled, when it is 'import'-ed by a query. The compiler output of each function (i.e., the MIL-tree) in the module file is cached, so that a subsequent call to a function in the same module could be compiled (much) faster. So, in an Mserver session, each module file is only compiled once (inline, user-defined-functions are not cached). Function cache only works with MPS front-end, not with the (new) algebra front-end.
For all functions called over XRPC, function cache is used. For normal queries (i.e., without any xrpc statement), function cache is used, if we can detect that the query contains a *single* call to a predefined function, with *only* simple literal paramters (i.e., each parameter is a single atomic-value). In Pathfinder runtime, we always check if function cache could be used, even if mclient has been started to use the algebra front-end. Once we found out that a query could be handeled with cached function, we automatically switch to the MPS front-end. And this is what has happend to your mclient-query06.xq, since it calls one module function with a single string parameter. Hence, both your normal query and your xrpc queries use the same cached MIL-trees, and most probably, the problem is caused by a not completely/correctly cleaned-up environment.
I don't know how to fix this problem yet. I only have some suggestions as workaround of this problem. I hope they could be helpful for you, so that you could continue your work. Basically, what we want to do here, is to prevent pathfinder from using function cache for your normal query. This could be done in different ways (I have tried both and it seems working):
1. do not use module function in your normal query, but directly inline the function bodies in the query. For instance, I have replaced function calls in mclient-query06.xq as the following:
$ cat mclient-query06.xq
(:
import module namespace ecv = "http://www.cs.utwente.nl/~belinfan/ecv"
at "/ufs/zhang/tmp/axelbel/ECVtest/xrpc/export/dyn/repo/dynxquery/axel/zambon/0";
ecv:view-from-save("eprints-export-conv-fmt.xml")
:)
let $eps :=
for $ep in doc("eprints-export-conv-fmt.xml")//tuple
where (
(
(
( some $c in $ep//creators_family satisfies contains(upper-case($c), "ZAMBON"))
)
)
)
return $ep
return
<div>{
<div class="items">{$eps//fmt/div}</div>
}</div>
advantage: query compilation time is short (~30ms with MPS, ~20ms with algebra), as no unused functions are compiled.
disadvantage: query maintainance, as you must update the normal query, if the corresponding functions in the module file have been changed.
2. this is very hacky: pass more complex parameter to the function call, so that it cannot be detected by the pathfinder runtime, as for which function cache could be used. This could be done very easily, for example:
import module namespace ecv = "http://www.cs.utwente.nl/~belinfan/ecv"
at "/ufs/zhang/tmp/axelbel/ECVtest/xrpc/export/dyn/repo/dynxquery/axel/zambon/0";
ecv:view-from-save( ("eprints-export-conv-fmt.xml") )
^ ^
Note the extra parentheses, now the parameter has become a sequence containing one string value. It is still valid, but will prevent the pathfinder runtime from using function cache :P
advantage: ease of maintainance, minimal change of code.
disadvantage: long compile time! For each normal query, the complete module must be load and compiled. With algebra, this costs ~300ms; with MPS, ~700ms.
Would you please let us know if you can live with any of my suggestions, so that we can increase the priority of this bug accordingly?
Regards,
Jennie
## Comment 12416
Date: 2009-09-27 23:07:08 +0200
From: Axel Belinfante <<axelbel>>
Thanks for your efforts to reproduce and explain the problem.
Having read it, I wonder whether this is related to what I reported as issue 2864991
( https://sourceforge.net/tracker/?func=detail&aid=2864991&group_id=56967&atid=482468 )
I'll have to think on how to proceed next.
I had already tried to work around the problem by using mapin instead of xrpc,
(which I where I encountered that problem 2864991, which I then reproduced using only mclient).
However, from your explanation I gather that, as long as I continue to use queries
that call a module function with only (a single) string parameter(s), I will continue to
encounter the problem, because pathfinder will try to use the caching.
to give some background: ", building a web application (via ruby-on-rails)
that allows a user to construct queries. these queries are saved
in abstract form of xml, as well as xquery module.
subsequently, these saved xqueries are used via a cgi script.
performance (speed) is important to us, especially in the last part:
the use of queries via cgi. ideally then they should appear to be
as fast as an ordinary html page request.
however, also when the user is constructing queries, and trying these out,
performance matters, to give the user a responsive application.
we (designers of the application) have been discussing the idea of
using pf to compile a query to when we save it in the web-application,
such that execution of the pre-compiled query can be very fast
(both when it is run from the web-app, as well as via the cgi script)
after this long intro, the question: would such compilation help to
avoid the issue?
(sidenote: when I tried to compile queries with the May 2009 release,
I encountered cases where the compiler would appear to run forever;
I did not yet retry with the latest release, but from release notes got
the impression that in this respect this should have improved?)
## Comment 12417
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2859986 at http://sourceforge.net/support/tracker.php?aid=2859986
## Comment 15969
Date: 2011-07-18 14:24:08 +0200
From: @sjoerdmullender
Development on MonetDB4 and MonetDB/XQuery has ceased. It is unlikely that
this bug will be fixed.
The last release of MonetDB/XQuery and MonetDB4 is Mar2011.
| xrpc use messes up mclient use | https://api.github.com/repos/MonetDB/MonetDB/issues/2193/comments | 0 | 2020-11-30T10:26:58Z | 2024-06-28T13:22:44Z | https://github.com/MonetDB/MonetDB/issues/2193 | 753,353,241 | 2,193 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-11 01:32:41 +0200
From: francois guerin <<fguerin>>
To: @yzchang
Version: -- development
Last updated: 2011-07-18 14:23:08 +0200
## Comment 12405
Date: 2009-09-11 13:32:41 +0200
From: francois guerin <<fguerin>>
Hi,
I have a xquery request that gives two differents results:
- the first one, within "mclient" works well, and gives me the right answer:
xquery>doc("frwikipedia_001.xml")//Sentence[@id="E1"]
- the second one, within a .xq module, fails and gives me the error below:
declare function foo:get_raw_enonce($document as xs:string, $eid as xs:string){
doc($document)//Sentence[@id=$eid]
};
get_raw_enponce("frwikipedia_001.xml","E1")
<?xml version="1.0" encoding="utf-8"?>
<env:Envelope xmlns:env="http://www.w3.org/2003/05/soap-envelope">
<env:Body>
<env:Fault>
<env:Code>
<env:Value>env:Sender</env:Value>
</env:Code>
<env:Reason>
<env:Text xml:lang="en">Error occurred during execution.
!ERROR: BATappend: access denied to tmp_2432, aborting.
!ERROR: BATappend_wrap: operation failed.
</env:Text>
</env:Reason>
</env:Fault>
</env:Body>
</env:Envelope>
do i need admin rights to perform this request?
or, one can't do the same things with mclient and with xrpc requests ?
(the source file is still there: http://fguerin.inria.perso.neuf.fr/final_wiki.tar.bz2 )
## Comment 12406
Date: 2009-09-11 14:20:31 +0200
From: @yzchang
Hello Francois,
No, you don't need admin rights to perform this request, since, I guess, your Mserver and mclient are running on the same host. After a quick look, your query should work with xrpc, as it does nothing exotic.
Unfortunately, I don't have time to investigate further in your problem right now. I will do it after Sep. 16th.
Do I recall correctly that you are running the latest release (i.e. August 2009) on Win. XP (32/64 bits?)? If I remenber it wrongly, please give me some OS, software version information in the mean time.
By the way, could you please also try to start Mserver with an extra option -d10 , and see if you get the same error?
Kind regards,
Jennie
## Comment 12407
Date: 2009-09-11 14:52:27 +0200
From: francois guerin <<fguerin>>
with the option -d10, the request within mclient is very slow (i didnt take time to wait for the answer, but it's at least 5min long)
I use the latest release.
My computer is a Mandriva - 64 bits - 4 proc.
## Comment 12408
Date: 2009-09-18 17:34:24 +0200
From: @sjoerdmullender
You declare the function using the name foo:get_raw_enonce, but use it using the name get_raw_enponce (note the extra p). Is that a typo in the bug report or perhaps the cause of the problem (in which case the error could be clearer)?
## Comment 12409
Date: 2009-09-21 10:42:34 +0200
From: @yzchang
No, it should be just a copy-paste error, because, otherwise, xrpc will send back a parse error.
I can reproduce the reported error. However, the query runs fine with xrpc (i.e., it produced seemly correct answer, i.s.o. error), if I start Mserver with -d10.
I haven't used gdb to look deeper into the problem, because the raw xml file is ~4.8GB. Can someone maybe give me some hints how I can debug this more quickly? Thanks a lot!
Jennie
## Comment 12410
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2856798 at http://sourceforge.net/support/tracker.php?aid=2856798
## Comment 15968
Date: 2011-07-18 14:23:08 +0200
From: @sjoerdmullender
Development on MonetDB4 and MonetDB/XQuery has ceased. It is unlikely that
this bug will be fixed.
The last release of MonetDB/XQuery and MonetDB4 is Mar2011.
| mclient vs. xrpc requests - !ERROR: BATappend | https://api.github.com/repos/MonetDB/MonetDB/issues/2192/comments | 0 | 2020-11-30T10:26:54Z | 2024-06-28T13:22:44Z | https://github.com/MonetDB/MonetDB/issues/2192 | 753,353,199 | 2,192 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-09 09:01:41 +0200
From: @skinkie
To: @njnes
Version: -- development
CC: @sjoerdmullender
Last updated: 2010-07-09 15:09:32 +0200
## Comment 12401
Date: 2009-09-09 21:01:41 +0200
From: @skinkie
sql>select cast('2004-05-01 23:00' as timestamp)+cast('1:00:00' as time);
!types timestamp(7,0) (timestamp) and tinyint(8,0) (bte) are not equal
select cast('2004-05-01 23:00' as timestamp)+3600;
+----------------------------+
| sql_add_single_value |
+============================+
| 2004-05-02 [000000000000](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=000000000000) |
+----------------------------+
## Comment 12402
Date: 2009-09-09 21:02:45 +0200
From: @skinkie
sf glitch
## Comment 12403
Date: 2009-09-09 21:04:02 +0200
From: @grobian
PEBKAC
## Comment 12404
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2855415 at http://sourceforge.net/support/tracker.php?aid=2855415
Bug has invalid status, setting status to "NEW".
Previous status was "DELETED".
## Comment 14322
Date: 2010-07-09 15:09:32 +0200
From: @sjoerdmullender
Bug was DELETED in SF.
| SQL: adding timestamp + time is impossible | https://api.github.com/repos/MonetDB/MonetDB/issues/2191/comments | 0 | 2020-11-30T10:26:51Z | 2024-06-28T13:22:43Z | https://github.com/MonetDB/MonetDB/issues/2191 | 753,353,158 | 2,191 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-09 06:55:08 +0200
From: @skinkie
To: GDK devs <<bugs-common>>
Version: -- development
Last updated: 2009-09-10 10:29:01 +0200
## Comment 12398
Date: 2009-09-09 18:55:08 +0200
From: @skinkie
I saw that when doing a large paste of multiple queries while mclient is waiting for a server result and redirected to a file, the input is not parsed correctly and it seems that it skips data.
I reproduce it by an all in one paste interleaving \>/bla with a long query.
## Comment 12399
Date: 2009-09-10 10:28:59 +0200
From: @sjoerdmullender
I don't believe for a second that this is a problem in mclient. I do believe this can be a problem.
I think the problem is in the interaction between the readline library and the kernel where readline sets the terminal in edit mode for each line of input and sets the terminal back at the end of the line. During this setting, on some systems, pending input may well be flushed. I have definitely seen this kind of behavior.
Therefore I'm closing this with Won't Fix since it is not anything we can do anything about.
## Comment 12400
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2855324 at http://sourceforge.net/support/tracker.php?aid=2855324
The original assignee of this bug does not have
an account here. Reassigning to the default assignee
for the component, bugs-common@monetdb.org.
Previous assignee was nobody@users.sourceforge.net.
| Mclient can overshoots its input buffer on large paste | https://api.github.com/repos/MonetDB/MonetDB/issues/2190/comments | 0 | 2020-11-30T10:26:47Z | 2024-06-28T13:22:42Z | https://github.com/MonetDB/MonetDB/issues/2190 | 753,353,099 | 2,190 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-09 02:31:41 +0200
From: francois guerin <<fguerin>>
To: Fabian Groffen <<fabian>>
Version: -- development
Last updated: 2009-09-12 12:24:32 +0200
## Comment 12388
Date: 2009-09-09 14:31:41 +0200
From: francois guerin <<fguerin>>
the problem was not fixed.
(cf. bug ID: 2848525)
now i have the errors below:
php_monetdb.c: In function 'zif_monetdb_close':
php_monetdb.c:724: warning: 'zend_get_parameters_ex' is deprecated (declared at /usr/include/php/Zend/zend_API.h:222)
php_monetdb.c: In function 'php_monetdb_get_link_info':
php_monetdb.c:772: warning: 'zend_get_parameters_ex' is deprecated (declared at /usr/include/php/Zend/zend_API.h:222)
php_monetdb.c: In function 'zif_monetdb_query':
php_monetdb.c:902: warning: 'zend_get_parameters_ex' is deprecated (declared at /usr/include/php/Zend/zend_API.h:222)
php_monetdb.c:909: warning: 'zend_get_parameters_ex' is deprecated (declared at /usr/include/php/Zend/zend_API.h:222)
php_monetdb.c: In function 'php_monetdb_get_result_info':
php_monetdb.c:1326: warning: 'zend_get_parameters_ex' is deprecated (declared at /usr/include/php/Zend/zend_API.h:222)
php_monetdb.c: In function 'php_monetdb_get_field_info':
php_monetdb.c:1410: warning: 'zend_get_parameters_ex' is deprecated (declared at /usr/include/php/Zend/zend_API.h:222)
php_monetdb.c: In function 'zif_monetdb_field_num':
php_monetdb.c:1482: warning: 'zend_get_parameters_ex' is deprecated (declared at /usr/include/php/Zend/zend_API.h:222)
php_monetdb.c: In function 'zif_monetdb_fetch_result':
php_monetdb.c:1514: warning: 'zend_get_parameters_ex' is deprecated (declared at /usr/include/php/Zend/zend_API.h:222)
php_monetdb.c:1515: warning: 'zend_get_parameters_ex' is deprecated (declared at /usr/include/php/Zend/zend_API.h:222)
php_monetdb.c: In function 'php_monetdb_fetch_hash':
php_monetdb.c:1719: error: 'zend_fcall_info' has no member named 'object_pp'
php_monetdb.c:1757: error: 'zend_fcall_info_cache' has no member named 'object_pp'
php_monetdb.c: In function 'php_monetdb_data_info':
php_monetdb.c:1858: warning: 'zend_get_parameters_ex' is deprecated (declared at /usr/include/php/Zend/zend_API.h:222)
php_monetdb.c:1859: warning: 'zend_get_parameters_ex' is deprecated (declared at /usr/include/php/Zend/zend_API.h:222)
php_monetdb.c: In function 'zif_monetdb_free_result':
php_monetdb.c:1956: warning: 'zend_get_parameters_ex' is deprecated (declared at /usr/include/php/Zend/zend_API.h:222)
gnumake[8]: *** [php_monetdb.lo] Error 1
gnumake[7]: *** [all-recursive] Error 1
gnumake[6]: *** [all] Error 2
gnumake[5]: *** [all-recursive] Error 1
gnumake[4]: *** [all] Error 2
gnumake[3]: *** [all-recursive] Error 1
gnumake[2]: *** [all] Error 2
gnumake[1]: *** [all-recursive] Error 1
gnumake: *** [all] Error 2
making MonetDB-client failed
Please refer to the full log at /private/var/tmp/MonetDB-1252498422/log
If you believe this is an error in the script or the software,
file a bug and attach the logfile.
Terminated
Thanks.
## Comment 12389
Date: 2009-09-09 14:35:53 +0200
From: @grobian
You just have a new problem with PHP.
checking for PHP... "found: 5.3.0"
checking for Zend/zend_exceptions.h... yes
checking for PHP includes and libraries... PHP_INCS=" -I/usr/include/php -I/usr/include/php/main -I/usr/include/php/TSRM -I/usr/include/php/Zend -I/usr/include/php/ext -I/usr/include/php/ext/date/lib" PHP_EXTENSIONDIR="$prefix/lib/php/extensions/no-debug-non-zts-20090626"
...
checking for PHP < 5.3.0... "no (5.3.0)"
so apparently my previous fix (unrelated to OSX 10.6) didn't work quite well for this.
## Comment 12390
Date: 2009-09-09 14:37:34 +0200
From: @grobian
you can work around this one by exporting EXTRA_ECONF="--with-php=no"
## Comment 12391
Date: 2009-09-09 14:39:16 +0200
From: @grobian
I found a possible cause for this problem. Hopefully it will be fixed tomorrow. Thanks!
## Comment 12392
Date: 2009-09-11 14:26:34 +0200
From: francois guerin <<fguerin>>
rmpf. still not working.
- even with the lastest nightly release.
- even with the ""--with-php=no" trick.
## Comment 12393
Date: 2009-09-11 14:29:03 +0200
From: @grobian
I'm sorry, can you show me the snippet where it goes wrong now?
## Comment 12394
Date: 2009-09-11 14:56:51 +0200
From: francois guerin <<fguerin>>
oh sorry, i forgot to attach the log file.
it's attached now (fguerin_log_11_sept.txt)
## Comment 12395
Date: 2009-09-11 15:13:37 +0200
From: @grobian
Thank you. I did another try to fix this. Sorry for the hassle.
## Comment 12396
Date: 2009-09-12 12:24:32 +0200
From: @grobian
I was just able to successfully compile clients. We can't compile MonetDB5 on a stock 10.6 though, since it doesn't come with libpcre, so you'll have to provide that via other ways.
Thanks!
## Comment 12397
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2855152 at http://sourceforge.net/support/tracker.php?aid=2855152
| mac10.6 and monetdb installation | https://api.github.com/repos/MonetDB/MonetDB/issues/2189/comments | 0 | 2020-11-30T10:26:43Z | 2024-06-27T11:41:39Z | https://github.com/MonetDB/MonetDB/issues/2189 | 753,353,054 | 2,189 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-09 10:23:33 +0200
From: @skinkie
To: @njnes
Version: -- development
Last updated: 2009-09-15 08:19:21 +0200
## Comment 12381
Date: 2009-09-09 10:23:33 +0200
From: @skinkie
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7f2f6ae60910 (LWP 27586)]
0x00007f2f6b0f6b02 in scanner_query_processed (s=0x2a7d008)
at ../../../src/server/sql_scan.mx:429
429 cur = s->rs->buf[s->rs->pos];
(gdb) bt
0 0x00007f2f6b0f6b02 in scanner_query_processed (s=0x2a7d008)
at ../../../src/server/sql_scan.mx:429
1 0x00007f2f6b08a3c8 in sqlcleanup (c=0x2a7afd8, err=0)
at ../../../../src/backends/monet5/sql.mx:1296
2 0x00007f2f6b0c6429 in SQLengineIntern (c=0x6057a8, be=0x7f2f64045018)
at ../../../../src/backends/monet5/sql_scenario.mx:1236
3 0x00007f2f6b0c6a50 in SQLengine (c=0x6057a8)
at ../../../../src/backends/monet5/sql_scenario.mx:1367
4 0x00007f2f7cec2f85 in runPhase (c=0x6057a8, phase=4)
at ../../../src/mal/mal_scenario.mx:602
5 0x00007f2f7cec3121 in runScenarioBody (c=0x6057a8)
at ../../../src/mal/mal_scenario.mx:648
6 0x00007f2f7cec32e1 in runScenario (c=0x6057a8)
at ../../../src/mal/mal_scenario.mx:672
7 0x00007f2f7ce7a185 in MSserveClient (dummy=0x6057a8)
at ../../../src/mal/mal_session.mx:496
8 0x00007f2f7c69c1f2 in wrapper_routine (data=0x7f2f6b5f9c40)
at ../../../src/gdk/gdk_posix.mx:983
9 0x00007f2f7b390624 in start_thread () from /lib/libpthread.so.0
10 0x00007f2f7a4ba56d in clone () from /lib/libc.so.6
11 0x0000000000000000 in ?? ()
Lets say what I did to create it was the following; I was working on an insert script which worked as a charm:
sh station.sh | /opt/monetdb-head/bin/mclient -lsql /dev/stdin
After that I removed the /dev/stdin, (accidently pressed enter), and saw the SEGV above;
The query that comes out of the script is:
INSERT INTO O3 SELECT date, 107, station107 FROM (SELECT date, station107 FROM rivm_012000_O3 UNION ALL SELECT date, station107 FROM rivm_012001_O3 UNION ALL SELECT date, station107 FROM rivm_022000_O3 UNION ALL SELECT date, station107 FROM rivm_022001_O3 UNION ALL SELECT date, station107 FROM rivm_032000_O3 UNION ALL SELECT date, station107 FROM rivm_032001_O3 UNION ALL SELECT date, station107 FROM rivm_042000_O3 UNION ALL SELECT date, station107 FROM rivm_042001_O3 UNION ALL SELECT date, station107 FROM rivm_052000_O3 UNION ALL SELECT date, station107 FROM rivm_052001_O3 UNION ALL SELECT date, station107 FROM rivm_062000_O3 UNION ALL SELECT date, station107 FROM rivm_062001_O3 UNION ALL SELECT date, station107 FROM rivm_072000_O3 UNION ALL SELECT date, station107 FROM rivm_072001_O3 UNION ALL SELECT date, station107 FROM rivm_082000_O3 UNION ALL SELECT date, station107 FROM rivm_082001_O3 UNION ALL SELECT date, station107 FROM rivm_092000_O3 UNION ALL SELECT date, station107 FROM rivm_092001_O3 UNION ALL SELECT date, station107 FROM rivm_102000_O3 UNION ALL SELECT date, station107 FROM rivm_102001_O3 UNION ALL SELECT date, station107 FROM rivm_112000_O3 UNION ALL SELECT date, station107 FROM rivm_112001_O3 UNION ALL SELECT date, station107 FROM rivm_122000_O3 UNION ALL SELECT date, station107 FROM rivm_122001_O3 UNION ALL SELECT date, station107 FROM rivm_apr_2002_O3 UNION ALL SELECT date, station107 FROM rivm_april2003_O3 UNION ALL SELECT date, station107 FROM rivm_april2004_O3 UNION ALL SELECT date, station107 FROM rivm_april2005_O3 UNION ALL SELECT date, station107 FROM rivm_april2006_O3 UNION ALL SELECT date, station107 FROM rivm_april2007_O3 UNION ALL SELECT date, station107 FROM rivm_april2008_O3 UNION ALL SELECT date, station107 FROM rivm_april2009_O3 UNION ALL SELECT date, station107 FROM rivm_aug_2002_O3 UNION ALL SELECT date, station107 FROM rivm_augustus2003_O3 UNION ALL SELECT date, station107 FROM rivm_augustus2004_O3 UNION ALL SELECT date, station107 FROM rivm_augustus2005_O3 UNION ALL SELECT date, station107 FROM rivm_augustus2006_O3 UNION ALL SELECT date, station107 FROM rivm_augustus2007_O3 UNION ALL SELECT date, station107 FROM rivm_augustus2008_O3 UNION ALL SELECT date, station107 FROM rivm_dec_2002_O3 UNION ALL SELECT date, station107 FROM rivm_december2003_O3 UNION ALL SELECT date, station107 FROM rivm_december2004_O3 UNION ALL SELECT date, station107 FROM rivm_december2005_O3 UNION ALL SELECT date, station107 FROM rivm_december2006_O3 UNION ALL SELECT date, station107 FROM rivm_december2007_O3 UNION ALL SELECT date, station107 FROM rivm_december2008_O3 UNION ALL SELECT date, station107 FROM rivm_feb_2002_O3 UNION ALL SELECT date, station107 FROM rivm_februari2004_O3 UNION ALL SELECT date, station107 FROM rivm_februari2005_O3 UNION ALL SELECT date, station107 FROM rivm_februari2006_O3 UNION ALL SELECT date, station107 FROM rivm_februari2007_O3 UNION ALL SELECT date, station107 FROM rivm_februari2008_O3 UNION ALL SELECT date, station107 FROM rivm_februari2009_O3 UNION ALL SELECT date, station107 FROM rivm_jan_2002_O3 UNION ALL SELECT date, station107 FROM rivm_januari2004_O3 UNION ALL SELECT date, station107 FROM rivm_januari2005_O3 UNION ALL SELECT date, station107 FROM rivm_januari2006_O3 UNION ALL SELECT date, station107 FROM rivm_januari2007_O3 UNION ALL SELECT date, station107 FROM rivm_januari2008_O3 UNION ALL SELECT date, station107 FROM rivm_januari2009_O3 UNION ALL SELECT date, station107 FROM rivm_jul_2002_O3 UNION ALL SELECT date, station107 FROM rivm_juli2003_O3 UNION ALL SELECT date, station107 FROM rivm_juli2004_O3 UNION ALL SELECT date, station107 FROM rivm_juli2005_O3 UNION ALL SELECT date, station107 FROM rivm_juli2006_O3 UNION ALL SELECT date, station107 FROM rivm_juli2007_O3 UNION ALL SELECT date, station107 FROM rivm_juli2008_O3 UNION ALL SELECT date, station107 FROM rivm_jun_2002_O3 UNION ALL SELECT date, station107 FROM rivm_juni2003_O3 UNION ALL SELECT date, station107 FROM rivm_juni2004_O3 UNION ALL SELECT date, station107 FROM rivm_juni2005_O3 UNION ALL SELECT date, station107 FROM rivm_juni2006_O3 UNION ALL SELECT date, station107 FROM rivm_juni2007_O3 UNION ALL SELECT date, station107 FROM rivm_juni2008_O3 UNION ALL SELECT date, station107 FROM rivm_maart2004_O3 UNION ALL SELECT date, station107 FROM rivm_maart2005_O3 UNION ALL SELECT date, station107 FROM rivm_maart2006_O3 UNION ALL SELECT date, station107 FROM rivm_maart2007_O3 UNION ALL SELECT date, station107 FROM rivm_maart2008_O3 UNION ALL SELECT date, station107 FROM rivm_maart2009_O3 UNION ALL SELECT date, station107 FROM rivm_mei_2002_O3 UNION ALL SELECT date, station107 FROM rivm_mei2003_O3 UNION ALL SELECT date, station107 FROM rivm_mei2004_O3 UNION ALL SELECT date, station107 FROM rivm_mei2005_O3 UNION ALL SELECT date, station107 FROM rivm_mei2006_O3 UNION ALL SELECT date, station107 FROM rivm_mei2007_O3 UNION ALL SELECT date, station107 FROM rivm_mei2008_O3 UNION ALL SELECT date, station107 FROM rivm_mei2009_O3 UNION ALL SELECT date, station107 FROM rivm_mrt_2002_O3 UNION ALL SELECT date, station107 FROM rivm_nov_2002_O3 UNION ALL SELECT date, station107 FROM rivm_november2003_O3 UNION ALL SELECT date, station107 FROM rivm_november2004_O3 UNION ALL SELECT date, station107 FROM rivm_november2005_O3 UNION ALL SELECT date, station107 FROM rivm_november2006_O3 UNION ALL SELECT date, station107 FROM rivm_november2007_O3 UNION ALL SELECT date, station107 FROM rivm_november2008_O3 UNION ALL SELECT date, station107 FROM rivm_okt_2002_O3 UNION ALL SELECT date, station107 FROM rivm_oktober2003_O3 UNION ALL SELECT date, station107 FROM rivm_oktober2004_O3 UNION ALL SELECT date, station107 FROM rivm_oktober2005_O3 UNION ALL SELECT date, station107 FROM rivm_oktober2006_O3 UNION ALL SELECT date, station107 FROM rivm_oktober2007_O3 UNION ALL SELECT date, station107 FROM rivm_oktober2008_O3 UNION ALL SELECT date, station107 FROM rivm_sep_2002_O3 UNION ALL SELECT date, station107 FROM rivm_september2003_O3 UNION ALL SELECT date, station107 FROM rivm_september2004_O3 UNION ALL SELECT date, station107 FROM rivm_september2005_O3 UNION ALL SELECT date, station107 FROM rivm_september2006_O3 UNION ALL SELECT date, station107 FROM rivm_september2007_O3 UNION ALL SELECT date, station107 FROM rivm_september2008_O3) as x WHERE station107 IS NOT NULL ORDER BY date;
...rather innocent and normally completely working.
## Comment 12382
Date: 2009-09-09 10:28:14 +0200
From: @grobian
what are the values of s->rs, s->rs->buf, s->rs->pos
have you tried this on vanilla sources, or your modified ones?
## Comment 12383
Date: 2009-09-09 10:42:21 +0200
From: @skinkie
0x00007f74d7a10b02 in scanner_query_processed (s=0x35932c8) at ../../../src/server/sql_scan.mx:429
429 cur = s->rs->buf[s->rs->pos];
(gdb) print s->rs->pos
$1 = 1461
(gdb) print *s->rs->buf[s->rs->pos]
Cannot access memory at address 0xb6a
(gdb) print s->rs->buf[s->rs->pos]
Cannot access memory at address 0xb6a
(gdb) print s->rs
$2 = (bstream *) 0x3590900
(gdb) print *s->rs
$3 = {s = 0x4ff, buf = 0x5b5 <Address 0x5b5 out of bounds>, size = 81737522629208, pos = 1461, len = 15842497851538791387, eof = -420114744, mode = 32628}
(gdb) print *s->rs->buf
Cannot access memory at address 0x5b5
(gdb) print s->rs->buf
$4 = 0x5b5 <Address 0x5b5 out of bounds>
(gdb) print s->rs->pos
$5 = 1461
Key observation;
- Bug can be reproduced to connect to Monet *BEFORE* the > prompt is there. So there is an:
Listening for connection requests on mapi:monetdb://127.0.0.1:50000/
- Mclient will reply:
Connection terminated
- Nuking data in after the > appears results in the segmentation fault.
Now I have identified a parental issue; the diplaying of the prompt take ages regardeless of the fact I nicely stopped the server or it crashed.
I'm not running vanilla...
## Comment 12384
Date: 2009-09-09 10:46:45 +0200
From: @grobian
That feels like a bug exposed in merovingian a while ago. You should only connect after the sql scenario becomes available. Not sure why it accepts you before that though. Does the bug happen if you wait long enough too? Or are we dealing with two bugs in one now?
## Comment 12385
Date: 2009-09-09 10:59:30 +0200
From: @skinkie
I just checked before my last reply if it would normally work if I actively looked for a >. Indeed that works.
" MonetDB/SQL module v2.33.0 loaded"
Is visible before the > appears.
And regarding to the bug in one; I think it is a bit odd that the actual loading now takes much longer than I have experienced before. I have 174 tables in my database, that is a bit more than usual ;)
## Comment 12386
Date: 2009-09-15 08:19:21 +0200
From: @njnes
The sql startup no sets the scenario name string first to a temporary name and as last action (after initializing sql)
to the correct sql/msql string. The startup of sql was also improved by caching the delta bats during startup. Closing without test as the testing infrastructure waits for the prompt to appear.
## Comment 12387
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2855021 at http://sourceforge.net/support/tracker.php?aid=2855021
| SQL: mserver accepts sql connections before scenario load | https://api.github.com/repos/MonetDB/MonetDB/issues/2188/comments | 0 | 2020-11-30T10:26:39Z | 2024-06-27T11:41:38Z | https://github.com/MonetDB/MonetDB/issues/2188 | 753,353,006 | 2,188 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-09 07:59:41 +0200
From: @skinkie
To: @njnes
Version: -- development
Last updated: 2009-09-15 08:37:51 +0200
## Comment 12378
Date: 2009-09-09 07:59:41 +0200
From: @skinkie
select timestamp '2000-01-01' + interval '1' year;
!TypeException:user.s1_1[3]:'mtime.addmonths' undefined in: _6:any := mtime.addmonths(_4:timestamp, _5:int)
!SQLException:SQLengine:Program contains errors
## Comment 12379
Date: 2009-09-15 08:37:51 +0200
From: @njnes
Added test to bugtracker-2009/addmonths_missing.SF-2854949.sql
Fixed bug by listing the correct mal function in the sql_types function list
## Comment 12380
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2854949 at http://sourceforge.net/support/tracker.php?aid=2854949
| SQL: + interval '1' year; Program contains errors | https://api.github.com/repos/MonetDB/MonetDB/issues/2187/comments | 0 | 2020-11-30T10:26:36Z | 2024-06-27T11:41:37Z | https://github.com/MonetDB/MonetDB/issues/2187 | 753,352,959 | 2,187 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-09 05:10:46 +0200
From: @skinkie
To: @njnes
Version: -- development
Last updated: 2009-09-17 09:20:23 +0200
## Comment 12375
Date: 2009-09-09 05:10:46 +0200
From: @skinkie
sql>select cast('2004-05-01 23:00' as timestamp)+cast('1:00:00' as time);
!types timestamp(7,0) (timestamp) and tinyint(8,0) (bte) are not equal
select cast('2004-05-01 23:00' as timestamp)+3600;
+----------------------------+
| sql_add_single_value |
+============================+
| 2004-05-02 [000000000000](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=000000000000) |
+----------------------------+
## Comment 12376
Date: 2009-09-17 21:20:23 +0200
From: @njnes
adding time to timestamp isn't semantically a valid operation. You should add intervals to time/timestamps instead
## Comment 12377
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2854884 at http://sourceforge.net/support/tracker.php?aid=2854884
| SQL: adding timestamp + time is impossible | https://api.github.com/repos/MonetDB/MonetDB/issues/2186/comments | 0 | 2020-11-30T10:26:32Z | 2024-06-28T13:22:42Z | https://github.com/MonetDB/MonetDB/issues/2186 | 753,352,907 | 2,186 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-09 12:19:05 +0200
From: @skinkie
To: @njnes
Version: -- development
Last updated: 2009-09-09 12:27:12 +0200
## Comment 12371
Date: 2009-09-09 00:19:05 +0200
From: @skinkie
Using:
create table bla(test integer, test2 integer);
COPY 1 RECORDS INTO bla from stdin using delimiters ',', '\n';
1,
I would expect that the it would insert:
1, NULL
Instead:
!SQLException:sql:value '' from line 1 field 2 not inserted, expecting type int
!SQLException:importTable:failed to import table
## Comment 12372
Date: 2009-09-09 00:27:12 +0200
From: @mlkersten
The response is correct, because you have not overruled the NULL representation.
The following might work.
COPY 1 RECORDS INTO bla from stdin using delimiters ',', '\n' NULL as '';
## Comment 12373
Date: 2009-09-09 09:42:46 +0200
From: @drstmane
Confirming Martin:
sql>create table bla(test integer, test2 integer);
Operation successful
sql>COPY 1 RECORDS INTO bla from stdin using delimiters ',', '\n';
more>1,
!SQLException:sql:value '' while parsing '' from line 0 field 1 not inserted, expecting type int
!SQLException:importTable:failed to import table
0 tuples
sql>COPY 1 RECORDS INTO bla from stdin using delimiters ',', '\n' NULL as '';
more>1,
Rows affected 1
sql>select * from bla;
+-------+-------+
| test | test2 |
+=======+=======+
| 1 | null |
+-------+-------+
1 tuple
sql>
## Comment 12374
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2854722 at http://sourceforge.net/support/tracker.php?aid=2854722
| SQL: Adjacent delimiters with copy into, do not create NULL | https://api.github.com/repos/MonetDB/MonetDB/issues/2185/comments | 0 | 2020-11-30T10:26:28Z | 2024-06-28T13:22:41Z | https://github.com/MonetDB/MonetDB/issues/2185 | 753,352,863 | 2,185 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-08 11:50:40 +0200
From: @grobian
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @romulogoncalves, @njnes
Last updated: 2011-01-25 14:57:19 +0100
## Comment 12367
Date: 2009-09-08 11:50:40 +0200
From: @grobian
(pegasus:sql/src/server) fabian% $INSTALL_DIR/bin/mclient -dthree
Welcome to mclient, the MonetDB/SQL interactive terminal
Database: MonetDB v5.15.0, 'three'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>\d
TABLE sys._columns
TABLE sys._tables
TABLE sys.args
TABLE sys.auths
VIEW sys.columns
TABLE sys.connections
TABLE sys.db_user_info
TABLE sys.dependencies
TABLE sys.functions
TABLE sys.idxs
TABLE sys.keycolumns
TABLE sys.keys
TABLE sys.privileges
TABLE sys.schemas
TABLE sys.sequences
VIEW sys.tables
TABLE sys.triggers
TABLE sys.types
TABLE sys.user_role
VIEW sys.users
sql>select * from sequences;
0 tuples
sql>drop sequence seq_4155
more>;
DROP SEQUENCE: no such sequence 'seq_4155'
sql>create table x (id serial);
operation successful
sql>\d x
CREATE TABLE "sys"."x" (
"id" int NOT NULL DEFAULT next value for "sys"."seq_4162",
CONSTRAINT "x_id_pkey" PRIMARY KEY ("id")
);
sql>drop table x;
operation successful
sql>drop sequence seq_4162;
MAPI = monetdb@localhost:50000
ACTION= read_line
QUERY = drop sequence seq_4162;
ERROR = !Connection terminated
(pegasus:sql/src/server) fabian% tail -n20 $INSTALL_DIR/var/log/MonetDB/merovingian.log
2009-09-08 11:48:35 MSG merovingian[17074]: database 'three' has crashed after start on 2009-09-08 11:46:46, attempting restart, up min/avg/max: 4s/17h/2w, crash average: 1.00 0.10 0.13 (32-28=4)
2009-09-08 11:48:35 MSG three[25888]: arguments: /export/scratch/stripe/fabian/monetdb/current/program-i86pc/bin/mserver5 --config=/export/scratch/stripe/fabian/monetdb/current/program-i86pc/etc/monetdb5.conf --dbname=three --dbinit=include sql; --set monet_daemon=yes --set mapi_open=false --set mapi_autosense=true --set mapi_port=50001 --set monet_vault_key=/export/scratch/stripe/fabian/monetdb/current/program-i86pc/var/MonetDB5/dbfarm/three/.vaultkey --set gdk_nr_threads=3 --set replication_master=true
2009-09-08 11:48:35 MSG three[25888]: MonetDB server v5.15.0, based on kernel v1.33.0
2009-09-08 11:48:35 MSG three[25888]: Serving database 'three', using 3 threads
2009-09-08 11:48:35 MSG three[25888]: Compiled for x86_64-pc-solaris2.11/64bit with 64bit OIDs dynamically linked
2009-09-08 11:48:35 MSG three[25888]: Detected 2.000 GiB (2147020800 bytes) main-memory.
2009-09-08 11:48:35 MSG three[25888]: Copyright (c) 1993-July 2008 CWI.
2009-09-08 11:48:35 MSG three[25888]: Copyright (c) August 2008-2009 MonetDB B.V., all rights reserved
2009-09-08 11:48:35 MSG three[25888]: Visit http://monetdb.cwi.nl/ for further information
2009-09-08 11:48:35 MSG three[25888]: Listening for connection requests on mapi:monetdb://127.0.0.1:50002/
2009-09-08 11:48:35 MSG three[25888]: MonetDB/SQL module v2.33.0 loaded
2009-09-08 11:48:35 MSG merovingian[17074]: proxying client localhost:48792 for database 'three' to mapi:monetdb://127.0.0.1:50002/three
2009-09-08 11:49:29 ERR three[25888]: Assertion failed: rid != oid_nil, file /export/scratch/mirror/fabian/monetdb/current/sql/src/storage/store.mx, line 3004, function sys_drop_sequence
2009-09-08 11:49:29 MSG merovingian[17074]: database 'three' (25888) was killed by signal 6
2009-09-08 11:49:29 MSG merovingian[17074]: client localhost:48792 has disconnected from proxy
## Comment 12368
Date: 2009-09-17 21:22:03 +0200
From: @njnes
this bug is related to dependency (table,sequence (stored in schema)) problems, ie reassigning to romulo
## Comment 12369
Date: 2009-09-17 21:25:22 +0200
From: @grobian
this looks to me like the in-memory catalog structure (the structs pointer thing) is not updated upon the cascade, and hence the drop thinks the object still exists, resulting in the crash when it tries to delete the object that has already been removed
## Comment 12370
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2854195 at http://sourceforge.net/support/tracker.php?aid=2854195
## Comment 14168
Date: 2010-07-01 17:09:05 +0200
From: @romulogoncalves
(In reply to comment 1)
> this bug is related to dependency (table,sequence (stored in schema)) problems,
> ie reassigning to romulo
Well, it is not related to the dependencies, at least not to my dependencies check. The dependencies over sequences is older than the dependency check for other database objects implemented by me.
If you drop column which is a serial, i.e., it has a sequence associated, should this sequence be dropped once you drop the column, or should the sequence be kept and to drop it you need to call DROP SEQUENCE?
Because if it is drop together with the column the following code is incomplete:
if (col->def && (seq_pos = strstr(col->def, next_value_for))) {
sql_sequence * seq = NULL;
char *seq_name = _strdup(seq_pos + (strlen(next_value_for) - strlen("seq_")));
seq_name[strlen(seq_name)-1] = '\0';
seq = find_sql_sequence(syss, seq_name);
if (seq)
INCOMPLETE --> sys_drop_sequence(tr, seq, drop_action);
_DELETE(seq_name);
}
You need to remove as you do when you DROP SEQUENCE, i.e., from the catalog loaded in MEM. Therefore, the two lines need to be added:
seq->base.wtime = syss->base.wtime = tr->wtime = tr->stime;
cs_del(&syss->seqs, n, seq->base.flag);
If you want to keep the sequence and only remove the dependency with the dropped column, sys_drop_sequence should not be called.
To fix the bug and close it I need to know which procedure should we take when we drop a serial column. Correct if my observation is wrong.
## Comment 14255
Date: 2010-07-07 15:42:37 +0200
From: @romulogoncalves
The bug is fixed with the check in:
Changeset: [6096dc911f06](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=6096dc911f06) for MonetDB
URL: [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=6096dc911f06](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=6096dc911f06)
Is there a test to cover the BUG? If not could you add it please?
## Comment 15259
Date: 2010-12-03 21:32:53 +0100
From: @njnes
missing test...
## Comment 15421
Date: 2011-01-25 14:56:51 +0100
From: @grobian
Changeset [e9b6e865219a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e9b6e865219a) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=e9b6e865219a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=e9b6e865219a)
Changeset description:
Add test for bug #2184
## Comment 15422
Date: 2011-01-25 14:57:19 +0100
From: @grobian
test added
| SQL: crash on drop sequence after drop table | https://api.github.com/repos/MonetDB/MonetDB/issues/2184/comments | 0 | 2020-11-30T10:26:25Z | 2024-06-27T11:41:34Z | https://github.com/MonetDB/MonetDB/issues/2184 | 753,352,827 | 2,184 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-07 12:03:57 +0200
From: @skinkie
To: @njnes
Version: -- development
Last updated: 2009-09-17 10:22:37 +0200
## Comment 12361
Date: 2009-09-07 12:03:57 +0200
From: @skinkie
Should this be prevented? Or is a cast a cast?
>select cast(power(2,63) as bigint);
+----------------------+
| power_single_value |
+======================+
| null |
+----------------------+
1 tuple
sql>select power(2,63);
+------------------------+
| power_single_value |
+========================+
| 9.2233720368547758e+18 |
+------------------------+
1 tuple
## Comment 12362
Date: 2009-09-07 22:58:57 +0200
From: @njnes
sql should give an error (ie overflow)
## Comment 12363
Date: 2009-09-08 11:14:29 +0200
From: @grobian
div_by_zero.SF-987304 no longer returns an error but returns NULL instead, can that be related to this bug?
## Comment 12364
Date: 2009-09-08 11:41:21 +0200
From: @drstmane
The div_by_zero is probably not related "code-wise", but possibly "conceptually":
Errors like division by zero (in batcalc.mx) or overflows (with arthmetics in batcalc.mx or casts in batcast.mx) are partly reported as errors, partly handled by resulting in nil/null, partly simply ignored.
For an efficient SQL implementation/support, we'd probably at least need two variants of each "bulk" ("multiplex") aritmetic and cast operation in batcalc.mx and batcast.mx: one that "silently" turns errors and overflows into nil/null, and one the raises an error/exception.
The SQL compiler then needs to decide which semantics it requires and call the respective primitive ...
## Comment 12365
Date: 2009-09-17 22:22:37 +0200
From: @njnes
fixed, we now test for nil better.. Added test in overflow.SF-2853458.sql
## Comment 12366
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2853458 at http://sourceforge.net/support/tracker.php?aid=2853458
| SQL: cast exceeding maximum value, returns null | https://api.github.com/repos/MonetDB/MonetDB/issues/2183/comments | 0 | 2020-11-30T10:26:21Z | 2024-06-27T11:41:33Z | https://github.com/MonetDB/MonetDB/issues/2183 | 753,352,773 | 2,183 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-07 11:18:03 +0200
From: Wouter Alink <<alink>>
To: Jan Rittinger <<jan.rittinger>>
Version: -- development
Last updated: 2009-09-10 11:17:30 +0200
## Comment 12356
Date: 2009-09-07 11:18:03 +0200
From: Wouter Alink <<alink>>
The following query yields duplicates (although I specified to return distinct-values):
let $x := ("aap","beer")
let $y := <a><b>beer</b><b>noot</b></a>//b
for $i in distinct-values(($x,$y)) return $i
It returns:
"beer",
"noot",
"aap",
"beer"
I assume this is a bug.
Note that the following query works fine:
let $x := ("aap","beer")
let $y := ("beer","noot")
for $i in distinct-values(($x,$y)) return $i
It returns:
"aap",
"beer",
"noot"
I used:
$ bin/Mserver --dbinit="module(pathfinder);" --set mapi_port=50333
MonetDB Server v4.30.1
based on GDK v1.30.1
Copyright (c) 1993-July 2008, CWI. All rights reserved.
Copyright (c) August 2008-2009, MonetDB B.V.. All rights reserved.
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs; dynamically linked.
Visit http://monetdb.cwi.nl/ for further information.
PF/Tijah module v0.13.0 loaded. http://dbappl.cs.utwente.nl/pftijah
MonetDB/XQuery module v0.28.3 loaded (default back-end is 'algebra')
XRPC administrative console at http://127.0.0.1:50334/admin
MonetDB>
## Comment 12357
Date: 2009-09-07 23:32:21 +0200
From: Jan Rittinger <<jan.rittinger>>
In the first example there is a mix of strings and items of type untypedAtomics. We did not compare items of different types until now.
The XQuery Specification however says: 'Values of type xs:untypedAtomic are compared as if they were of type xs:string.' Note: the result is of type untypedAtomic if the input was untypedAtomic.
I extended the implementation with a special case for a mix of strings and untypedAtomics which casts items of type untypedAtomic to string before the duplicate elimination. In comparison to the specification we however loose the untypedAtomic information during duplicate elimination (even if no string matches an untypedAtomic).
Wouter can you check whether this fixes your problem.
## Comment 12358
Date: 2009-09-08 00:19:23 +0200
From: Wouter Alink <<alink>>
Thanks Jan. I just added a test. I used an older MonetDB version to generate the output, and corrected it manually. The test might therefore use a different order for the results which should be approved.
## Comment 12359
Date: 2009-09-10 11:17:27 +0200
From: Wouter Alink <<alink>>
testing indicates that this bug has been fixed
## Comment 12360
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2853441 at http://sourceforge.net/support/tracker.php?aid=2853441
| PF: distinct values on two sequences | https://api.github.com/repos/MonetDB/MonetDB/issues/2182/comments | 0 | 2020-11-30T10:26:18Z | 2024-06-27T11:41:32Z | https://github.com/MonetDB/MonetDB/issues/2182 | 753,352,728 | 2,182 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-06 03:34:02 +0200
From: @yzchang
To: @yzchang
Version: -- development
Last updated: 2009-09-18 01:49:05 +0200
## Comment 12348
Date: 2009-09-06 15:34:02 +0200
From: @yzchang
On my CWI desktop machine, using latest stable branch (monetdb-install.sh --enable-xquery --enable-optimize --nightly=stable --prefix='...'), using the algebra front-end, all updates (insert, delete, rename, replace) gives error, *after* updates have been applied to the document:
$ Mserver --dbinit='module(pathfinder);' --set xquery_cacheMB=0 --dbname='blah'
MonetDB Server v4.32.0
based on GDK v1.32.0
Copyright (c) 1993-July 2008, CWI. All rights reserved.
Copyright (c) August 2008-2009, MonetDB B.V.. All rights reserved.
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs; dynamically linked.
Visit http://monetdb.cwi.nl/ for further information.
PF/Tijah module v0.13.0 loaded. http://dbappl.cs.utwente.nl/pftijah
MonetDB/XQuery module v0.28.3 loaded (default back-end is 'algebra')
XRPC administrative console at http://127.0.0.1:50001/admin
MonetDB>shred_doc("/ufs/zhang/tmp/hello.xml" , "hello.xml", "hello", 10LL);
shredding with holes: pagesize=16384, fillsize=14746
Elapsed time = 024ms 888us [001ms 382us/node]
Shredded 1 XML document (hello.xml), total time after commit=0.038s
MonetDB>
$ mclient -lx -t
xquery>doc("hello.xml")
more><>
<?xml version="1.0" encoding="utf-8"?>
<hello x="y" lang="en">
<foo>text of node foo<bar>text of node bar</bar></foo>
<!-- a comment node -->
<?pi action=beep?>
<world/>
<path>\asfda\asdfasf\asdf</path>
</hello>
Trans 3.270 msec
Shred 1.991 msec
Query 8.308 msec
Print 0.237 msec
Timer 16.923 msec
xquery>do insert <a/> into doc("hello.xml")/hello
more><>
MAPI = monetdb@localhost:50000
QUERY = do insert <a/> into doc("hello.xml")/hello
ERROR = !ERROR: interpret: no matching MIL operator to '-(lng, void)'.
!MAYBE YOU MEAN:
! -(dbl) : dbl
! -(lng) : lng
! -(flt) : flt
! -(wrd) : wrd
! -(int) : int
! -(sht) : sht
! -(bte) : bte
! -(chr) : chr
! -(lng, dbl) : dbl
! -(lng, flt) : dbl
! -(lng, lng) : lng
! -(lng, wrd) : lng
! -(lng, int) : lng
! -(lng, sht) : lng
! -(lng, bte) : lng
! -(lng, chr) : lng
! -(dbl, dbl) : dbl
! -(dbl, flt) : dbl
! -(dbl, lng) : dbl
! -(dbl, wrd) : dbl
! -(dbl, int) : dbl
! -(dbl, sht) : dbl
! -(dbl, bte) : dbl
! -(dbl, chr) : dbl
! -(flt, dbl) : dbl
! -(flt, flt) : flt
! -(flt, lng) : flt
! -(flt, wrd) : flt
! -(flt, int) : flt
! -(flt, sht) : flt
! -(flt, bte) : flt
! -(flt, chr) : flt
! -(wrd, dbl) : dbl
! -(wrd, flt) : flt
! -(wrd, lng) : lng
! -(wrd, wrd) : wrd
! -(wrd, int) : wrd
! -(wrd, sht) : wrd
! -(wrd, bte) : wrd
! -(wrd, chr) : wrd
! -(int, dbl) : dbl
! -(int, flt) : flt
! -(int, lng) : lng
! -(int, wrd) : wrd
! -(int, int) : int
! -(int, sht) : int
! -(int, bte) : int
! -(int, chr) : int
! -(sht, dbl) : dbl
! -(sht, flt) : flt
! -(sht, lng) : lng
! -(sht, wrd) : wrd
! -(sht, int) : int
! -(sht, sht) : sht
! -(sht, bte) : sht
! -(sht, chr) : sht
! -(bte, dbl) : dbl
! -(bte, flt) : flt
! -(bte, lng) : lng
! -(bte, wrd) : wrd
! -(bte, int) : int
! -(bte, sht) : sht
! -(bte, chr) : bte
! -(bte, bte) : bte
! -(chr, dbl) : dbl
! -(chr, flt) : flt
! -(chr, lng) : lng
! -(chr, wrd) : wrd
! -(chr, int) : int
! -(chr, sht) : sht
! -(chr, bte) : bte
! -(chr, chr) : chr
Timer 63.229 msec
xquery>do delete doc("hello.xml")//world
more><>
MAPI = monetdb@localhost:50000
QUERY = do delete doc("hello.xml")//world
ERROR = !ERROR: interpret: no matching MIL operator to '-(lng, void)'.
!MAYBE YOU MEAN:
! -(dbl) : dbl
! -(lng) : lng
! -(flt) : flt
! -(wrd) : wrd
! -(int) : int
! -(sht) : sht
! -(bte) : bte
! -(chr) : chr
! -(lng, dbl) : dbl
! -(lng, flt) : dbl
! -(lng, lng) : lng
! -(lng, wrd) : lng
! -(lng, int) : lng
! -(lng, sht) : lng
! -(lng, bte) : lng
! -(lng, chr) : lng
! -(dbl, dbl) : dbl
! -(dbl, flt) : dbl
! -(dbl, lng) : dbl
! -(dbl, wrd) : dbl
! -(dbl, int) : dbl
! -(dbl, sht) : dbl
! -(dbl, bte) : dbl
! -(dbl, chr) : dbl
! -(flt, dbl) : dbl
! -(flt, flt) : flt
! -(flt, lng) : flt
! -(flt, wrd) : flt
! -(flt, int) : flt
! -(flt, sht) : flt
! -(flt, bte) : flt
! -(flt, chr) : flt
! -(wrd, dbl) : dbl
! -(wrd, flt) : flt
! -(wrd, lng) : lng
! -(wrd, wrd) : wrd
! -(wrd, int) : wrd
! -(wrd, sht) : wrd
! -(wrd, bte) : wrd
! -(wrd, chr) : wrd
! -(int, dbl) : dbl
! -(int, flt) : flt
! -(int, lng) : lng
! -(int, wrd) : wrd
! -(int, int) : int
! -(int, sht) : int
! -(int, bte) : int
! -(int, chr) : int
! -(sht, dbl) : dbl
! -(sht, flt) : flt
! -(sht, lng) : lng
! -(sht, wrd) : wrd
! -(sht, int) : int
! -(sht, sht) : sht
! -(sht, bte) : sht
! -(sht, chr) : sht
! -(bte, dbl) : dbl
! -(bte, flt) : flt
! -(bte, lng) : lng
! -(bte, wrd) : wrd
! -(bte, int) : int
! -(bte, sht) : sht
! -(bte, chr) : bte
! -(bte, bte) : bte
! -(chr, dbl) : dbl
! -(chr, flt) : flt
! -(chr, lng) : lng
! -(chr, wrd) : wrd
! -(chr, int) : int
! -(chr, sht) : sht
! -(chr, bte) : bte
! -(chr, chr) : chr
Timer 39.003 msec
xquery>doc("hello.xml")
more><>
<?xml version="1.0" encoding="utf-8"?>
<hello x="y" lang="en">
<foo>text of node foo<bar>text of node bar</bar></foo>
<!-- a comment node -->
<?pi action=beep?>
^^^^^^^^^^^^^^^^^
'<world/>' has been deleted
<path>\asfda\asdfasf\asdf</path>
<a/></hello>
^^^^^
previous insertion was applied
Trans 3.279 msec
Shred 1.957 msec
Query 8.039 msec
Print 0.247 msec
Timer 16.519 msec
xquery>do rename doc("hello.xml")//foo into "foo2"
more><>
MAPI = monetdb@localhost:50000
QUERY = do rename doc("hello.xml")//foo into "foo2"
ERROR = !ERROR: interpret: no matching MIL operator to '-(lng, void)'.
!MAYBE YOU MEAN:
! -(dbl) : dbl
! -(lng) : lng
! -(flt) : flt
! -(wrd) : wrd
! -(int) : int
! -(sht) : sht
! -(bte) : bte
! -(chr) : chr
! -(lng, dbl) : dbl
! -(lng, flt) : dbl
! -(lng, lng) : lng
! -(lng, wrd) : lng
! -(lng, int) : lng
! -(lng, sht) : lng
! -(lng, bte) : lng
! -(lng, chr) : lng
! -(dbl, dbl) : dbl
! -(dbl, flt) : dbl
! -(dbl, lng) : dbl
! -(dbl, wrd) : dbl
! -(dbl, int) : dbl
! -(dbl, sht) : dbl
! -(dbl, bte) : dbl
! -(dbl, chr) : dbl
! -(flt, dbl) : dbl
! -(flt, flt) : flt
! -(flt, lng) : flt
! -(flt, wrd) : flt
! -(flt, int) : flt
! -(flt, sht) : flt
! -(flt, bte) : flt
! -(flt, chr) : flt
! -(wrd, dbl) : dbl
! -(wrd, flt) : flt
! -(wrd, lng) : lng
! -(wrd, wrd) : wrd
! -(wrd, int) : wrd
! -(wrd, sht) : wrd
! -(wrd, bte) : wrd
! -(wrd, chr) : wrd
! -(int, dbl) : dbl
! -(int, flt) : flt
! -(int, lng) : lng
! -(int, wrd) : wrd
! -(int, int) : int
! -(int, sht) : int
! -(int, bte) : int
! -(int, chr) : int
! -(sht, dbl) : dbl
! -(sht, flt) : flt
! -(sht, lng) : lng
! -(sht, wrd) : wrd
! -(sht, int) : int
! -(sht, sht) : sht
! -(sht, bte) : sht
! -(sht, chr) : sht
! -(bte, dbl) : dbl
! -(bte, flt) : flt
! -(bte, lng) : lng
! -(bte, wrd) : wrd
! -(bte, int) : int
! -(bte, sht) : sht
! -(bte, chr) : bte
! -(bte, bte) : bte
! -(chr, dbl) : dbl
! -(chr, flt) : flt
! -(chr, lng) : lng
! -(chr, wrd) : wrd
! -(chr, int) : int
! -(chr, sht) : sht
! -(chr, bte) : bte
! -(chr, chr) : chr
Timer 76.213 msec
xquery>doc("hello.xml")/hello
more><>
<hello x="y" lang="en">
<foo2>text of node foo<bar>text of node bar</bar></foo2>
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
rename has been done
<!-- a comment node -->
<?pi action=beep?>
<path>\asfda\asdfasf\asdf</path>
<a/></hello>
Trans 4.082 msec
Shred 1.931 msec
Query 11.054 msec
Print 0.261 msec
Timer 20.973 msec
xquery>do replace doc("hello.xml")//bar with <noot/>
more><>
MAPI = monetdb@localhost:50000
QUERY = do replace doc("hello.xml")//bar with <noot/>
ERROR = !ERROR: interpret: no matching MIL operator to '-(lng, void)'.
!MAYBE YOU MEAN:
! -(dbl) : dbl
! -(lng) : lng
! -(flt) : flt
! -(wrd) : wrd
! -(int) : int
! -(sht) : sht
! -(bte) : bte
! -(chr) : chr
! -(lng, dbl) : dbl
! -(lng, flt) : dbl
! -(lng, lng) : lng
! -(lng, wrd) : lng
! -(lng, int) : lng
! -(lng, sht) : lng
! -(lng, bte) : lng
! -(lng, chr) : lng
! -(dbl, dbl) : dbl
! -(dbl, flt) : dbl
! -(dbl, lng) : dbl
! -(dbl, wrd) : dbl
! -(dbl, int) : dbl
! -(dbl, sht) : dbl
! -(dbl, bte) : dbl
! -(dbl, chr) : dbl
! -(flt, dbl) : dbl
! -(flt, flt) : flt
! -(flt, lng) : flt
! -(flt, wrd) : flt
! -(flt, int) : flt
! -(flt, sht) : flt
! -(flt, bte) : flt
! -(flt, chr) : flt
! -(wrd, dbl) : dbl
! -(wrd, flt) : flt
! -(wrd, lng) : lng
! -(wrd, wrd) : wrd
! -(wrd, int) : wrd
! -(wrd, sht) : wrd
! -(wrd, bte) : wrd
! -(wrd, chr) : wrd
! -(int, dbl) : dbl
! -(int, flt) : flt
! -(int, lng) : lng
! -(int, wrd) : wrd
! -(int, int) : int
! -(int, sht) : int
! -(int, bte) : int
! -(int, chr) : int
! -(sht, dbl) : dbl
! -(sht, flt) : flt
! -(sht, lng) : lng
! -(sht, wrd) : wrd
! -(sht, int) : int
! -(sht, sht) : sht
! -(sht, bte) : sht
! -(sht, chr) : sht
! -(bte, dbl) : dbl
! -(bte, flt) : flt
! -(bte, lng) : lng
! -(bte, wrd) : wrd
! -(bte, int) : int
! -(bte, sht) : sht
! -(bte, chr) : bte
! -(bte, bte) : bte
! -(chr, dbl) : dbl
! -(chr, flt) : flt
! -(chr, lng) : lng
! -(chr, wrd) : wrd
! -(chr, int) : int
! -(chr, sht) : sht
! -(chr, bte) : bte
! -(chr, chr) : chr
Timer 55.182 msec
xquery>doc("hello.xml")
more><>
<?xml version="1.0" encoding="utf-8"?>
<hello x="y" lang="en">
<foo2>text of node foo<noot/></foo2>
^^^^^^^^^
replace has been done
<!-- a comment node -->
<?pi action=beep?>
<path>\asfda\asdfasf\asdf</path>
<a/></hello>
Trans 4.688 msec
Shred 1.965 msec
Query 8.381 msec
Print 0.244 msec
Timer 17.948 msec
Could someone please check if this could be reproduced?
FYI, the updates work fine with the MPS front-end (for this mclient session, the doc. has been deleted and re-shredded):
$ mclient -lx -t -g
xquery>doc("hello.xml")
more><>
<?xml version="1.0" encoding="utf-8"?>
<hello x="y" lang="en">
<foo>text of node foo<bar>text of node bar</bar></foo>
<!-- a comment node -->
<?pi action=beep?>
<world/>
<path>\asfda\asdfasf\asdf</path>
</hello>
Trans 9.883 msec
Shred 7.381 msec
Query 2.549 msec
Print 0.832 msec
Timer 25.371 msec
xquery>do insert <a/> into doc("hello.xml")/hello
more><>
Trans 11.521 msec
Shred 5.279 msec
Query 16.776 msec
Update 18.536 msec
Timer 62.439 msec
xquery>doc("hello.xml")
more><>
<?xml version="1.0" encoding="utf-8"?>
<hello x="y" lang="en">
<foo>text of node foo<bar>text of node bar</bar></foo>
<!-- a comment node -->
<?pi action=beep?>
<world/>
<path>\asfda\asdfasf\asdf</path>
<a/></hello>
Trans 4.615 msec
Shred 5.154 msec
Query 2.556 msec
Print 0.819 msec
Timer 16.668 msec
xquery>do delete doc("hello.xml")//world
more><>
Trans 5.566 msec
Shred 5.344 msec
Query 5.401 msec
Update 12.175 msec
Timer 35.438 msec
xquery>doc("hello.xml")
more><>
<?xml version="1.0" encoding="utf-8"?>
<hello x="y" lang="en">
<foo>text of node foo<bar>text of node bar</bar></foo>
<!-- a comment node -->
<?pi action=beep?>
<path>\asfda\asdfasf\asdf</path>
<a/></hello>
Trans 4.419 msec
Shred 5.135 msec
Query 2.547 msec
Print 0.826 msec
Timer 16.462 msec
xquery>do rename doc("hello.xml")//foo into 'foo2'
more><>
Trans 5.700 msec
Shred 5.329 msec
Query 5.074 msec
Update 13.145 msec
Timer 36.176 msec
xquery>doc("hello.xml")
more><>
<?xml version="1.0" encoding="utf-8"?>
<hello x="y" lang="en">
<foo2>text of node foo<bar>text of node bar</bar></foo2>
<!-- a comment node -->
<?pi action=beep?>
<path>\asfda\asdfasf\asdf</path>
<a/></hello>
Trans 4.407 msec
Shred 5.168 msec
Query 2.566 msec
Print 0.860 msec
Timer 16.514 msec
xquery>do replace doc("hello.xml")//bar with <noot/>
more><>
Trans 7.433 msec
Shred 5.321 msec
Query 8.120 msec
Update 15.398 msec
Timer 47.425 msec
xquery>doc("hello.xml")
more><>
<?xml version="1.0" encoding="utf-8"?>
<hello x="y" lang="en">
<foo2>text of node foo<noot/></foo2>
<!-- a comment node -->
<?pi action=beep?>
<path>\asfda\asdfasf\asdf</path>
<a/></hello>
Trans 4.410 msec
Shred 5.160 msec
Query 2.584 msec
Print 0.855 msec
Timer 16.553 msec
xquery>\q
Regards,
Jennie
## Comment 12349
Date: 2009-09-07 14:30:17 +0200
From: @sjoerdmullender
The problem has to do with the -t flag given to mclient and seems to be that the value time_print is not initialized in the generated MIL code when doing an updating query.
## Comment 12350
Date: 2009-09-07 16:28:07 +0200
From: @sjoerdmullender
Assigning to Jan because the problem has to do with mil_dce.c.
When using mclient -t or pf -T (i.e. when requesting timing information) in combination with an updating query, the assignments to time_print are eliminated, even though time_print is used. The two assignments to time_print are the initial assignment of 0 when the variable is declared, and an assignment near the call to UpdateTape.
Jan, just compare the output of
echo 'do insert <a/> into doc("hello.xml")/hello' | pf -T
end
echo 'do insert <a/> into doc("hello.xml")/hello' | pf -T -e0
and look for time_print.
## Comment 12351
Date: 2009-09-07 22:52:12 +0200
From: Jan Rittinger <<jan.rittinger>>
Sjoerd, thanks for the analysis -- I checked in a fix.
Jennie could you please check whether everything works as expected.
## Comment 12352
Date: 2009-09-10 11:44:33 +0200
From: @sjoerdmullender
No word from Jennie yet, but I added a test (tests/BugTracker update-with-timing.SF-2852928.XQUERY) and as far as I am concerned, the bug is fixed.
So, Jennie, close if you're happy.
## Comment 12353
Date: 2009-09-10 20:58:50 +0200
From: @yzchang
Sorry, I didn't have time to check this. I will check it and close the bug next week.
## Comment 12354
Date: 2009-09-18 13:47:18 +0200
From: @yzchang
checked and it works, so I'm very happy:) close the bug, test is already added by Sjoerd.
## Comment 12355
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2852928 at http://sourceforge.net/support/tracker.php?aid=2852928
| PF: updates broken using algebra front-end? | https://api.github.com/repos/MonetDB/MonetDB/issues/2181/comments | 0 | 2020-11-30T10:26:14Z | 2024-06-27T11:41:31Z | https://github.com/MonetDB/MonetDB/issues/2181 | 753,352,691 | 2,181 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-06 03:24:23 +0200
From: @skinkie
To: @mlkersten
Version: -- development
Last updated: 2009-11-15 03:39:05 +0100
## Comment 12341
Date: 2009-09-06 03:24:23 +0200
From: @skinkie
Entering on the MonetDB server:
user.main();
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7faf8fe906f0 (LWP 15035)]
0x00007faf8f99ca60 in runMALsequence (cntxt=Cannot access memory at address 0x7fff7d7db5a8
) at ../../../src/mal/mal_interpreter.mx:415
415 {
(gdb) bt
0 0x00007faf8f99ca60 in runMALsequence (cntxt=Cannot access memory at address 0x7fff7d7db5a8
) at ../../../src/mal/mal_interpreter.mx:415
1 0x00007faf8f99bf06 in runMAL (cntxt=0x605560, mb=0x15960a8, startpc=1, mbcaller=0x15960a8, env=0x7fff7d7e0530, pcicaller=0x1bf9698)
at ../../../src/mal/mal_interpreter.mx:308
2 0x00007faf8f9a9972 in runMALsequence (cntxt=0x605560, mb=0x15960a8, startpc=1, stoppc=0, stk=0x7fff7d7e0530, env=0x7fff7d7e3070, pcicaller=0x1bf9698)
at ../../../src/mal/mal_interpreter.mx:1770
3 0x00007faf8f99bf06 in runMAL (cntxt=0x605560, mb=0x15960a8, startpc=1, mbcaller=0x15960a8, env=0x7fff7d7e3070, pcicaller=0x1bf9698)
at ../../../src/mal/mal_interpreter.mx:308
etc.
I found this one by pasting in a function that didn't work. And reduced it actually executing the function.
## Comment 12342
Date: 2009-09-06 08:05:06 +0200
From: @mlkersten
This is the general stack and does not give any clue on what object it is attempting to access.
Also it seems a glitch depending on an unspecified sequence of actions.
The erroneous function might give an indication if the parser/type resolver,etc.... left something
unmarked.
The bug is archived.
## Comment 12343
Date: 2009-09-06 08:54:19 +0200
From: @drstmane
The segfault is easily reproducible by simply starting mserver5 and then calling `user.main();` on the MAL prompt.
It also occurs with the Aug2009 release.
While calling user.main(); interactively / explicitly is probably not intended, we should at least try to avoid the segfault in this case.
looking at the complete trace, the problem might be related to a stack overflow due to a too deep recursion:
CVS HEAD:
$ gdb mserver5
GNU gdb Fedora (6.8-32.fc10)
Copyright (C) 2008 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-redhat-linux-gnu"...
(gdb) r
Starting program: /net/rig.ins.cwi.nl/export/scratch0/manegold/Monet/Testing/Current/prefix.--enable-strict_--disable-optimize_--enable-debug_--enable-assert/bin/mserver5
[Thread debugging using libthread_db enabled]
[New Thread 0x7ffff67d37b0 (LWP 3966)]
[New Thread 0x7fffec2ed950 (LWP 4019)]
MonetDB server v5.15.0, based on kernel v1.33.0
Serving database 'demo', using 4 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs dynamically linked
Detected 7.753 GiB (8324386816 bytes) main-memory.
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2009 MonetDB B.V., all rights reserved
Visit http://monetdb.cwi.nl/ for further information
warning: please don't forget to set your vault key!
(see /ufs/manegold/_/scratch0/Monet/Testing/Current/prefix.--enable-strict_--disable-optimize_--enable-debug_--enable-assert/etc/monetdb5.conf)
[New Thread 0x7fffe3229950 (LWP 4020)]
Listening for connection requests on mapi:monetdb://127.0.0.1:50000/
>user.main();
Program received signal SIGSEGV, Segmentation fault.
0x00007ffff7ced3c5 in runMALsequence (cntxt=Cannot access memory at address 0x7fffff3fdf28
) at /ufs/manegold/_/scratch0/Monet/Testing/Current/source/MonetDB5/src/mal/mal_interpreter.mx:415
415 {
Missing separate debuginfos, use: debuginfo-install bzip2-libs-1.0.5-3.fc10.x86_64 e2fsprogs-libs-1.41.4-6.fc10.x86_64 glibc-2.9-3.x86_64 keyutils-libs-1.2-3.fc9.x86_64 krb5-libs-1.6.3-18.fc10.x86_64 libselinux-2.0.78-1.fc10.x86_64 ncurses-libs-5.6-20.20080927.fc10.x86_64 openssl-0.9.8g-14.fc10.x86_64 pcre-7.8-1.fc10.x86_64 readline-5.2-13.fc9.x86_64 zlib-1.2.3-18.fc9.x86_64
(gdb) bt
0 0x00007ffff7ced3c5 in runMALsequence (cntxt=Cannot access memory at address 0x7fffff3fdf28
) at /ufs/manegold/_/scratch0/Monet/Testing/Current/source/MonetDB5/src/mal/mal_interpreter.mx:415
1 0x00007ffff7cec8a8 in runMAL (cntxt=0x6048a0, mb=0x611388, startpc=1, mbcaller=0x611388, env=0x7fffff402f20, pcicaller=0x65d268) at /ufs/manegold/_/scratch0/Monet/Testing/Current/source/MonetDB5/src/mal/mal_interpreter.mx:308
2 0x00007ffff7cf98ff in runMALsequence (cntxt=0x6048a0, mb=0x611388, startpc=1, stoppc=0, stk=0x7fffff402f20, env=0x7fffff405ad0, pcicaller=0x65d268)
at /ufs/manegold/_/scratch0/Monet/Testing/Current/source/MonetDB5/src/mal/mal_interpreter.mx:1770
3 0x00007ffff7cec8a8 in runMAL (cntxt=0x6048a0, mb=0x611388, startpc=1, mbcaller=0x611388, env=0x7fffff405ad0, pcicaller=0x65d268) at /ufs/manegold/_/scratch0/Monet/Testing/Current/source/MonetDB5/src/mal/mal_interpreter.mx:308
[...]
2248 0x00007ffff7cf98ff in runMALsequence (cntxt=0x6048a0, mb=0x611388, startpc=1, stoppc=0, stk=0xd75818, env=0xd75818, pcicaller=0x0) at /ufs/manegold/_/scratch0/Monet/Testing/Current/source/MonetDB5/src/mal/mal_interpreter.mx:1770
2249 0x00007ffff7cec8a8 in runMAL (cntxt=0x6048a0, mb=0x611388, startpc=1, mbcaller=0x0, env=0xd75818, pcicaller=0x0) at /ufs/manegold/_/scratch0/Monet/Testing/Current/source/MonetDB5/src/mal/mal_interpreter.mx:308
2250 0x00007ffff7cebbff in MALengine (c=0x6048a0) at /ufs/manegold/_/scratch0/Monet/Testing/Current/source/MonetDB5/src/mal/mal_session.mx:681
2251 0x00007ffff7d30265 in runPhase (c=0x6048a0, phase=4) at /ufs/manegold/_/scratch0/Monet/Testing/Current/source/MonetDB5/src/mal/mal_scenario.mx:602
2252 0x00007ffff7d303f2 in runScenarioBody (c=0x6048a0) at /ufs/manegold/_/scratch0/Monet/Testing/Current/source/MonetDB5/src/mal/mal_scenario.mx:647
2253 0x00007ffff7d30589 in runScenario (c=0x6048a0) at /ufs/manegold/_/scratch0/Monet/Testing/Current/source/MonetDB5/src/mal/mal_scenario.mx:672
2254 0x00007ffff7ceb04a in MSserveClient (dummy=0x6048a0) at /ufs/manegold/_/scratch0/Monet/Testing/Current/source/MonetDB5/src/mal/mal_session.mx:496
2255 0x0000000000402fa5 in main (argc=1, av=0x7fffffffdda8) at /ufs/manegold/_/scratch0/Monet/Testing/Current/source/MonetDB5/src/tools/mserver5.mx:637
(gdb)
Aug2009:
$ gdb mserver5
GNU gdb Fedora (6.8-32.fc10)
Copyright (C) 2008 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-redhat-linux-gnu"...
(gdb) r
Starting program: /net/rig.ins.cwi.nl/export/scratch0/manegold/Monet/Testing/Stable/prefix.--enable-strict_--disable-optimize_--enable-debug_--enable-assert/bin/mserver5
[Thread debugging using libthread_db enabled]
[New Thread 0x7ffff67d37b0 (LWP 4283)]
[New Thread 0x7fffec2ed950 (LWP 4336)]
MonetDB server v5.14.0, based on kernel v1.32.0
Serving database 'demo', using 4 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs dynamically linked
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2009 MonetDB B.V., all rights reserved
Visit http://monetdb.cwi.nl/ for further information
warning: please don't forget to set your vault key!
(see /ufs/manegold/_/scratch0/Monet/Testing/Stable/prefix.--enable-strict_--disable-optimize_--enable-debug_--enable-assert/etc/monetdb5.conf)
[New Thread 0x7fffe33a6950 (LWP 4337)]
Listening for connection requests on mapi:monetdb://127.0.0.1:50000/
>user.main();
Program received signal SIGSEGV, Segmentation fault.
0x00007ffff7ced319 in runMALsequence (cntxt=Cannot access memory at address 0x7fffff3fecd8
) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB5/src/mal/mal_interpreter.mx:415
415 {
Missing separate debuginfos, use: debuginfo-install bzip2-libs-1.0.5-3.fc10.x86_64 e2fsprogs-libs-1.41.4-6.fc10.x86_64 glibc-2.9-3.x86_64 keyutils-libs-1.2-3.fc9.x86_64 krb5-libs-1.6.3-18.fc10.x86_64 libselinux-2.0.78-1.fc10.x86_64 ncurses-libs-5.6-20.20080927.fc10.x86_64 openssl-0.9.8g-14.fc10.x86_64 pcre-7.8-1.fc10.x86_64 readline-5.2-13.fc9.x86_64 zlib-1.2.3-18.fc9.x86_64
(gdb) bt
0 0x00007ffff7ced319 in runMALsequence (cntxt=Cannot access memory at address 0x7fffff3fecd8
) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB5/src/mal/mal_interpreter.mx:415
1 0x00007ffff7cec7fc in runMAL (cntxt=0x6047e0, mb=0x610ad8, startpc=1, mbcaller=0x610ad8, env=0x7fffff3ffdf0, pcicaller=0x660d68) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB5/src/mal/mal_interpreter.mx:308
2 0x00007ffff7cf97bd in runMALsequence (cntxt=0x6047e0, mb=0x610ad8, startpc=1, stoppc=0, stk=0x7fffff3ffdf0, env=0x7fffff400a30, pcicaller=0x660d68)
at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB5/src/mal/mal_interpreter.mx:1770
3 0x00007ffff7cec7fc in runMAL (cntxt=0x6047e0, mb=0x610ad8, startpc=1, mbcaller=0x610ad8, env=0x7fffff400a30, pcicaller=0x660d68) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB5/src/mal/mal_interpreter.mx:308
[...]
8020 0x00007ffff7cf97bd in runMALsequence (cntxt=0x6047e0, mb=0x610ad8, startpc=1, stoppc=0, stk=0xd76b68, env=0xd76b68, pcicaller=0x0) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB5/src/mal/mal_interpreter.mx:1770
8021 0x00007ffff7cec7fc in runMAL (cntxt=0x6047e0, mb=0x610ad8, startpc=1, mbcaller=0x0, env=0xd76b68, pcicaller=0x0) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB5/src/mal/mal_interpreter.mx:308
8022 0x00007ffff7cebb54 in MALengine (c=0x6047e0) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB5/src/mal/mal_session.mx:691
8023 0x00007ffff7d2fe49 in runPhase (c=0x6047e0, phase=4) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB5/src/mal/mal_scenario.mx:602
8024 0x00007ffff7d2ffd6 in runScenarioBody (c=0x6047e0) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB5/src/mal/mal_scenario.mx:647
8025 0x00007ffff7d3016d in runScenario (c=0x6047e0) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB5/src/mal/mal_scenario.mx:672
8026 0x00007ffff7ceafe2 in MSserveClient (dummy=0x6047e0) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB5/src/mal/mal_session.mx:496
8027 0x0000000000402e22 in main (argc=1, av=0x7fffffffdda8) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB5/src/tools/mserver5.mx:623
(gdb)
## Comment 12344
Date: 2009-09-06 09:04:16 +0200
From: @mlkersten
Confirmed. It is the famous recursive call. There is code in the interpreter to catch it, but was turned into comments a long time ago because it consumed too much time, hampering the execution of small functions.
## Comment 12345
Date: 2009-09-06 09:40:38 +0200
From: @mlkersten
The stable version has been extended with a cheaper test on runtime stack overflow.
We top the number of recursive function calls and watch out for a large MAL
variable stack.
The reported bug is cast into a test src/mal/Tests/pithole.mal
## Comment 12346
Date: 2009-11-15 15:39:05 +0100
From: @mlkersten
The pithole.mal test has been added.For all but a few platforms the defense is working.
Closing this report, as further actions should/could be triggered by the test web.
## Comment 12347
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2852707 at http://sourceforge.net/support/tracker.php?aid=2852707
| non existing mal function segv | https://api.github.com/repos/MonetDB/MonetDB/issues/2180/comments | 0 | 2020-11-30T10:26:10Z | 2024-06-28T13:22:40Z | https://github.com/MonetDB/MonetDB/issues/2180 | 753,352,646 | 2,180 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-04 02:27:26 +0200
From: @yzchang
To: Jan Flokstra <<jflokstra>>
Version: -- development
Last updated: 2011-04-28 16:38:58 +0200
## Comment 12333
Date: 2009-09-04 14:27:26 +0200
From: @yzchang
I sometimes get the error that there is no matching MIL operator for some Tijah doc tape MIL proc.
This happens some what unpredictable, i.e., it doesn't happen every time.
However, once I start to get this error, it won't disappear even by restarting Mserver.
I have seen this error on different systems, e.g. FC10 and Mac OS10.5.
The error occurs both with the algebra and the MPS backend.
In the following mclient sessions, I was trying to add a document to a new DB.
It returns the error, while the documents have been added.
$ mclient -lx
xquery>pf:add-doc("/ufs/zhang/xmark/tkde/xmark2mb.xml", "test.xml", 'test',10)
more><>
MAPI = monetdb@localhost:50000
QUERY = pf:add-doc("/ufs/zhang/xmark/tkde/xmark2mb.xml", "test.xml", 'test',10)
ERROR = !ERROR: interpret: no matching MIL operator to 'ALG_tj_docmgmt_tape(BAT[str,BAT], BAT[oid,BAT], BAT[oid,str], BAT[oid,str], BAT[oid,str], BAT[oid,lng])'.
xquery>count(doc("test.xml"))
more><>
1
xquery>\q
$ mclient -lx -g
xquery>pf:add-doc("/ufs/zhang/xmark/tkde/xmark2mb.xml", "test2.xml", 'test',10)
more><>
MAPI = monetdb@localhost:50000
QUERY = pf:add-doc("/ufs/zhang/xmark/tkde/xmark2mb.xml", "test2.xml", 'test',10)
ERROR = !ERROR: interpret: no matching MIL operator to 'tj_play_doc_tape(BAT[oid,BAT], BAT[oid,oid], BAT[oid,int], BAT[oid,lng], BAT[oid,str])'.
xquery>count(doc("test2.xml"))
more><>
1
xquery>pf:documents()
more><>
<document updatable="true" url="/ufs/zhang/xmark/tkde/xmark2mb.xml" collection="test">test.xml</document>,
<document updatable="true" url="/ufs/zhang/xmark/tkde/xmark2mb.xml" collection="test">test2.xml</document>
xquery>\q
## Comment 12334
Date: 2009-09-04 14:49:54 +0200
From: @drstmane
could you please check and report whether the problem also occurs with
property checking enabled, i.e., when starting Mserver with -d10 ?
If this prevent the error, the problem might be a not correctly
set denseness property "somewhere" in some code that creats a BAT,
triggering a function resolution mismatch between an expected dense BAT-head (or -tail)
and a provided arbitrary (not knownly dense) BAT-head (or -tail) ...
## Comment 12335
Date: 2009-09-04 15:25:28 +0200
From: @yzchang
Unfortunately, starting Mserver with -d10 doesn't make difference.
## Comment 12336
Date: 2009-09-07 10:54:53 +0200
From: Jan Flokstra <<jflokstra>>
There ineed seems to be a problem with dense heads here. It is only strange it only happens with pftijah. The 'subject' function was created with the function 'play_update_tape' in pf_support.mx as example. This function uses [void,xxx] bats as parameter.
PROC play_update_tape(bat[void, bat] ws, bat[void, oid] item, bat[void, int] kind, bat[void,lng] int_values, bat[void,str] str_values) : void
So the signature of the failing function in modules/pftijah/pftijah.mx is
PROC ALG_tj_docmgmt_tape(BAT[str,bat] tape,
BAT[void,BAT] ws,
BAT[void,str] location,
BAT[void,str] docnames,
BAT[void,str] colnames,
BAT[void,lng] percentages) : bit
I think you can change the voids in the headtypes to oid and then look what happens. When this solves the problem without a lot of deep debugging I haqve no problem with this.
## Comment 12337
Date: 2009-09-07 11:27:47 +0200
From: @drstmane
The main question is, what does PROC play_update_tape() expect and/or what can it cope with in the BAT heads?
- empty void-nil heads (TYPE_void with seqbase == oid_nil)
- dense non-materialized OIDs (aka. "Virtual ID", or VoID; i.e., TYPE_void with seqbase != oid_nil)
- dense materialized OIDs (TYPE_oid, dense sequence with denseness property set and seqbase != oid_nil)
- arbitrary materialized OIDs (TYPE_oid, arbitrary values, denseness property not set, seqbase not relevant/used)
?
The answer to this questions will determine the proper way to fix this bug ...
## Comment 12338
Date: 2009-09-07 11:30:28 +0200
From: @drstmane
Note the orignal error message is actually about PROC / command tj_play_doc_tape() and ALG_tj_docmgmt_tape(), not play_update_tape() ...
## Comment 12339
Date: 2009-09-07 11:45:27 +0200
From: @lsidir
tj_play_doc_tape() ,ALG_tj_docmgmt_tape() and play_update_tape() should take bat[void,xxx] as parameters. In my experience, failing to do so it means that some part of the translation from algebra to mil is not working properly. changing from void to oid the signature will not solve the problem, because eventually at some point you will take wrongly aligned bats. Sometimes the solution is to perform the correct casts in the milgen part, but I haven't seen in detail this bug.
## Comment 12340
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2851106 at http://sourceforge.net/support/tracker.php?aid=2851106
## Comment 15768
Date: 2011-04-28 16:38:58 +0200
From: @sjoerdmullender
Development on MonetDB4 and MonetDB/XQuery has ceased. It is unlikely that this bug will be fixed.
The last release of MonetDB/XQuery and MonetDB4 is Mar2011.
| PF: problems with Tijah doc tape | https://api.github.com/repos/MonetDB/MonetDB/issues/2179/comments | 0 | 2020-11-30T10:26:07Z | 2024-06-28T13:22:39Z | https://github.com/MonetDB/MonetDB/issues/2179 | 753,352,612 | 2,179 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-04 01:17:49 +0200
From: @skinkie
To: GDK devs <<bugs-common>>
Version: -- development
CC: y.kargin
Last updated: 2012-11-27 16:14:36 +0100
## Comment 12320
Date: 2009-09-04 01:17:49 +0200
From: @skinkie
The following query:
insert into namefinder select lower(name), lat, long, v from (select min(way_nds.way) as way, name, bit_and(zcurve, 0xfffffffee0000000) as smooth from (select ways.id as way, v as name from ways, way_tags where ways.id = way_tags.way and way_tags.k = 'name') as x, way_nds, nodes_legacy where x.way = way_nds.way and to_node = id group by name, smooth) as z, way_nds, nodes_legacy LEFT JOIN way_tags ON way = way_tags.way where z.way = way_nds.way and way_tags.k = 'is_in' and to_node = nodes_legacy.id and idx = 1;
results in:
mserver5: gdk.h:778: BUNlast: Assertion `b->U->count <= (((BUN) 2147483647) - 1)' failed.
Program received signal SIGABRT, Aborted.
[Switching to Thread 0x7fffe13d7950 (LWP 28546)]
0x0000003c19c32f05 in raise () from /lib64/libc.so.6
Missing separate debuginfos, use: debuginfo-install bzip2-libs-1.0.5-3.fc10.x86_64 cyrus-sasl-lib-2.1.22-19.fc10.x86_64 e2fsprogs-libs-1.41.4-6.fc10.x86_64 glibc-2.9-3.x86_64 keyutils-libs-1.2-3.fc9.x86_64 krb5-libs-1.6.3-18.fc10.x86_64 libcurl-7.19.6-1.fc10.x86_64 libidn-0.6.14-8.x86_64 libselinux-2.0.78-1.fc10.x86_64 libssh2-0.18-7.fc9.x86_64 libxml2-2.7.3-2.fc10.x86_64 libxslt-1.1.24-2.fc10.x86_64 ncurses-libs-5.6-20.20080927.fc10.x86_64 nspr-4.7.4-1.fc10.x86_64 nss-3.12.3.99.3-2.10.4.fc10.x86_64 openldap-2.4.12-1.fc10.x86_64 openssl-0.9.8g-14.fc10.x86_64 pcre-7.8-1.fc10.x86_64 raptor-1.4.18-1.fc10.x86_64 readline-5.2-13.fc9.x86_64 zlib-1.2.3-18.fc9.x86_64
(gdb) bt
0 0x0000003c19c32f05 in raise () from /lib64/libc.so.6
1 0x0000003c19c34a73 in abort () from /lib64/libc.so.6
2 0x0000003c19c2bef9 in __assert_fail () from /lib64/libc.so.6
3 0x00007ffff76bd334 in BUNlast (b=0x13a2188) at gdk.h:778
4 0x00007ffff76b11c7 in mergejoin (l=0x139ef58, r=0x7fffd4000ae0, bn=0x13a2188, nil_on_miss=0x0, estimate=2147483647, limit=0x0) at ../../../src/gdk/gdk_relop.mx:232
5 0x00007ffff76abbc6 in mergejoin (l=0x139ef58, r=0x7fffd4000ae0, bn=0x0, nil_on_miss=0x0, estimate=2147483647, limit=0x0) at ../../../src/gdk/gdk_relop.mx:94
6 0x00007ffff76bd798 in batmergejoin (l=0x7fffd4000ab8, r=0x13a64f0, estimate=2147483647, swap=1 '\001', limit=0x0) at ../../../src/gdk/gdk_relop.mx:394
7 0x00007ffff76f9c5a in batjoin (l=0x7fffd4000ab8, r=0x13a64f0, estimate=2147483647, swap=1 '\001') at ../../../src/gdk/gdk_relop.mx:1205
8 0x00007ffff76f9d64 in BATjoin (l=0x7fffd4000ab8, r=0x13a64f0, estimate=2147483647) at ../../../src/gdk/gdk_relop.mx:1225
9 0x00007fffebc1c2c1 in CMDjoin (result=0x7fffe13d4a68, left=0x7fffd4000ab8, right=0x13a64f0, estimate=0x7fffe13d4ab8) at ../../../../src/modules/kernel/algebra.mx:1552
10 0x00007fffebc28014 in ALGjoinestimate (result=0x13cb638, lid=0x13cb628, rid=0x13cb618, estimate=0x7fffe13d4ab8) at ../../../../src/modules/kernel/algebra.mx:2999
11 0x00007fffebc2813e in ALGjoin (result=0x13cb638, lid=0x13cb628, rid=0x13cb618) at ../../../../src/modules/kernel/algebra.mx:3017
12 0x00007ffff7d09de2 in DFLOWstep (t=0x138de38, fs=0x7fffe23dbfe8) at ../../../src/mal/mal_interpreter.mx:1644
13 0x00007ffff7d0e926 in runDFLOWworker (t=0x138de38) at ../../../src/mal/mal_interpreter.mx:839
14 0x00007ffff779463b in wrapper_routine (data=0x7fffe23dc160) at ../../../src/gdk/gdk_posix.mx:983
15 0x0000003c1a4073da in start_thread () from /lib64/libpthread.so.0
16 0x0000003c19ce62bd in clone () from /lib64/libc.so.6
Which basically got into such behavior after I added the left join. The following query just works:
insert into namefinder select lower(name), lat, long from (select min(way_nds.way) as way, name, bit_and(zcurve, 0xfffffffee0000000) as smooth from (select ways.id as way, v as name from ways, way_tags where ways.id = way_tags.way and way_tags.k = 'name') as x, way_nds, nodes_legacy where x.way = way_nds.way and to_node = id group by name, smooth) as z, way_nds, nodes_legacy where z.way = way_nds.way and to_node = nodes_legacy.id and idx = 1;
## Comment 12321
Date: 2009-09-04 09:40:03 +0200
From: @drstmane
apparently at least one intermediate result of your first query (with the left join) exceeds the count that MonetDB can handle on your (not specified) system.
## Comment 12322
Date: 2009-09-04 16:13:41 +0200
From: @skinkie
I am using oid32 on Capella; would it be possible to get a more eighter:
- a more user friendly message
- stopping the execution and gracefully continue
- pursue another plan that uses a smaller intermediate result
- split the task(?)
## Comment 12323
Date: 2009-09-07 11:24:58 +0200
From: @sjoerdmullender
I added a test in gdk_relop.mx to prevent a BAT from growing beyond BUN_MAX so that the assertion should now be prevented and the query should be aborted one iteration before you would otherwise have gotten the crash.
Can you test whether this works?
I'll propagate the fix to the development branch shortly.
## Comment 12324
Date: 2009-09-07 12:18:51 +0200
From: @skinkie
I see a load of about 100% now, but it doesn't seem to crash like it did before.
## Comment 12325
Date: 2009-09-07 14:12:32 +0200
From: @skinkie
>mserver5: ../../../src/gdk/gdk_bat.mx:291: BATnew: Assertion `cap <= (((BUN) 2147483647) - 1)' failed.
Aborted
...if it should be already in current, then it does not work.
## Comment 12326
Date: 2009-09-07 16:06:46 +0200
From: @sjoerdmullender
That's a different assertion. I.e. my fix does seem to work but there are more places like that.
## Comment 12327
Date: 2009-09-07 17:24:41 +0200
From: @sjoerdmullender
By the way, it would be useful to know where BATnew is called from in your latest assertion failed.
## Comment 12328
Date: 2009-09-08 01:08:18 +0200
From: @skinkie
(gdb) bt
0 0x0000003c19c32f05 in raise () from /lib64/libc.so.6
1 0x0000003c19c34a73 in abort () from /lib64/libc.so.6
2 0x0000003c19c2bef9 in __assert_fail () from /lib64/libc.so.6
3 0x00007ffff7686a79 in BATnew (ht=7, tt=7, cap=2699054870)
at ../../../src/gdk/gdk_bat.mx:291
4 0x00007ffff76aac18 in mergejoin (l=0x12b2d58, r=0x7fffd40010d0, bn=0x0,
nil_on_miss=0x0, estimate=2147483647, limit=0x0)
at ../../../src/gdk/gdk_relop.mx:140
5 0x00007ffff76bc4dd in batmergejoin (l=0x7fffd40010a8, r=0x12b2d80,
estimate=2147483647, swap=1 '\001', limit=0x0)
at ../../../src/gdk/gdk_relop.mx:396
6 0x00007ffff76f899f in batjoin (l=0x7fffd40010a8, r=0x12b2d80,
estimate=2147483647, swap=1 '\001') at ../../../src/gdk/gdk_relop.mx:1207
7 0x00007ffff76f8aa9 in BATjoin (l=0x7fffd40010a8, r=0x12b2d80,
estimate=2147483647) at ../../../src/gdk/gdk_relop.mx:1227
8 0x00007fffebc1b2c1 in CMDjoin (result=0x7fffe197ba68, left=0x7fffd40010a8,
right=0x12b2d80, estimate=0x7fffe197bab8)
at ../../../../src/modules/kernel/algebra.mx:1552
9 0x00007fffebc27014 in ALGjoinestimate (result=0x12e23a0, lid=0x12e2390,
rid=0x12e2380, estimate=0x7fffe197bab8)
at ../../../../src/modules/kernel/algebra.mx:2999
10 0x00007fffebc2713e in ALGjoin (result=0x12e23a0, lid=0x12e2390,
rid=0x12e2380) at ../../../../src/modules/kernel/algebra.mx:3017
---Type <return> to continue, or q <return> to quit---
11 0x00007ffff7d08f60 in DFLOWstep (t=0x12b7c58, fs=0x7fffe2982fe8)
at ../../../src/mal/mal_interpreter.mx:1650
12 0x00007ffff7d0dac0 in runDFLOWworker (t=0x12b7c58)
at ../../../src/mal/mal_interpreter.mx:845
13 0x00007ffff7793383 in wrapper_routine (data=0x7fffe2983160)
at ../../../src/gdk/gdk_posix.mx:983
14 0x0000003c1a4073da in start_thread () from /lib64/libpthread.so.0
15 0x0000003c19ce62bd in clone () from /lib64/libc.so.6
hope that helps.
## Comment 12329
Date: 2009-09-08 11:21:36 +0200
From: @sjoerdmullender
I added more checks against overflow. Can you check again?
## Comment 12330
Date: 2009-09-18 17:38:30 +0200
From: @sjoerdmullender
Stefan, what is the situation now?
## Comment 12331
Date: 2009-09-18 17:45:47 +0200
From: @skinkie
After your patch it returns after an extremely long time:
0 affected rows
The first time the error came much earlier. And the output is not as expected.
## Comment 12332
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2850513 at http://sourceforge.net/support/tracker.php?aid=2850513
## Comment 18141
Date: 2012-11-27 16:14:36 +0100
From: Yagiz <<y.kargin>>
Is this still a problem? If yes, we need the schema definition and data to reproduce the problem.
| BUNlast: Assertion `b->U->count <= (((BUN) 2147483647) - 1)' | https://api.github.com/repos/MonetDB/MonetDB/issues/2178/comments | 0 | 2020-11-30T10:26:04Z | 2024-06-27T11:41:27Z | https://github.com/MonetDB/MonetDB/issues/2178 | 753,352,574 | 2,178 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-03 11:32:52 +0200
From: @skinkie
To: @njnes
Version: -- development
Last updated: 2009-09-17 10:32:46 +0200
## Comment 12316
Date: 2009-09-03 23:32:52 +0200
From: @skinkie
sql>select bit_and(3749090034127126942, 0xffffffffffffffff);
+----------------------+
| bit_and_single_value |
+======================+
| 3748508925034660238 |
+----------------------+
## Comment 12317
Date: 2009-09-07 12:00:08 +0200
From: @skinkie
That looks working now!
(See new bug, I did find a glitch)
## Comment 12318
Date: 2009-09-17 22:32:46 +0200
From: @njnes
test bit_and.SF-2850341.sql added
## Comment 12319
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2850341 at http://sourceforge.net/support/tracker.php?aid=2850341
| SQL: bit_and returns invalid values | https://api.github.com/repos/MonetDB/MonetDB/issues/2177/comments | 0 | 2020-11-30T10:26:00Z | 2024-06-27T11:41:26Z | https://github.com/MonetDB/MonetDB/issues/2177 | 753,352,522 | 2,177 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-09-01 05:39:29 +0200
From: francois guerin <<fguerin>>
To: Fabian Groffen <<fabian>>
Version: -- development
Last updated: 2009-09-07 03:14:08 +0200
## Comment 12310
Date: 2009-09-01 17:39:29 +0200
From: francois guerin <<fguerin>>
Hi,
I've just installed the newest version of MacOSX (ie 10.6, which is claimed to be a pure 64bits architecture, smaller, better, faster, stronger, etc.)
for i was no sure my last installation of MonetDB would be still compatible, i re-installed it (monetdb-install.sh --enable-xquery --enable-optimize) and got the error:
In file included from MapiClient.c:32:
ReadlineTools.h:9: error: expected '=', ',', ';', 'asm' or '__attribute__' before '*' token
ReadlineTools.h:10: error: expected ')' before '*' token
MapiClient.c: In function 'doFileByLines':
MapiClient.c:1292: error: 'rl_completion_func_t' undeclared (first use in this function)
MapiClient.c:1292: error: (Each undeclared identifier is reported only once
MapiClient.c:1292: error: for each function it appears in.)
MapiClient.c:1292: error: 'func' undeclared (first use in this function)
MapiClient.c:1297: warning: implicit declaration of function 'suspend_completion'
MapiClient.c:1300: warning: implicit declaration of function 'continue_completion'
MapiClient.c:1556: warning: assignment discards qualifiers from pointer target type
gnumake[5]: *** [MapiClient.o] Error 1
gnumake[4]: *** [all] Error 2
gnumake[3]: *** [all-recursive] Error 1
gnumake[2]: *** [all] Error 2
gnumake[1]: *** [all-recursive] Error 1
gnumake: *** [all] Error 2
making MonetDB-client failed
(Take a look at the attached file my log file)
thanks.
## Comment 12311
Date: 2009-09-01 17:49:24 +0200
From: @grobian
problem is readline related, probably we don't really deal very well with a new version of libedit that immitates as readline even more
## Comment 12312
Date: 2009-09-02 14:30:12 +0200
From: francois guerin <<fguerin>>
perhaps it's hidden, but i dont find anywhere where you declare the type "rl_completion_func_t" ...
(it's the function quoted in my logfile)
i applied a grep in my installation folder and this is what i have obtained:
./MonetDB-client/clients/src/mapiclient/MapiClient.c: rl_completion_func_t *func = NULL;
./MonetDB-client/clients/src/mapiclient/MapiClient.mx: rl_completion_func_t *func = NULL;
./MonetDB-client/clients/src/mapiclient/ReadlineTools.c:rl_completion_func_t *
./MonetDB-client/clients/src/mapiclient/ReadlineTools.c: rl_completion_func_t *func = rl_attempted_completion_function;
./MonetDB-client/clients/src/mapiclient/ReadlineTools.c:continue_completion(rl_completion_func_t * func)
./MonetDB-client/clients/src/mapiclient/ReadlineTools.h:rl_completion_func_t *suspend_completion(void);
./MonetDB-client/clients/src/mapiclient/ReadlineTools.h:void continue_completion(rl_completion_func_t * func);
./MonetDB-client/clients/src/mapiclient/ReadlineTools.mx:rl_completion_func_t *suspend_completion(void);
./MonetDB-client/clients/src/mapiclient/ReadlineTools.mx:void continue_completion(rl_completion_func_t * func);
./MonetDB-client/clients/src/mapiclient/ReadlineTools.mx:rl_completion_func_t *
./MonetDB-client/clients/src/mapiclient/ReadlineTools.mx: rl_completion_func_t *func = rl_attempted_completion_function;
./MonetDB-client/clients/src/mapiclient/ReadlineTools.mx:continue_completion(rl_completion_func_t * func)
./MonetDB-SQL/sql/src/backends/monet5/sql_readline.c:rl_completion_func_t *suspend_completion(void);
./MonetDB-SQL/sql/src/backends/monet5/sql_readline.c:void continue_completion(rl_completion_func_t * func);
./MonetDB-SQL/sql/src/backends/monet5/sql_readline.mx:rl_completion_func_t *suspend_completion(void);
./MonetDB-SQL/sql/src/backends/monet5/sql_readline.mx:void continue_completion(rl_completion_func_t * func);
./MonetDB5-server/MonetDB5/src/mal/mal_readline.c:rl_completion_func_t *suspend_completion(void);
./MonetDB5-server/MonetDB5/src/mal/mal_readline.c:void continue_completion(rl_completion_func_t * func);
./MonetDB5-server/MonetDB5/src/mal/mal_readline.c:rl_completion_func_t *
./MonetDB5-server/MonetDB5/src/mal/mal_readline.c: rl_completion_func_t *func = rl_attempted_completion_function;
./MonetDB5-server/MonetDB5/src/mal/mal_readline.c:continue_completion(rl_completion_func_t * func)
./MonetDB5-server/MonetDB5/src/mal/mal_readline.mx:rl_completion_func_t *suspend_completion(void);
./MonetDB5-server/MonetDB5/src/mal/mal_readline.mx:void continue_completion(rl_completion_func_t * func);
./MonetDB5-server/MonetDB5/src/mal/mal_readline.mx:rl_completion_func_t *
./MonetDB5-server/MonetDB5/src/mal/mal_readline.mx: rl_completion_func_t *func = rl_attempted_completion_function;
./MonetDB5-server/MonetDB5/src/mal/mal_readline.mx:continue_completion(rl_completion_func_t * func)
the "rl_completion_func_t" is nowhere to be found.
## Comment 12313
Date: 2009-09-02 14:39:10 +0200
From: francois guerin <<fguerin>>
ok, my fault.
the "rl_completion_func_t" is located in "rltypedefs.h"
(it's not in monetdb sources, but formerly in my computer)
## Comment 12314
Date: 2009-09-07 15:14:08 +0200
From: @grobian
ok, I added a configure check for tl_completion_func_t, that should do the trick. Available tomorrow with --nightly=stable (argument to monetdb-install.sh) or in Aug2009-SP1
## Comment 12315
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2848525 at http://sourceforge.net/support/tracker.php?aid=2848525
| mac10.6 and monetdb installation | https://api.github.com/repos/MonetDB/MonetDB/issues/2176/comments | 0 | 2020-11-30T10:25:56Z | 2024-06-27T11:41:25Z | https://github.com/MonetDB/MonetDB/issues/2176 | 753,352,488 | 2,176 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-31 03:48:16 +0200
From: nickloeve <<nickloeve>>
To: Fabian Groffen <<fabian>>
Version: -- development
Last updated: 2009-10-02 03:35:53 +0200
## Comment 12300
Date: 2009-08-31 15:48:16 +0200
From: nickloeve <<nickloeve>>
I tried to install the superball release, and the build failed for the client, as the PHP bindings use deprecated interfaces.
## Comment 12301
Date: 2009-08-31 16:26:43 +0200
From: @drstmane
The major problem is not the deprecated interfaces (these are "merely" warnings), but rather:
php_monetdb.c: In function ‘php_monetdb_fetch_hash’:
php_monetdb.c:1719: error: ‘zend_fcall_info’ has no member named ‘object_pp’
php_monetdb.c:1757: error: ‘zend_fcall_info_cache’ has no member named ‘object_pp’
Looking at the code (I couldn't find any respective documentation) it appears that PHP chose to rename the "object_pp" members of structs "zend_fcall_info" & "zend_fcall_info_cache" between versions 5.2.9 and 5.3.0 ...
We'll need to extend our configure scripts to detect this "subtle" difference ...
## Comment 12302
Date: 2009-08-31 16:41:39 +0200
From: nickloeve <<nickloeve>>
Ah, indeed, I didn't look too closely (as you can see) :)
## Comment 12303
Date: 2009-08-31 17:04:07 +0200
From: @grobian
seems we're the only one on (google) earth with this problem though
## Comment 12304
Date: 2009-08-31 17:10:59 +0200
From: nickloeve <<nickloeve>>
I was thinking of having a bash at converting this to a PDO plugin. It might be worth looking into now, if you have to do much porting.
From what I can tell, some others with the issue included the php-gtk extension and an experimental sqlite ext that has since been killed.
## Comment 12305
Date: 2009-08-31 17:19:53 +0200
From: @grobian
I think I fixed the issue, I synced with the 5_3 branch from PHP. However, we have a native PHP driver in development (written in PHP) which is most probably more stable than the C implementation.
## Comment 12306
Date: 2009-09-07 14:30:44 +0200
From: @grobian
ok, my previous fix broke more than it solved (nothing), so I developed a new strategy: configure now requires PHP < 5.3.0 to build the php module.
## Comment 12307
Date: 2009-10-02 15:31:08 +0200
From: @sjoerdmullender
What is the status? Can this be closed?
## Comment 12308
Date: 2009-10-02 15:35:48 +0200
From: @grobian
assuming it does since Aug2009-SP1 release
## Comment 12309
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2847738 at http://sourceforge.net/support/tracker.php?aid=2847738
| SuperBall build fails with PHP 5.3.0 | https://api.github.com/repos/MonetDB/MonetDB/issues/2175/comments | 0 | 2020-11-30T10:25:53Z | 2024-06-27T11:41:24Z | https://github.com/MonetDB/MonetDB/issues/2175 | 753,352,447 | 2,175 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-30 02:21:35 +0200
From: @skinkie
To: @njnes
Version: -- development
Last updated: 2009-09-17 10:33:41 +0200
## Comment 12297
Date: 2009-08-30 02:21:35 +0200
From: @skinkie
CREATE TABLE "sys"."way_nds" (
"way" int,
"idx" int,
"to_node" int
);
CREATE TABLE "sys"."nodes_legacy" (
"id" int NOT NULL,
"long" double,
"lat" double,
"uid" int,
"timestamp" TIMESTAMP WITH TIME ZONE,
"zcurve" bigint
);
select way_nds.way, zcurve from way_nds, nodes_legacy where to_node = id and zcurve between (zcurve - 4096) and (zcurve + 4096);
mserver5: ../../../src/server/rel_bin.mx:517: exp_bin: Assertion `0' failed.
Program received signal SIGABRT, Aborted.
[Switching to Thread 0x7fce851046f0 (LWP 7486)]
0x00007fce821ae645 in raise () from /lib/libc.so.6
(gdb) bt
0 0x00007fce821ae645 in raise () from /lib/libc.so.6
1 0x00007fce821afb63 in abort () from /lib/libc.so.6
2 0x00007fce821a76d9 in __assert_fail () from /lib/libc.so.6
3 0x00007fce72ee5ff1 in exp_bin (sql=0x2c1cfe8, e=0x2d00fd8, left=0x2cf0cb8, right=0x2d26998, grp=0x0, sel=0x0) at ../../../src/server/rel_bin.mx:517
4 0x00007fce72ee717d in rel2bin_join (sql=0x2c1cfe8, rel=0x2d03ac8, refs=0x2d0e758) at ../../../src/server/rel_bin.mx:797
5 0x00007fce72ef0d60 in subrel_bin (sql=0x2c1cfe8, rel=0x2d03ac8, refs=0x2d0e758) at ../../../src/server/rel_bin.mx:3389
6 0x00007fce72ee9680 in rel2bin_project (sql=0x2c1cfe8, rel=0x2d01108, refs=0x2d0e758) at ../../../src/server/rel_bin.mx:1409
7 0x00007fce72ef0e46 in subrel_bin (sql=0x2c1cfe8, rel=0x2d01108, refs=0x2d0e758) at ../../../src/server/rel_bin.mx:3410
8 0x00007fce72eea6df in rel2bin_topn (sql=0x2c1cfe8, rel=0x2d013c8, refs=0x2d0e758) at ../../../src/server/rel_bin.mx:1735
9 0x00007fce72ef0ed0 in subrel_bin (sql=0x2c1cfe8, rel=0x2d013c8, refs=0x2d0e758) at ../../../src/server/rel_bin.mx:3422
10 0x00007fce72ef10bf in output_rel_bin (sql=0x2c1cfe8, rel=0x2d013c8) at ../../../src/server/rel_bin.mx:3469
11 0x00007fce72e35158 in sql_symbol2stmt (c=0x2c1cfe8, sym=0x2d17058) at ../../../../src/backends/monet5/sql.mx:1243
12 0x00007fce72e708cb in SQLparser (c=0x605560) at ../../../../src/backends/monet5/sql_scenario.mx:1030
13 0x00007fce84c56f85 in runPhase (c=0x605560, phase=1) at ../../../src/mal/mal_scenario.mx:602
14 0x00007fce84c57082 in runScenarioBody (c=0x605560) at ../../../src/mal/mal_scenario.mx:639
15 0x00007fce84c572e1 in runScenario (c=0x605560) at ../../../src/mal/mal_scenario.mx:672
16 0x00007fce84c0e185 in MSserveClient (dummy=0x605560) at ../../../src/mal/mal_session.mx:496
17 0x00000000004035ba in main (argc=2, av=0x7fffc3780878) at ../../../src/tools/mserver5.mx:637
ps. yes binary operations would be more than welcome ;)
## Comment 12298
Date: 2009-09-03 22:45:10 +0200
From: @njnes
fixed bug by properly handling this special case of a select (its a buggy query, the selections doesn't do anything). Added a test crash_on_complex_join.SF-2847106
## Comment 12299
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2847106 at http://sourceforge.net/support/tracker.php?aid=2847106
| SQL: crash upon join with advanced between | https://api.github.com/repos/MonetDB/MonetDB/issues/2174/comments | 0 | 2020-11-30T10:25:49Z | 2024-06-27T11:41:23Z | https://github.com/MonetDB/MonetDB/issues/2174 | 753,352,402 | 2,174 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-29 11:38:41 +0200
From: @skinkie
To: @mlkersten
Version: -- development
Last updated: 2009-09-03 10:23:25 +0200
## Comment 12294
Date: 2009-08-29 23:38:41 +0200
From: @skinkie
2246, 'created_by', 'JOSM'
Where before above line using: COPY 1 RECORDS INTO way_tags from '/tmp/bla1.csv' USING DELIMITERS ',', '\n', '''';
Was imported as:
2246
created_by
JOSM
It is now imported as
2246
[space]'created_by'
[space]'JOSM'
While testing different variations the following came on the server console:
>SQLworker exists (null)
It seems that when the space is removed after the seperator the field is imported with the other separator undone as well.
## Comment 12295
Date: 2009-09-03 22:23:25 +0200
From: @mlkersten
Yes, leading white spaces are never ignored anymore!
This means that a string field started after a separator should immediately start with a quote,
if it is the intention to enter a quoted string.
## Comment 12296
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2847050 at http://sourceforge.net/support/tracker.php?aid=2847050
| SQL: regression in string parsing at COPY INTO | https://api.github.com/repos/MonetDB/MonetDB/issues/2173/comments | 0 | 2020-11-30T10:25:46Z | 2024-06-28T13:23:53Z | https://github.com/MonetDB/MonetDB/issues/2173 | 753,352,357 | 2,173 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-29 10:45:25 +0200
From: @skinkie
To: @njnes
Version: -- development
Last updated: 2009-08-29 11:47:59 +0200
## Comment 12291
Date: 2009-08-29 22:45:25 +0200
From: @skinkie
Make a file /tmp/bla with:
123271, 3.5667269, 50.3409331, 1267, '2007-08-15T19:58:59', 3749088843704927238
CREATE TABLE nodes_legacy (id integer, long double, lat double, uid integer, timestamp timestamptz, zcurve bigint);
COPY 1 RECORDS INTO nodes_legacy from '/tmp/bla.csv' USING DELIMITERS ',', '\n', '''';
!SQLException:sql:value ' '2007-08-15T19:58:59'' from line 1 field 5 not inserted, expecting type timestamp.
!SQLException:importTable:failed to import table
0 tuples
Oh really...
insert into nodes_legacy values (123271, 3.5667269, 50.3409331, 1267, '2007-08-15T19:58:59', 3749088843704927238);
1 affected row
## Comment 12292
Date: 2009-08-29 23:47:57 +0200
From: @skinkie
See string space bug. Removing the space fixes this one.
## Comment 12293
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2847009 at http://sourceforge.net/support/tracker.php?aid=2847009
| SQL: regression in timestamp parsing from COPY INTO | https://api.github.com/repos/MonetDB/MonetDB/issues/2172/comments | 0 | 2020-11-30T10:25:42Z | 2024-06-28T13:23:53Z | https://github.com/MonetDB/MonetDB/issues/2172 | 753,352,309 | 2,172 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-28 03:38:49 +0200
From: Matthew W. Jones <<matburt>>
To: @njnes
Version: -- development
Last updated: 2009-09-17 10:34:10 +0200
## Comment 12282
Date: 2009-08-28 15:38:49 +0200
From: Matthew W. Jones <<matburt>>
I believe I may have found a bug in the sequence though... I want to make sure that this value will not have problems when working with large integers:
sql> CREATE SEQUENCE "test_seq" as bigint;
sql> select next value for test_seq;
+------------------------------+
| next_value_for_testdb |
+==============================+
| 1 |
+------------------------------+
sql> alter sequence test_seq restart with 3000000000;
sql> select next value for test_seq;
+------------------------------+
| next_value_for_testdb |
+==============================+
| 2147483647 |
+------------------------------+
I can always increment the sequence higher, but if I try to alter it to anything past the maximum value for a 32-bit integer then it will reset to it.
sql>select next value for test_seq;
+------------------------------+
| next_value_for_testdb |
+==============================+
| 2147483648 |
+------------------------------+
1 tuple
sql>select next value for test_seq;
+------------------------------+
| next_value_for_testdb |
+==============================+
| 2147483649 |
+------------------------------+
## Comment 12283
Date: 2009-08-28 16:03:32 +0200
From: @grobian
This might actually be a bug related to serial being a INT, not a BIGINT. We should probably implement bigserial, and fix the sequence code here not to accept the too large value.
## Comment 12284
Date: 2009-08-28 17:29:41 +0200
From: Matthew W. Jones <<matburt>>
I'm actually doing this outside of an auto increment or serial field in a table.... I'm just strictly creating, querying, and altering the sequence. Looking in the 'sequences' table shows that all of its integer fields are bigints.... it also seems to increment past a 32-bit integer.... the error must be in the code that takes the new value during the sql query for 'alter sequence restart ...'
## Comment 12285
Date: 2009-08-28 18:26:20 +0200
From: Matthew W. Jones <<matburt>>
Here's some more interesting information:
sql>explain alter sequence my_other_test_seq restart with 3000000000 ;
+-------------------------------------------------------------------------------------------------------+
| function user.s0_1{autoCommit=true}():void; |
| sql.restart("sys","my_other_test_seq",2147483647:lng); |
| sql.exportOperation(""); |
| end s0_1; |
+-------------------------------------------------------------------------------------------------------+
sql>
## Comment 12286
Date: 2009-08-28 19:04:55 +0200
From: Matthew W. Jones <<matburt>>
Selecting the next value seems to do the right thing:
sql>select next value for my_other_test_seq ;
+------------------------------+
| next_value_for_datawarehouse |
+==============================+
| 2147483649 |
+------------------------------+
1 tuple
sql>explain select next value for my_other_test_seq;
+-------------------------------------------------------------------------------------------------------+
| function user.s5_1{autoCommit=true}():void; |
| _4 := sql.next_value("datawarehouse","my_other_test_seq"); |
| sql.exportValue(1,".","next_value_for_datawarehouse","bigint",64,0,6,_4,""); |
| end s5_1; |
+-------------------------------------------------------------------------------------------------------+
sql>
## Comment 12287
Date: 2009-08-28 19:06:03 +0200
From: Matthew W. Jones <<matburt>>
It's also interesting that selecting next value uses the name of the schema instead of the name of the sequence when selecting the next value.
## Comment 12288
Date: 2009-09-03 22:38:08 +0200
From: @njnes
Fixed by properly handling long integers in the parsers 'restart with' code.
## Comment 12289
Date: 2009-09-03 22:38:39 +0200
From: @njnes
added test restart_with_large_int.SF-2846238.sql
## Comment 12290
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2846238 at http://sourceforge.net/support/tracker.php?aid=2846238
| alter sequence restart on large integer | https://api.github.com/repos/MonetDB/MonetDB/issues/2171/comments | 0 | 2020-11-30T10:25:39Z | 2024-06-27T11:41:20Z | https://github.com/MonetDB/MonetDB/issues/2171 | 753,352,263 | 2,171 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-28 11:07:08 +0200
From: Henning Rode <<hrode>>
To: GDK devs <<bugs-common>>
Version: -- development
Last updated: 2011-09-07 11:27:18 +0200
## Comment 12277
Date: 2009-08-28 11:07:08 +0200
From: Henning Rode <<hrode>>
running the following query:
let $doc1 := doc("/net/asbjorn/export/scratch0/henning/collections/trecvid09/mastershots/processed3/441.mp7.xml")
let $doc2 := doc("/net/asbjorn/export/scratch0/henning/collections/trecvid09/concepts_mixed/concepts_sorted.xml")
let $rsc := exactly-one($doc1//*:resource)
let $cc := zero-or-one($doc2//video[@uri = $rsc/@uri])//c
return
<resource>
{for $seg in $rsc/segment
let $cc_shot := $cc[@shotID = $seg/@uri]
return
<segment>
{for $c in $cc_shot return
<descr>
<concept>{string($c/@name)}</concept>
<score>{string($c/@prob)}</score>
</descr>
}
</segment>
}
</resource>
i get the error message:
ERROR = !ERROR: merged_union: BAT/column 6 of table 5 must have a dense head.
!ERROR: CMDmulti_merged_union: operation failed.
unfortunately, the error only occurs when using the files specified in the beginning of the query. when people outside CWI try to fix this bug, i can provide the files on request.
## Comment 12278
Date: 2009-08-28 13:42:44 +0200
From: @drstmane
Henning,
could you please check and report whether the problem also occurs with property checking enabled, i.e., when starting Mserver with -d10 ?
If this prevent the error, the problem is most probably a not correctly set denseness property "somewhere" in some code that creats a BAT ...
## Comment 12279
Date: 2009-08-28 13:53:43 +0200
From: Henning Rode <<hrode>>
OK, good guess. It is working correctly when starting the Mserver with -d10.
## Comment 12280
Date: 2009-09-04 14:45:27 +0200
From: @drstmane
Well, then we'll need to find which function fails to set the denseness property on a BAT it creates, while merged_union as a ("transitive") consumer of that BAT expects the denseness ...
## Comment 12281
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2846126 at http://sourceforge.net/support/tracker.php?aid=2846126
The original assignee of this bug does not have
an account here. Reassigning to the default assignee
for the component, bugs-common@monetdb.org.
Previous assignee was nobody@users.sourceforge.net.
## Comment 16191
Date: 2011-09-07 11:27:18 +0200
From: @sjoerdmullender
Development on MonetDB4 and MonetDB/XQuery has ceased. It is unlikely that
this bug will be fixed.
The last release of MonetDB/XQuery and MonetDB4 is Mar2011.
| CMDmulti_merged_union: operation failed | https://api.github.com/repos/MonetDB/MonetDB/issues/2170/comments | 0 | 2020-11-30T10:25:35Z | 2024-06-28T13:23:52Z | https://github.com/MonetDB/MonetDB/issues/2170 | 753,352,221 | 2,170 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-26 05:04:38 +0200
From: Matthew W. Jones <<matburt>>
To: clients devs <<bugs-clients>>
Version: -- development
CC: @gijzelaerr
Last updated: 2010-05-13 13:56:54 +0200
## Comment 12272
Date: 2009-08-26 17:04:38 +0200
From: Matthew W. Jones <<matburt>>
Attempting to install the monetdb python client the build/installation fails:
python setup.py install
[...]
byte-compiling /usr/lib/python2.5/site-packages/monetdb/mapi.py to mapi.pyc
byte-compiling /usr/lib/python2.5/site-packages/monetdb/mapi3.py to mapi3.pyc
File "/usr/lib/python2.5/site-packages/monetdb/mapi3.py", line 84
except socket.error as error:
^
SyntaxError: invalid syntax
byte-compiling /usr/lib/python2.5/site-packages/monetdb/monetdb_exceptions.py to monetdb_exceptions.pyc
[...]
## Comment 12273
Date: 2009-08-26 17:06:19 +0200
From: Matthew W. Jones <<matburt>>
except EXCEPTION as other
is supported in Python 3.0 but not 2.5
## Comment 12274
Date: 2009-08-26 17:11:51 +0200
From: Matthew W. Jones <<matburt>>
There are several places in this file where that syntax is used.
## Comment 12275
Date: 2009-08-26 22:41:11 +0200
From: @sjoerdmullender
The file mapi3.py is the version specifically for Python 3.X, so it is not so strange that it uses Python 3.X syntax. But it should be possible to install the package using Python 2.X.
Assigning to Gijs to come up with a solution.
## Comment 12276
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2844980 at http://sourceforge.net/support/tracker.php?aid=2844980
## Comment 14020
Date: 2010-05-13 13:56:54 +0200
From: @gijzelaerr
This has been fixed a long time ago, didn't see this bugreport.
| Python client install error | https://api.github.com/repos/MonetDB/MonetDB/issues/2169/comments | 0 | 2020-11-30T10:25:32Z | 2024-06-27T11:41:18Z | https://github.com/MonetDB/MonetDB/issues/2169 | 753,352,192 | 2,169 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-25 02:32:35 +0200
From: @grobian
To: Merovingian devs <<bugs-merovingian>>
Version: -- development
Last updated: 2010-05-04 13:07:01 +0200
## Comment 12269
Date: 2009-08-25 14:32:35 +0200
From: @grobian
the mero_port option set in the monetdb5.conf file is being ignored by merovingian, and as a result it always defaults to using port 50000
## Comment 12270
Date: 2009-08-25 14:34:51 +0200
From: @grobian
problem fixed, will be delivered in the next release
## Comment 12271
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2844202 at http://sourceforge.net/support/tracker.php?aid=2844202
## Comment 13988
Date: 2010-05-04 13:07:01 +0200
From: @grobian
Fixed in Aug2009-SP1 release.
| Merovingian: mero_port configuration option ignored | https://api.github.com/repos/MonetDB/MonetDB/issues/2168/comments | 0 | 2020-11-30T10:25:29Z | 2024-06-27T11:41:17Z | https://github.com/MonetDB/MonetDB/issues/2168 | 753,352,144 | 2,168 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-19 03:47:52 +0200
From: @drstmane
To: @njnes
Version: -- development
Last updated: 2009-09-20 09:08:51 +0200
## Comment 12266
Date: 2009-08-19 15:47:52 +0200
From: @drstmane
some SQL tests --- a fixed set, but I could not recognize the pattern, yet --- fail on our only big-endian machine (Sun T1000) with
"Cannot access descriptor",
i.e., a BATdescriptor() call fails to derive a BAT pointer for a given BAT id:
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.32.32.d.1-SunOS5.10/src_test/add_column_with_seq.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.32.32.d.1-SunOS5.10/src_test_BugTracker/multivalue_insert.SF-1509864.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.32.32.d.1-SunOS5.10/src_test_BugTracker/limit_timestampz.SF-1774217.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.32.32.d.1-SunOS5.10/src_test_BugTracker/proc_insert_into.SF-1958572.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.32.32.d.1-SunOS5.10/src_test_BugTracker-2009/distinct.SF-2546483.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.32.32.d.1-SunOS5.10/src_test_BugTracker-2009/serial_with_order_by_not_working.SF-2585592.err.00.html
The same hold for the CWI-internal test ViaM/test/Tests/create.sql.py, although the error message differs slightly:
ERROR = !MALException:algebra.uselect:Object not found
I'm actually not sure, whether the bug is in SQL or M5 ... or elsewhere ... ?
## Comment 12267
Date: 2009-09-19 22:57:54 +0200
From: @njnes
fixed a problem in bat_next_value_for (use lng instead of bat), then problems should be related to sequence numbers
## Comment 12268
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2840310 at http://sourceforge.net/support/tracker.php?aid=2840310
| SQL: "Cannot access descriptor" on big endian | https://api.github.com/repos/MonetDB/MonetDB/issues/2167/comments | 0 | 2020-11-30T10:25:24Z | 2024-06-27T11:41:16Z | https://github.com/MonetDB/MonetDB/issues/2167 | 753,352,085 | 2,167 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-19 03:42:57 +0200
From: Jan-Mark <<jmswams>>
To: @mlkersten
Version: -- development
Last updated: 2011-10-27 15:11:59 +0200
## Comment 12257
Date: 2009-08-19 15:42:57 +0200
From: Jan-Mark <<jmswams>>
If I run the install script and than the plain vanilla embedded server example (from http://monetdb.cwi.nl/projects/monetdb//MonetDB/Documentation/A-Simple-Example.html) it crashes. Below the "screen dump." As you can see the file sql_types.c line 207 contains a dereference of a variable named 'types' and it's current value is NULL. Since the install script does not leave me with the source code, I can't really look at the source code. But the bigger question is: "Is the example code in the daily test set, or in any test set?" because I think it should.
(I ran the install script with these parameters: ./monetdb-install.sh --enable-sql --cvs=May2009-SP2 --enable-debug --enable-testing)
I installed on both 2.6.28-11-generic 42-Ubuntu SMP as well as FreeBSD. Both have the same problem. (Note you do have to build a non generic kernel voor FreeBSD to run MonetDB.)
My guess, is a test for NULL is needed. Probably something that is not needed in the non-embedded version.
Regards Jan-Mark
Here is a "screen dump."
$ gdb embedtest2
GNU gdb 6.8-debian
Copyright (C) 2008 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. Type "show copying"
and "show warranty" for details.
This GDB was configured as "i486-linux-gnu"...
(gdb) run
Starting program: /home/jms/src/RaxUnixServer/embedtest2
[Thread debugging using libthread_db enabled]
[New Thread 0xb6e3e8d0 (LWP 15588)]
[New Thread 0xb6e3db90 (LWP 15591)]
[New Thread 0xb20f6b90 (LWP 15592)]
[Thread 0xb6e3db90 (LWP 15591) exited]
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0xb20f6b90 (LWP 15592)]
0xb7eeb886 in sql_find_subtype (res=0xb20f6268, name=0xb21f0b00 "int",
digi1970-01-01 01:00:00 +0100, scale=0) at sql_types.c:207
207 for (n = types->h; n; n = n->next) {
(gdb) print types
$1 = (list *) 0x0
## Comment 12258
Date: 2009-08-19 16:14:16 +0200
From: @drstmane
Hi Jan-Mark,
thanks for your report, we will double-check our embedded example as soon as time allows us to do so.
Til then, a few remarks:
- The MonetDB source are available for download from http://monetdb.cwi.nl/downloads/sources/
- Your install command might not do what you expect/intend to do:
According to the documentation (see `monetdb-install.sh --help`), the --cvs switch does not take any arguments, but rather installs " checkout a CVS snapshot of the current branch" --- in MonetDB lingo, "current (branch)" usually refers to the CVS HEAD (aka., development trunk), not the (latest) release branch. Hence, I assume that your "May2009-SP2" argument is (silently instead of issuing a warnig or error) ignored, and you installed a CVS snapshot instead of a released version of MonetDB --- you can check this by running `mserver5 --version`: v5.12.4 -> May2009-SP2; v5.15.0 "Current" CVS development snapshot.
To install the May2009-SP2 release, you'd (e.g.) need to download the respective "SuperSourceBall" from http://monetdb.cwi.nl/downloads/sources/May2009-SP2/MonetDB-May2009-SuperBall-SP2.tar.bz2, unpack it, and run the extracted monetdb-install.sh without --cvs option.
Stefan
## Comment 12259
Date: 2009-08-19 16:54:00 +0200
From: @grobian
from monetdb-install.sh --help
--cvs[=tag] checkout a CVS snapshot of the current branch or
optionally given tag
:)
On request of loredana we added this, as this is possible since we use unified tags over all our repos
## Comment 12260
Date: 2009-08-19 16:56:08 +0200
From: @grobian
> Since the install script does not leave me
> with the source code, I can't really look at the source code
the installscript leaves /var/tmp/MonetDB-XXXXXXXXX behind containing the build tree, and the build log
## Comment 12261
Date: 2009-08-19 17:36:41 +0200
From: @drstmane
mea cupla --- I didn't recall that feature and accidently check the Feb2009 version --- handling too many versions in parallel for tracing historical performance and stability behavior ...
However, then the argument must be "May2009" (to get the head of the May2009 branch with by now in the code base of the May2009-Sp2 release, plus any bug-fixes that might have been applied after that release, or "May2009_4" to get the actual code base of the May2009-SP2 release ("4" is the <release> part of the <major>.<minor>.<release> version numbers any "-SP2" release).
There is no "May2009-SP2" CVS tag.
## Comment 12262
Date: 2009-08-26 14:53:35 +0200
From: Jan-Mark <<jmswams>>
FYI
$ mserver5 --version
MonetDB server v5.14.0 (32-bit), based on kernel v1.32.0 (32-bit oids)
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2009 MonetDB B.V., all rights reserved
Visit http://monetdb.cwi.nl/ for further information
Configured for prefix: /home/jms/MonetDB
Libraries:
libpcre: 7.8 2008-09-05 (compiled with 7.8)
openssl: OpenSSL 0.9.8g 19 Oct 2007 (compiled with OpenSSL 0.9.8g 19 Oct 2007)
Compiled by: jms@columnstoresrv001 (i686-pc-linux-gnu)
Compilation: gcc -g
Linking : /usr/bin/ld
## Comment 12263
Date: 2009-09-03 23:26:50 +0200
From: @njnes
I agree this should be added to the testing.
## Comment 12264
Date: 2009-11-11 19:48:11 +0100
From: @njnes
New status is
niels@alf::/tmp> myprog
MAPI = monetdb@localhost:-1
ACTION= mapi_start_talking
ERROR = !SQLException:optimizer:Optimizer 'default_pipe' does not exist, use default pipe instead
## Comment 12265
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2840309 at http://sourceforge.net/support/tracker.php?aid=2840309
## Comment 16484
Date: 2011-10-27 15:11:59 +0200
From: @sjoerdmullender
We don't build the embedded examples anymore, and moreover, necessary code to compile those examples have been removed in changeset [d9fa2eb57ae2](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d9fa2eb57ae2) and [52cf8254e285](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=52cf8254e285)
| Embedded example won't run. | https://api.github.com/repos/MonetDB/MonetDB/issues/2166/comments | 0 | 2020-11-30T10:25:20Z | 2024-06-28T13:23:51Z | https://github.com/MonetDB/MonetDB/issues/2166 | 753,352,049 | 2,166 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-19 01:05:40 +0200
From: @MarcinZukowski
To: @sjoerdmullender
Version: -- development
Last updated: 2009-10-07 12:27:05 +0200
## Comment 12252
Date: 2009-08-19 13:05:40 +0200
From: @MarcinZukowski
Try the following file
@f z
@= i_do_exist
I DO EXIST
@= i_do_not_exist
I DO NOT EXIST
@c
@:i_do_exist@
@:i_dont_exist@
Then running:
$ Mx -c z.mx
Mx:Unresolved reference:i_dont_exist[z.mx:11].
z.mx: ./z.c - created
What happens is that z.c is created even though the error appeared. As a result, when compiling first with "make" and then with "make -k", the z.c file exists, making the compilation continue. Still, z.c is incomplete, leading to weird effects.
Mx (like any other compiler) should not generate output files if it detects an error condition
## Comment 12253
Date: 2009-10-02 10:42:09 +0200
From: @sjoerdmullender
I just checked in a fix in the Aug2009 stable branch. Can you check?
With this fix, the output file is not touched when there was an error (i.e. if it existed already, it is kept).
## Comment 12254
Date: 2009-10-07 11:29:29 +0200
From: @MarcinZukowski
Works for me, thanks!
## Comment 12255
Date: 2009-10-07 12:27:01 +0200
From: @sjoerdmullender
No tests for Mx issues.
## Comment 12256
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2840225 at http://sourceforge.net/support/tracker.php?aid=2840225
| Mx creates output files even on errors | https://api.github.com/repos/MonetDB/MonetDB/issues/2165/comments | 0 | 2020-11-30T10:25:16Z | 2024-06-27T11:41:14Z | https://github.com/MonetDB/MonetDB/issues/2165 | 753,351,999 | 2,165 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-18 03:01:33 +0200
From: @skinkie
To: @njnes
Version: -- development
Last updated: 2009-11-22 10:24:30 +0100
## Comment 12248
Date: 2009-08-18 15:01:33 +0200
From: @skinkie
create table kvk (kvks integer);
I have filled it with about 2.4mil rows.
Executing the following query fails:
select distinct rank() over (order by kvks) as rank_number, kvks from (select kvks from kvk) as x limit 10;
Trace:
>sql.start(); >trace select distinct rank() over (order by kvks) as rank_number, kvks from (select kvks from kvk) as x limit 10;
[ 9457 usec mdb.setTimer(true); ]
[ 12 usec @0 mdb.setThread(true); ]
[ 31 usec @0 barrier _69 := language.dataflow(); ]
[ 144 usec @0 _2:bat[:oid,:int] <tmp_2652>[2414680] := sql.bind("sys","kvk","kvks",0); ]
[ 51 usec @0 _7:bat[:oid,:int] <tmp_5163>[0] := sql.bind("sys","kvk","kvks",2); ]
[ 377 usec @0 _9<tmp_5506>[2414680] := algebra.kdifference(_2=<tmp_2652>:bat[:oid,:int][2414680],_7=<tmp_5163>:bat[:oid,:int][0]); ]
[ 15 usec @0 _2:bat[:oid,:int] := nil:BAT; ]
[ 23 usec @0 _10<tmp_5507>[2414680] := algebra.kunion(_9=<tmp_5506>[2414680],_7=<tmp_5163>:bat[:oid,:int][0]); ]
[ 18 usec @0 _9 := nil:BAT; ]
[ 2 usec @0 _7:bat[:oid,:int] := nil:BAT; ]
[ 91 usec @0 _11:bat[:oid,:int] <tmp_5164>[0] := sql.bind("sys","kvk","kvks",1); ]
[ 28 usec @0 _13<tmp_5506>[2414680] := algebra.kunion(_10=<tmp_5507>[2414680],_11=<tmp_5164>:bat[:oid,:int][0]); ]
[ 8 usec @0 _10 := nil:BAT; ]
[ 2 usec @0 _11:bat[:oid,:int] := nil:BAT; ]
[ 13 usec @0 _14:bat[:oid,:oid] <tmp_5166>[0] := sql.bind_dbat("sys","kvk",1); ]
[ 130 usec @0 _15<tmpr_5166>[0] := bat.reverse(_14=<tmp_5166>:bat[:oid,:oid][0]); ]
[ 3 usec @0 _14:bat[:oid,:oid] := nil:BAT; ]
[ 14 usec @0 _16<tmp_5507>[2414680] := algebra.kdifference(_13=<tmp_5506>[2414680],_15=<tmpr_5166>[0]); ]
[ 6 usec @0 _13 := nil:BAT; ]
[ 2 usec @0 _15 := nil:BAT; ]
[ 622740 usec @0 _17<tmpr_5506>[2414680] := algebra.sortTail(_16=<tmp_5507>[2414680]); ]
[ 2582255 usec @0 return rank_grp<tmp_5510>[2414680] := sql.rank(b=<tmpr_5506>[2414680]); ]
[ 2582584 usec @0 _18:bat[:oid,:int] <tmp_5510>[2414680] := batcalc.rank_grp(_17=<tmpr_5506>[2414680]); ]
[ 44 usec @0 _17 := nil:BAT; ]
[ 1658133 usec @0 (ext23<tmp_5511>[1916813],grp21<tmp_5512>[2414680]) := group.new(_18=<tmp_5510>:bat[:oid,:int][2414680]); ]
Explain:
explain select distinct rank() over (order by kvks) as rank_number, kvks from (select kvks from kvk) as x limit 10;
+---------------------------------------------------------------------+
| function user.s1_1{autoCommit=true}():void; |
| barrier _69 := language.dataflow(); |
| _2:bat[:oid,:int] := sql.bind("sys","kvk","kvks",0); |
| _7:bat[:oid,:int] := sql.bind("sys","kvk","kvks",2); |
| _9 := algebra.kdifference(_2,_7); |
| _2:bat[:oid,:int] := nil:BAT; |
| _10 := algebra.kunion(_9,_7); |
| _9 := nil:BAT; |
| _7:bat[:oid,:int] := nil:BAT; |
| _11:bat[:oid,:int] := sql.bind("sys","kvk","kvks",1); |
| _13 := algebra.kunion(_10,_11); |
| _10 := nil:BAT; |
| _11:bat[:oid,:int] := nil:BAT; |
| _14:bat[:oid,:oid] := sql.bind_dbat("sys","kvk",1); |
| _15 := bat.reverse(_14); |
| _14:bat[:oid,:oid] := nil:BAT; |
| _16 := algebra.kdifference(_13,_15); |
| _13 := nil:BAT; |
| _15 := nil:BAT; |
| _17 := algebra.sortTail(_16); |
| _18:bat[:oid,:int] := batcalc.rank_grp(_17); |
| _17 := nil:BAT; |
| (ext23,grp21) := group.new(_18); |
| (ext26,grp24) := group.derive(ext23,grp21,_16); |
| grp24 := nil:BAT; |
| ext23 := nil:BAT; |
| grp21 := nil:BAT; |
| _23 := bat.mirror(ext26); |
| ext26 := nil:BAT; |
| _24 := algebra.join(_23,_18); |
| _18:bat[:oid,:int] := nil:BAT; |
| _25 := algebra.slice(_24,0,9); |
| _24 := nil:BAT; |
| _26 := algebra.join(_23,_16); |
| _23 := nil:BAT; |
| _16 := nil:BAT; |
| _27 := algebra.semijoin(_26,_25); |
| _26 := nil:BAT; |
| exit _69; |
| _28 := sql.resultSet(2,1,_25); |
| sql.rsColumn(_28,".","rank_number","int",32,0,_25); |
| _25 := nil:BAT; |
| sql.rsColumn(_28,"sys.x","kvks","int",32,0,_27); |
| _27 := nil:BAT; |
| _36 := io.stdout(); |
| sql.exportResult(_36,_28); |
| end s1_1; |
+---------------------------------------------------------------------+
Plan:
plan select distinct rank() over (order by kvks) as rank_number, kvks from (select kvks from kvk) as x limit 10;
+------------------------------------------------------------------------------+
| top N ( |
| | distinct project ( |
| | | project ( |
| | | | table(sys.kvk) [ kvk.kvks, kvk.%TID% NOT NULL ] |
| | | ) [ kvk.kvks as x.kvks, kvk.%TID% NOT NULL ] |
| | ) [ rank(x.kvks ASC) as rank_number, x.kvks ] |
| ) [ 10 ] |
+------------------------------------------------------------------------------+
## Comment 12249
Date: 2009-09-20 21:14:51 +0200
From: @njnes
This bug can only be reproduced using the data.
## Comment 12250
Date: 2009-11-22 22:24:29 +0100
From: @njnes
The generated code is correct. We did fix a problem in grp_mark recently which probably improved the performance
of the distinct (grp.derive) statement. As the test is large we cannot add this to the nightly testing.
## Comment 12251
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2839602 at http://sourceforge.net/support/tracker.php?aid=2839602
| SQL: distinct in rank query causes troubles | https://api.github.com/repos/MonetDB/MonetDB/issues/2164/comments | 0 | 2020-11-30T10:25:13Z | 2024-06-28T13:23:50Z | https://github.com/MonetDB/MonetDB/issues/2164 | 753,351,952 | 2,164 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-17 07:00:35 +0200
From: @drstmane
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2009-08-19 08:35:55 +0200
## Comment 12243
Date: 2009-08-17 19:00:35 +0200
From: @drstmane
With BAT property checking (re-)enabled in the M5 MAL interpreter,
the following SQL tests fail with property errors:
sql/src/test/BugDay_2005-10-06_2.9.3/Tests/case.SF-929151.sql:
"
QUERY = select case i
ERROR = !MALException:mal.propertyCheck:Object not found
!ERROR: BATpropcheck: BAT tmpr_2432(-1306)[str,void] with 5 tuples was incorrectly marked nonil!
"
sql/src/odbc/samples/Tests/odbcsample1.SQL.sh:
"
SQLColumns: Error: SQLstate HY000, Errnr 0, Message [MonetDB][ODBC Driver 1.0]!MALException:mal.propertyCheck:Object not found
"
## Comment 12244
Date: 2009-08-18 18:09:30 +0200
From: @drstmane
Nightly testing revealed more SQL test that trigger BAT property checking errors,
all of them complaining about
!ERROR: BATpropcheck: BAT ... was incorrectly marked nonil!
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.64.d.1-Fedora10/src_odbc_samples/odbcsample1.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.64.d.1-Fedora10/src_test_BugDay_2005-10-06_2.9.3/case.SF-929151.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.64.d.1-Fedora10/src_jdbc_tests/Test_JdbcClient.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.64.d.1-Fedora10/src_test_BugDay_2005-10-06_2.9.3/alter_table_describe.SF-1146092.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.64.d.1-Fedora10/src_test_BugDay_2005-10-06_2.9.3/not_null.SF-933194.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.64.d.1-Fedora10/src_test_BugDay_2005-10-06_2.9.3/type_dump_test.SF-989257.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.32.32.d.1-Fedora10/src_test_BugTracker-2008/insert_into_unique_column.SF-2075180.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.32.d.1-Fedora10/src_test_BugDay_2005-10-06_2.9.3/div_by_zero.SF-987304.err.00.html
## Comment 12245
Date: 2009-08-18 23:17:57 +0200
From: @drstmane
Hopefully mostly fixed by
===================================================================
2009/08/18 - stmane: MonetDB5/src/modules/kernel/batcalc.mx,1.181.2.3
MonetDB5/src/modules/kernel/batcast.mx,1.27.4.1
unset nonil whenever we insert a nil into the result
(and only then);
this fixes some of the "!ERROR: BATpropcheck: BAT ... was incorrectly marked nonil!"
reported in ID: 2839103 "SQL: property errors: BAT incorrectly marked nonil"
https://sourceforge.net/tracker/?func=detail&aid=2839103&group_id=56967&atid=482468
===================================================================
2009/08/18 - stmane: MonetDB5/src/modules/kernel/batifthen.mx,1.24.4.1
fixed nil-check for strings
this fixes some more of the "!ERROR: BATpropcheck: BAT ... was incorrectly marked nonil!"
reported in ID: 2839103 "SQL: property errors: BAT incorrectly marked nonil"
https://sourceforge.net/tracker/?func=detail&aid=2839103&group_id=56967&atid=482468
===================================================================
tonight's testing will reveal the details ...
## Comment 12246
Date: 2009-08-19 08:35:54 +0200
From: @drstmane
testing confirmed fixes.
closing.
## Comment 12247
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2839103 at http://sourceforge.net/support/tracker.php?aid=2839103
| SQL: property errors: BAT incorrectly marked nonil | https://api.github.com/repos/MonetDB/MonetDB/issues/2163/comments | 0 | 2020-11-30T10:25:09Z | 2024-06-27T11:41:12Z | https://github.com/MonetDB/MonetDB/issues/2163 | 753,351,908 | 2,163 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-17 05:51:34 +0200
From: @drstmane
To: @mlkersten
Version: -- development
Last updated: 2009-09-25 11:51:14 +0200
## Comment 12239
Date: 2009-08-17 17:51:34 +0200
From: @drstmane
With BAT property checking (re-)enabled in the M5 MAL interpreter,
test MonetDB5/src/modules/mal/Tests/chopper03.mal now fails with
!FATAL: BATpropcheck: BAT tmpr_12(-10) has inconsistent descriptor 65536 (200000)
65536 dec (200000 oct) is 1<<14, hence,
"
((b->H->type && (b->U->capacity > (b->H->heap.size>>b->H->shift))) |
(b->T->type && (b->U->capacity > (b->T->heap.size>>b->T->shift))))
"
## Comment 12240
Date: 2009-08-18 18:01:35 +0200
From: @drstmane
for completeness and reference:
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/MonetDB5/.mTests103/GNU.64.64.d.1-Fedora10/src_modules_mal/chopper03.err.00.html
## Comment 12241
Date: 2009-09-25 11:51:14 +0200
From: @drstmane
The problem seems to be fixed by
===================================================================
2009/09/04 - mlkersten: MonetDB/src/gdk/gdk.mx,1.287.2.5
MonetDB/src/gdk/gdk_align.mx,1.103.2.3
Properly setting the bounds of an existing BAT view object.
===================================================================
2009/09/04 - mlkersten: MonetDB5/src/modules/mal/chopper.mx,1.63.4.1
MonetDB5/src/modules/mal/tablet.mx,1.196.2.3
MonetDB5/src/modules/mal/Tests/chopper03.mal,1.7.20.1
Properly setting the VIEW bounds and handling final stage of iterator.
===================================================================
2009/09/25 - stmane: MonetDB5/src/modules/mal/Tests/chopper03.mal,1.7.20.2
undoing
"
2009/09/04 - mlkersten: MonetDB5/src/modules/mal/Tests/chopper03.mal,1.7.20.1
Properly setting the VIEW bounds and handling final stage of iterator.
"
to re-activate test
===================================================================
## Comment 12242
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2839063 at http://sourceforge.net/support/tracker.php?aid=2839063
| M5: property error in src/modules/mal/Tests/chopper03.mal | https://api.github.com/repos/MonetDB/MonetDB/issues/2162/comments | 0 | 2020-11-30T10:25:06Z | 2024-06-27T11:41:11Z | https://github.com/MonetDB/MonetDB/issues/2162 | 753,351,876 | 2,162 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-14 02:54:09 +0200
From: Wouter Alink <<alink>>
To: Fabian Groffen <<fabian>>
Version: -- development
Last updated: 2009-08-16 10:49:41 +0200
## Comment 12234
Date: 2009-08-14 14:54:09 +0200
From: Wouter Alink <<alink>>
When adding a user in a transaction... it is created regardless of the transaction outcome. (see example below)
(on the May2009 branch). I assume it should obey the transaction outcome.
$ mclient -lsql -dtest3
sql>START TRANSACTION;
0 tuples
sql>CREATE USER "aap" WITH PASSWORD 'aap' NAME 'AAP user' SCHEMA "aap";
0 tuples
sql>ROLLBACK;
0 tuples
sql>CREATE USER "aap" WITH PASSWORD 'aap' NAME 'AAP user' SCHEMA "aap";
!CREATE USER: user 'aap' already exists
0 tuples
sql>
## Comment 12235
Date: 2009-08-14 14:58:18 +0200
From: Wouter Alink <<alink>>
oops, replaced to many values in the given example, the example below works on a clean database (using schema 'sys' instead of 'aap')
START TRANSACTION;
CREATE USER "aap" WITH PASSWORD 'aap' NAME 'AAP user' SCHEMA "sys";
ROLLBACK;
CREATE USER "aap" WITH PASSWORD 'aap' NAME 'AAP user' SCHEMA "sys";
## Comment 12236
Date: 2009-08-16 10:32:56 +0200
From: @njnes
users are m5 objects, ie not controlled by m5/sql. As users are outside the sql standard we can consider this a feature.
## Comment 12237
Date: 2009-08-16 10:49:41 +0200
From: @grobian
agreed
## Comment 12238
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2837594 at http://sourceforge.net/support/tracker.php?aid=2837594
| SQL: "create user" not transactional | https://api.github.com/repos/MonetDB/MonetDB/issues/2161/comments | 0 | 2020-11-30T10:25:02Z | 2024-06-28T13:23:50Z | https://github.com/MonetDB/MonetDB/issues/2161 | 753,351,838 | 2,161 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-14 01:42:04 +0200
From: Wouter Alink <<alink>>
To: @njnes
Version: -- development
Last updated: 2009-08-16 04:24:41 +0200
## Comment 12231
Date: 2009-08-14 13:42:04 +0200
From: Wouter Alink <<alink>>
The following query sequence makes the server crash on the May2009 branch:
CREATE VIEW test AS SELECT 'none' AS score, 'none' AS avalue, 'none' AS displayname;
SELECT avalue FROM test GROUP BY avalue;
The debugger says:
(gdb) bt
0 0x0000000000750077 in ?? ()
1 0x00007f2b36dcf675 in ATOMlen (t=255, src=0x7f2b240f9c28) at /home/alink/workspace/MonetDB/MonetDB/src/gdk/gdk_atoms.mx:960
2 0x00007f2b36dcfc20 in ATOMdup (t=255, p=0x7f2b240f9c28) at /home/alink/workspace/MonetDB/MonetDB/src/gdk/gdk_atoms.mx:1064
3 0x00007f2b247bdff8 in res_col_create (tr=0x7f2b24140aa8, t=0x7f2b23ae7b68, tn=0x7f2b23af8f98 ".find_NE_from_DOC_2_RESULT_facet7",
name=0x7f2b23af3be8 "score", typename=0x7f2b23af3848 "char", digi1970-01-01 01:00:04 +0100, scale=0, mtype=255, val=0x7f2b240f9c28)
at /home/alink/workspace/MonetDB/sql/src/storage/bat/res_table.mx:88
4 0x00007f2b2473e21d in mvc_result_value (m=0x7f2b24145b38, tn=0x7f2b23af8f98 ".find_NE_from_DOC_2_RESULT_facet7", name=0x7f2b23af3be8 "score",
typename=0x7f2b23af3848 "char", digi1970-01-01 01:00:04 +0100, scale=0, p=0x7f2b240f9c28, mtype=255)
at /home/alink/workspace/MonetDB/sql/src/backends/monet5/sql_result.mx:1215
5 0x00007f2b246f3beb in mvc_result_value_wrap (cntxt=0x604a20, mb=0x7f2b240fa2e8, stk=0x7f2b240f9a98, pci=0x7f2b23af38c8)
at /home/alink/workspace/MonetDB/sql/src/backends/monet5/sql.mx:2067
6 0x0000000000970fb4 in runMALsequence (cntxt=0x604a20, mb=0x7f2b240fa2e8, startpc=1, stoppc=0, stk=0x7f2b240f9a98, env=0x0, pcicaller=0x0)
at /home/alink/workspace/MonetDB/MonetDB5/src/mal/mal_interpreter.mx:1740
7 0x0000000000968d0d in callMAL (cntxt=0x604a20, mb=0x7f2b240fa2e8, env=0x7f2b23d64da8, argv=0x7f2b23d64df0, debug=0 '\0')
at /home/alink/workspace/MonetDB/MonetDB5/src/mal/mal_interpreter.mx:394
8 0x00007f2b2472d498 in SQLexecutePrepared (c=0x604a20, be=0x7f2b23a6cfc8, q=0x7f2b23ae6768)
at /home/alink/workspace/MonetDB/sql/src/backends/monet5/sql_scenario.mx:1190
9 0x00007f2b2472d707 in SQLengineIntern (c=0x604a20, be=0x7f2b23a6cfc8)
at /home/alink/workspace/MonetDB/sql/src/backends/monet5/sql_scenario.mx:1243
10 0x00007f2b2472dbd1 in SQLengine (c=0x604a20) at /home/alink/workspace/MonetDB/sql/src/backends/monet5/sql_scenario.mx:1353
11 0x00000000009aa07d in runPhase (c=0x604a20, phase=4) at /home/alink/workspace/MonetDB/MonetDB5/src/mal/mal_scenario.mx:602
12 0x00000000009aa20a in runScenarioBody (c=0x604a20) at /home/alink/workspace/MonetDB/MonetDB5/src/mal/mal_scenario.mx:647
13 0x00000000009aa3a1 in runScenario (c=0x604a20) at /home/alink/workspace/MonetDB/MonetDB5/src/mal/mal_scenario.mx:672
14 0x0000000000966cf1 in MSserveClient (dummy=0x604a20) at /home/alink/workspace/MonetDB/MonetDB5/src/mal/mal_session.mx:493
15 0x00000036f06073da in start_thread () from /lib64/libpthread.so.0
16 0x00000036efae62bd in clone () from /lib64/libc.so.6
(gdb)
## Comment 12232
Date: 2009-08-16 16:24:41 +0200
From: @njnes
fixed we now make sure we always group on bats. Added test
group_by_on_values.SF-2837561.sql
## Comment 12233
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2837561 at http://sourceforge.net/support/tracker.php?aid=2837561
| SQL: group by on view with constants | https://api.github.com/repos/MonetDB/MonetDB/issues/2160/comments | 0 | 2020-11-30T10:24:59Z | 2024-06-27T11:41:09Z | https://github.com/MonetDB/MonetDB/issues/2160 | 753,351,792 | 2,160 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-13 08:06:19 +0200
From: @drstmane
To: @njnes
Version: -- development
Last updated: 2009-11-07 02:11:17 +0100
## Comment 12228
Date: 2009-08-13 20:06:19 +0200
From: @drstmane
test src/test/xquery/Tests/q01.sql fails since checkin on 2009/06/08 on the CVS HEAD;
cf.,
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.64.d.1-Fedora10/src_test_xquery/q01.out.00.html
## Comment 12229
Date: 2009-11-07 14:11:17 +0100
From: @drstmane
closing as the test seems to work fine now/again with the Nov2009 branch
## Comment 12230
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2837050 at http://sourceforge.net/support/tracker.php?aid=2837050
| SQL: src/test/xquery/Tests/q01.sql fails | https://api.github.com/repos/MonetDB/MonetDB/issues/2159/comments | 0 | 2020-11-30T10:24:55Z | 2024-06-27T11:41:08Z | https://github.com/MonetDB/MonetDB/issues/2159 | 753,351,754 | 2,159 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-13 01:55:04 +0200
From: Roy Walter <<gooseflight>>
To: Henning Rode <<hrode>>
Version: -- development
Last updated: 2009-09-15 04:47:47 +0200
## Comment 12224
Date: 2009-08-13 13:55:04 +0200
From: Roy Walter <<gooseflight>>
Use of the 'or' operator in the context of a tijah phrase query produces same row count as 'and'. For example:
tijah:queryall("//[about(., 'wind farm') or about(., 'wind turbine')]")
find only records where both phrases are present.
## Comment 12225
Date: 2009-08-14 13:53:51 +0200
From: @drstmane
tagged subject
## Comment 12226
Date: 2009-09-15 16:47:08 +0200
From: Henning Rode <<hrode>>
the issue is fixed (test case added).
i also fixed a not yet reported bug, related to phrase search: the previous join between query terms and term IDs changed the order of query terms in some cases, which caused incorrect results for phrase searches. this problem is fixed now as well.
## Comment 12227
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2836870 at http://sourceforge.net/support/tracker.php?aid=2836870
| PFtijah: Use of 'or' in phrase search yelds incorrect result | https://api.github.com/repos/MonetDB/MonetDB/issues/2158/comments | 0 | 2020-11-30T10:24:52Z | 2024-06-27T11:41:07Z | https://github.com/MonetDB/MonetDB/issues/2158 | 753,351,717 | 2,158 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-13 01:47:17 +0200
From: @drstmane
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2009-08-18 05:58:46 +0200
## Comment 12217
Date: 2009-08-13 13:47:17 +0200
From: @drstmane
Most probably since this checkin
===================================================================
2009/06/09 - nielsnes: sql/src/backends/monet5/sql_gencode.mx,1.319.2.4
sql/src/backends/monet5/sql_optimizer.mx,1.218.2.2
sql/src/server/rel_bin.mx,1.78.2.1
sql/src/server/sql_rel2bin.mx,1.132.2.1
sql/src/server/sql_statement.mx,1.184.2.1
ng problems with 'order' dependend statements (such as mkey xor and batcalc*
)
this solves problems with alter_table_drop_constraint as well as
SQL: test type_dump_test.SF-989257.SQL.sh fails since Feb 28
projection joins are now handled by leftjoins
===================================================================
order-by tests fail on 64-bit systems with 32-bit OIDs (the advised default used with our 64-bit Windows installers); cf.
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/Int.64.32.d.1-Windows5.2/src_benchmarks_ATIS/select_group.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.32.d.1-Fedora10/src_test_ADT2006/slam.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.32.d.1-Fedora10/src_test_Dependencies/Dependencies.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.32.d.1-Fedora10/src_test_UserDump/create.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.32.d.1-Fedora10/src_test_bugs/subselect_multiple_unionall_where_1=1-bug-sf-1005596.out.00.html
The problem might be related to these comments in the above checkin in sql/src/server/sql_rel2bin.mx
"
/* TODO the semijoin may break the order!! */
/* TODO semijoin may break order */
"
maybe, we need to introduce an order-preserving "left"-semijoin?
## Comment 12218
Date: 2009-08-16 23:27:58 +0200
From: @drstmane
A quick test indeed suggests that the kernel (GDK) seems to choose a "reversed" and hence not order-preserving implementation of semijoin, and this results in incorrect order --- at least disabling the reversed semijoin implementation does yield the correct results.
Hence, introducing (and using) an order-preserving "left"semijoin (basically a semijoin that is not allowed to internally use the "reversed" implementation that loops over the right iso. the left input) should help.
Open question is, why the problems only occur on 64-bit systems with 32-bit OIDs ...
## Comment 12219
Date: 2009-08-17 08:50:14 +0200
From: @drstmane
Unfortunately, the problem is even stranger:
Looking more detailed at the first failing test
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/Int.64
.32.d.1-Windows5.2/src_benchmarks_ATIS/select_group.out.00.html,
it appears that semijoin (BATsemijoin) is not even used; hence, what seemed to be a fix (see my prev. comment) was rather by coincidence that the test worked.
In fact, when run single-threaded (gdk_nr_threads=1), the test always fails (only on 64-bit systems with 32-bit OIDs); however when run multi-threaded (ghk_nr_thread>1) (on a multi-core system), the test seems to work randomly in roughtly 50% of the attempts ...
... I'm afraid it will be a nasty task to debug and eventually fix this one ... |-(
## Comment 12220
Date: 2009-08-17 09:18:32 +0200
From: @drstmane
Ok, running the failing query
select category,count(*) as totalnr from aircraft where engines=2 group by category having count(*)>4 order by category;
in isolation against a freshly restarted mserver5 (after loading the data) reveals that the first run gives the wrong order as in the test, but any subsequent run (against the same running mserver5) yield the correctly ordered result.
TRACEing the query and enabling algorithm tracing in the server (--algorithm) reveals as (only) significant difference between the 1st & 2nd run, that the 1st run creates some hash table before showing any other tracing output, while the 2nd does not:
BAThash: create hash(3);
... I cannot say that this clams me down, much ...
... to be continued ...
## Comment 12221
Date: 2009-08-18 00:10:36 +0200
From: @drstmane
Most probably fixed by
===================================================================
2009/08/17 - stmane: MonetDB/src/gdk/gdk_batop.mx,1.173.2.3
Added missing GDK_VARSHIFT for string heap access in BATordered() .
This should fix
ID: 2836863 "SQL: ORDER BY tests fail on 64-bit systems with 32-bit OIDs"
https://sourceforge.net/tracker/?func=detail&aid=2836863&group_id=56967&atid=482468
===================================================================
To be closed once testing has confirmed the fix.
## Comment 12222
Date: 2009-08-18 17:58:46 +0200
From: @drstmane
testing confirmed fix. closing.
## Comment 12223
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2836863 at http://sourceforge.net/support/tracker.php?aid=2836863
| SQL: ORDER BY tests fail on 64-bit systems with 32-bit OIDs | https://api.github.com/repos/MonetDB/MonetDB/issues/2157/comments | 0 | 2020-11-30T10:24:48Z | 2024-06-27T11:41:06Z | https://github.com/MonetDB/MonetDB/issues/2157 | 753,351,670 | 2,157 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-13 12:33:24 +0200
From: Roy Walter <<gooseflight>>
To: Fabian Groffen <<fabian>>
Version: -- development
Last updated: 2009-08-17 11:35:46 +0200
## Comment 12211
Date: 2009-08-13 12:33:24 +0200
From: Roy Walter <<gooseflight>>
A useful addition to the JBDC docs would be:
"To make a JDBC connection to a MonetDB server other than localhost, mclient sessions must be allowed from outside the host server. The following flag must therefore be set in monetdb.conf:
mapi_open=yes
The installation default is mapi_open=no."
## Comment 12212
Date: 2009-08-14 09:32:02 +0200
From: @grobian
Hi Roy,
Do you by chance have the URL where you found that phrase at hand?
On a side-note, when using merovingian, this problem does not occur.
## Comment 12213
Date: 2009-08-14 18:22:49 +0200
From: @sjoerdmullender
Assigned to mr. JDBC.
## Comment 12214
Date: 2009-08-14 19:19:10 +0200
From: Roy Walter <<gooseflight>>
Hi Fabian
The phrase is just my suggested addition.
To put it in context, I'm using the Windows version of MonetDB/XQuery so merovingian is not on my radar.
## Comment 12215
Date: 2009-08-17 11:35:46 +0200
From: @grobian
I patched the documentation with the mapi_open note. Updated documentation should appear at least with the next release.
## Comment 12216
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2836833 at http://sourceforge.net/support/tracker.php?aid=2836833
| documentation: connect JDBC other than localhost | https://api.github.com/repos/MonetDB/MonetDB/issues/2156/comments | 0 | 2020-11-30T10:24:45Z | 2024-06-27T11:41:05Z | https://github.com/MonetDB/MonetDB/issues/2156 | 753,351,637 | 2,156 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-12 06:16:23 +0200
From: John van Schie <<johnvanschie>>
To: Peter Boncz <<boncz>>
Version: -- development
Last updated: 2011-04-28 16:38:59 +0200
## Comment 12208
Date: 2009-08-12 18:16:23 +0200
From: John van Schie <<johnvanschie>>
Hi,
I'm using the May2009-SP2 FC10 x86_64 RPM release of MonetDB4/XQuery. The machine has 93GB free on disk and 32 GB RAM.
During the usage of this instance, we have encountered a segfault. I have included an backtrace of each thread from the crashing process and the tail of the query log. In this snippet (I have removed all queries that returned a value before the first query that didn't), you can see that just before the crash, we were trying to shred a document in a collection while another document was still being shred in the same collection.
Currently the process is still running in GDB, so if you need more information, please ask.
Cheers,
John
## Comment 12209
Date: 2009-08-14 13:57:40 +0200
From: @drstmane
tagged subject
## Comment 12210
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2836315 at http://sourceforge.net/support/tracker.php?aid=2836315
## Comment 15771
Date: 2011-04-28 16:38:59 +0200
From: @sjoerdmullender
Development on MonetDB4 and MonetDB/XQuery has ceased. It is unlikely that this bug will be fixed.
The last release of MonetDB/XQuery and MonetDB4 is Mar2011.
| XQ: Segfault in HEAPcopy () while shredding documents | https://api.github.com/repos/MonetDB/MonetDB/issues/2155/comments | 0 | 2020-11-30T10:24:41Z | 2024-06-28T13:23:49Z | https://github.com/MonetDB/MonetDB/issues/2155 | 753,351,604 | 2,155 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-12 04:35:29 +0200
From: @arjenderijke
To: @mlkersten
Version: -- development
Last updated: 2009-08-12 08:26:05 +0200
## Comment 12205
Date: 2009-08-12 16:35:29 +0200
From: @arjenderijke
I was doing a "copy into table from file" command in sql with a large file on a system where the partition with the dbfarm had insufficient diskspace. When the disc was full it reported the following in the merovingian.log file (8 times):
2009-08-10 21:38:05 MSG bestdr7[23245]: !OS: No space left on device
2009-08-10 21:38:05 MSG bestdr7[23245]: !ERROR: GDKsave: error -1 on: name=01/13
/11332, ext=theap.tmp, mode=0
But it did not stop the query. Instead it continued to run with a lot of swapping of data to and from the harddisc. It should have stopped the query and indicate to the client that an error had occured.
The code was a checkout of the current branch of 10 august 2009 on a Fedora 10 machine.
## Comment 12206
Date: 2009-08-12 20:26:05 +0200
From: @mlkersten
The COPY INTO was too optimistic. It did not check the result of BATextend, which meant that each new
tuple caused a (failing) BATextend... a useless action and evidently an IO drainer.
Closing the bug, for it is hard to cast in a nightly test.
## Comment 12207
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2836259 at http://sourceforge.net/support/tracker.php?aid=2836259
| Monet does not handle "Out of disc space" error correct | https://api.github.com/repos/MonetDB/MonetDB/issues/2154/comments | 0 | 2020-11-30T10:24:38Z | 2024-06-28T13:23:48Z | https://github.com/MonetDB/MonetDB/issues/2154 | 753,351,568 | 2,154 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-10 01:13:09 +0200
From: Wouter Alink <<alink>>
To: Gabriele Modena <<gabriele_modena>>
Version: -- development
Last updated: 2009-11-17 12:24:31 +0100
## Comment 12199
Date: 2009-08-10 01:13:09 +0200
From: Wouter Alink <<alink>>
Regarding the use of the XQuery language in the PHP interface:
Conforming to the current JDBC behavior (which allows both XQuery and SQL queries): a different row should be returned for each item in a sequence of results. An example: a query like "1 to 4" should return a result with 1 column and 4 rows, containing 1, 2, 3, and 4 respectively. Currently it doesn't seem to do so.
See test in pathfinder: tests/clients/Tests/php_monetdb.XQUERY.sh
And see example in clients: src/examples/php/xquerysample.php
## Comment 12200
Date: 2009-08-17 11:38:13 +0200
From: @grobian
Since we cannot find why this doesn't work, I suggest to fix it differently (by recoding it from scratch ;) )
Gabriele, how about adding support for this to your new driver?
## Comment 12201
Date: 2009-08-17 14:29:01 +0200
From: @gmodena
I'm having a look at it right now. It should be just a matter of letting the native php diver "speak" M4 protocol correctly and then parse the result of an xquery.
## Comment 12202
Date: 2009-11-12 10:55:14 +0100
From: @peterboncz
Gabriele: what is the status of this bug?
## Comment 12203
Date: 2009-11-17 00:24:31 +0100
From: @drstmane
The PHP tests now seems to work fine for both SQL & XQuery:
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.64.d.1-Fedora10/src_test_mapi/php_monetdb.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d.1-Fedora10/tests_clients/php_monetdb.out.00.html
Exceptions on Gentoo & Windows are probably due to PHP "glitched" on those systems:
Fatal error: Call to undefined function socket_create() in /home/monet/testing/monet.GNU.64.64.d.1.13984/clients/.GNU.64.64.d.1-Gentoo2.0.1/share/php/monetdb/php_mapi.inc on line 675
Fatal error: Call to undefined function socket_create() in G:\monet\monet.Int.32.32.d.1.13984\clients\.Int.32.32.d.1-Windows5.1\share\php\monetdb\php_mapi.inc on line 675
cf.,
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.64.d.1-Gentoo2.0.1/src_test_mapi/php_monetdb.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_clients/php_monetdb.out.00.html
CLosing.
## Comment 12204
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2834654 at http://sourceforge.net/support/tracker.php?aid=2834654
| PHP+XQUERY: multiple results not returned correctly | https://api.github.com/repos/MonetDB/MonetDB/issues/2153/comments | 0 | 2020-11-30T10:24:34Z | 2024-06-27T11:41:02Z | https://github.com/MonetDB/MonetDB/issues/2153 | 753,351,507 | 2,153 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-07 07:36:33 +0200
From: @skinkie
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2009-08-13 04:11:37 +0200
## Comment 12184
Date: 2009-08-07 19:36:33 +0200
From: @skinkie
/opt/monetdb-head/include/MonetDB/mapilib/Mapi.h:182: error: expected declaration specifiers or '...' before 'mapi_int64'
/opt/monetdb-head/include/MonetDB/mapilib/Mapi.h:186: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'mapi_fetch_all_rows'
/opt/monetdb-head/include/MonetDB/mapilib/Mapi.h:188: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'mapi_get_row_count'
/opt/monetdb-head/include/MonetDB/mapilib/Mapi.h:189: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'mapi_get_last_id'
/opt/monetdb-head/include/MonetDB/mapilib/Mapi.h:190: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'mapi_rows_affected'
It seems HAVE_LONG_LONG is not defined. Thus the compilation of client programs break.
## Comment 12185
Date: 2009-08-11 14:16:08 +0200
From: @drstmane
Compilation of clients seems to work fine on all our nightly testing platforms, both with th Aug2009 release branch ("Stable") and the development trunk (CVS HEAD, "Current"):
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/clients/.Makes3/index_short.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/clients/.Makes3/index_short.html
Could you please give more details about which client programs fail to compile in which environment (OS, compiler, MonetDB version)?
Thanks!
## Comment 12186
Date: 2009-08-11 14:38:55 +0200
From: @skinkie
Projectmembers of your team are aware of the issue, since more people complained privately. It is unrelated to the compilation of MonetDB, but it hits for anyone that does external development. Now I have heard yesterday that it would be solved by including client(s)_config.h but since this header is not installed, there is nothing to include.
## Comment 12187
Date: 2009-08-11 14:48:48 +0200
From: @drstmane
Ah, it's about other client programs (than those provided with the MonetDB code base) that use Mapi.h --- sorry, I didn't get this ...
Well, the proper solution would IMHO be to create a setup for your project X that is similar to the standard MonetDB setup, in particular also coming with a configure in a "MonetDB style", that checks the cofniguration of your platform and creates its own X_config.h accordingly. Your client programs should then include this X_config.h as the very first include, and thsuu before including Mapi.h.
See our "template" CVS module for a sample "template" starting point for your project X setup.
## Comment 12188
Date: 2009-08-11 14:56:00 +0200
From: @skinkie
'Client programs' are ambigious I agree. Anyway this 'change' ought to be documented. At least in the next changelog, because such configuration was not a requirement before. And I think the same type should be used, so the previously mentioned client(s)_config.h should be installed?
## Comment 12189
Date: 2009-08-11 15:13:58 +0200
From: @drstmane
clients_config.h is created by configure, hence is cannot be "installed", since it has to reflect the environment of the machine you are compiling on, not the machine the release was built on.
The type to be used in this case is the very 64-bit integer type that your platform provides; hence, your configure needs to tell Mapi.h (via your X_config.h) which 64-bit integer type your platform has.
I agree that we should document this in the Mapi documentation accordingly, once (before) we release the Aug2009 release.
## Comment 12190
Date: 2009-08-11 15:18:01 +0200
From: @drstmane
... of course, if your are compiling clients from sources, and you don't have/want your own separate project source tree, and you don't mind changing your CVS checkout or untared sources of clients locally, you can also simply put your own client programs into some sub-directory in the clients tree, alter the related Makefile.ag's (if CVS checkout) or Makefile.in's (if untared source) accordingly, and then simply include the clients clients_config.h ...
## Comment 12191
Date: 2009-08-11 18:11:21 +0200
From: @skinkie
In my perspective such 'innovations' are not required my programming against MySQL or PostgreSQL. Including a single header is enough to target their client interface. Just including <monetdb.h> or <mapi.h> is so much more userfriendly...
## Comment 12192
Date: 2009-08-11 18:22:16 +0200
From: @drstmane
Until we re-design our code base and setup to ensure portability to oh so many different platforms, I'm afraid with MonetDB we have to live with the situation as it is now.
Any detailed constructive ideas and/or hand-on help to simplify things while keeping the functionality and (in this case) portability (at least) as-is are more than welcome.
## Comment 12193
Date: 2009-08-11 18:46:12 +0200
From: @skinkie
The major point is actually a few lines above:
/* avoid using "ifdef WIN32" so that this file does not need our config.h */
So adding the actual platform check inside the headerfile would be an effective solution. Or making the header file platform specific. Removing all ifdefs and providing only a header for that specific MonetDB installation. Since the actual data files are not portable between platforms why should the header file be?
Anyway this piece was somewhere on a forum, and has resemblance with what is already there:
if defined(_MSC_VER) || defined(__BORLANDC__)
typedef unsigned __int64 ulong64;
typedef signed __int64 long64;
else
typedef unsigned long long ulong64;
typedef signed long long long64;
endif
The main point now: there is no else, that makes it break when nothing is defined.
## Comment 12194
Date: 2009-08-12 22:19:04 +0200
From: @drstmane
Ok, the following patch might do the trick --- before checking it in, I'd like to (have) verif(y|ied), though, that it indeed makes sense to assume that (all) Windows compilers provide type __int64, while (all) Unix compilers provide type long long ...
$ cvs diff clients/src/mapilib/Mapi.mx
Index: clients/src/mapilib/Mapi.mx
===================================================================
RCS file: /cvsroot/monetdb/clients/src/mapilib/Mapi.mx,v
retrieving revision 1.79.2.2
diff -u -r1.79.2.2 Mapi.mx
--- clients/src/mapilib/Mapi.mx 6 Aug 2009 07:03:15 -0000 1.79.2.2
+++ clients/src/mapilib/Mapi.mx 12 Aug 2009 20:09:24 -0000
@@ -1120,6 +1120,10 @@
/* avoid using "ifdef WIN32" so that this file does not need our config.h */
if defined(_MSC_VER) || defined(__CYGWIN__) || defined(__MINGW32__)
+if defined(__ICL) && __ICL >= 1010
+define HAVE_LONG_LONG 1
+endif
+define HAVE___INT64 1
ifndef LIBMAPI
define mapi_export extern __declspec(dllimport)
else
@@ -1127,6 +1131,7 @@
endif
else
define mapi_export extern
+define HAVE_LONG_LONG 1
endif
## Comment 12195
Date: 2009-08-12 22:29:12 +0200
From: @sjoerdmullender
I would greatly prefer a solution where the include file "just works". Stefan M's patch may well do the trick.
I think we can indeed assume that all _MSC_VER setting compilers define __int64, and all gcc's define long long (if they don't, they're probably too old anyway).
But I would rather get rid of the use of the HAVE_* macros in Mapi.h since they might conflict with a config.h file that the application might have and include.
## Comment 12196
Date: 2009-08-12 22:39:11 +0200
From: @drstmane
e.g. like this?
$ cvs diff clients/src/mapilib/Mapi.mx
Index: clients/src/mapilib/Mapi.mx
===================================================================
RCS file: /cvsroot/monetdb/clients/src/mapilib/Mapi.mx,v
retrieving revision 1.79.2.2
diff -u -r1.79.2.2 Mapi.mx
--- clients/src/mapilib/Mapi.mx 6 Aug 2009 07:03:15 -0000 1.79.2.2
+++ clients/src/mapilib/Mapi.mx 12 Aug 2009 20:33:47 -0000
@@ -1120,6 +1120,13 @@
/* avoid using "ifdef WIN32" so that this file does not need our config.h */
if defined(_MSC_VER) || defined(__CYGWIN__) || defined(__MINGW32__)
+if defined(__ICL) && __ICL >= 1010
+typedef unsigned long long mapi_uint64;
+typedef long long mapi_int64;
+else
+typedef unsigned __int64 mapi_uint64;
+typedef __int64 mapi_int64;
+endif
ifndef LIBMAPI
define mapi_export extern __declspec(dllimport)
else
@@ -1127,16 +1134,8 @@
endif
else
define mapi_export extern
-endif
-
-ifdef HAVE_LONG_LONG
typedef unsigned long long mapi_uint64;
typedef long long mapi_int64;
-else
-ifdef HAVE___INT64
-typedef unsigned __int64 mapi_uint64;
-typedef __int64 mapi_int64;
-endif
endif
/* three structures used for communicating date/time information */
## Comment 12197
Date: 2009-08-13 16:11:37 +0200
From: @drstmane
fixed in CVS (Aug2009 branch; propagation pending):
to define types mapi_int64 & mapi_uint64,
we no longer depend on HAVE_LONG_LONG and HAVE___INT64
as defined by our configure in our *_config.h,
but rather assume that
- Microsoft & Intel compilers under Windows have type __int64
- gcc and other (Unix-) compilers (usually) have type long long
## Comment 12198
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2833831 at http://sourceforge.net/support/tracker.php?aid=2833831
| mapilib/Mapi.h missing HAVE_LONG_LONG | https://api.github.com/repos/MonetDB/MonetDB/issues/2152/comments | 0 | 2020-11-30T10:24:31Z | 2024-06-27T11:41:01Z | https://github.com/MonetDB/MonetDB/issues/2152 | 753,351,473 | 2,152 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-07 03:43:18 +0200
From: @skinkie
To: @njnes
Version: -- development
Last updated: 2009-08-11 01:59:11 +0200
## Comment 12177
Date: 2009-08-07 15:43:18 +0200
From: @skinkie
COPY 664 RECORDS INTO haltesgeo FROM '/home/skinkie/clean.csv' USING DELIMITERS ';', '\n';
!SQLException:sql:value '4.9.036' while parsing '4.9.036' from line 80 field 2 not inserted, expecting type dbl
!SQLException:importTable:failed to import table
0 tuples
Now this line is actually 81 in the file.
## Comment 12178
Date: 2009-08-07 16:04:57 +0200
From: @njnes
could you please create an example which could be run in the test set?
## Comment 12179
Date: 2009-08-07 16:09:51 +0200
From: @njnes
fix checked into current. Could you validate that it works?
## Comment 12180
Date: 2009-08-07 16:10:53 +0200
From: @njnes
return current inserted count + 1, for line number. This should still be corrected for the skip value!
## Comment 12181
Date: 2009-08-07 16:12:50 +0200
From: @skinkie
create table number (isanumber integer);
COPY 2 RECORDS INTO number FROM '/home/skinkie/notanumber' USING DELIMITERS '\n';
notanumber:
1
bla
!SQLException:sql:value 'bla' while parsing 'bla' from line 1 field 0 not inserted, expecting type int
should be:
!SQLException:sql:value 'bla' while parsing 'bla' from line 2 field 1 not inserted, expecting type int
(I think field one sounds much better too)
## Comment 12182
Date: 2009-08-11 14:12:14 +0200
From: @drstmane
added test in the Aug2009 release branch as
sql/src/test/BugTracker-2009/Tests/copy_into_error_off_by_one.SF-2833696.*
Still fails in the Aug2009 release branch,
but is fixed with the new bulk loader in the development trunk.
## Comment 12183
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2833696 at http://sourceforge.net/support/tracker.php?aid=2833696
| Copy into error off by one | https://api.github.com/repos/MonetDB/MonetDB/issues/2151/comments | 0 | 2020-11-30T10:24:27Z | 2024-06-27T11:41:00Z | https://github.com/MonetDB/MonetDB/issues/2151 | 753,351,436 | 2,151 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-06 11:01:43 +0200
From: @mlkersten
To: @njnes
Version: -- development
Last updated: 2009-08-06 09:06:50 +0200
## Comment 12174
Date: 2009-08-06 11:01:43 +0200
From: @mlkersten
The experiments with SF-1 scripts show that foreign key checking
on part <-> partsupp fails. It is the first FK check in the file alter.sql.
## Comment 12175
Date: 2009-08-06 21:06:50 +0200
From: @mlkersten
The underlying scripts had a temporary patch for another experiment. It resulted in non-loading of a table.
## Comment 12176
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2833063 at http://sourceforge.net/support/tracker.php?aid=2833063
| SQL: Foreign key checking fails | https://api.github.com/repos/MonetDB/MonetDB/issues/2150/comments | 0 | 2020-11-30T10:24:23Z | 2024-06-28T13:23:47Z | https://github.com/MonetDB/MonetDB/issues/2150 | 753,351,396 | 2,150 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-04 03:16:28 +0200
From: Gleb Lanbin <<glanbin>>
To: @njnes
Version: -- development
Last updated: 2009-08-11 02:18:18 +0200
## Comment 12168
Date: 2009-08-04 15:16:28 +0200
From: Gleb Lanbin <<glanbin>>
Hello,
we have a strange problem with MonetDB JDBC driver, when we use standard statement to insert numeric data value all works. but with prepared statement we got the problem with value 0 (null) for decimal and numeric data type, so when we try to insert data with null in PreparedStatement we got SQL exception. I attached file with sample source code which generates such error.
Thanks
## Comment 12169
Date: 2009-08-04 15:20:25 +0200
From: @grobian
this is most probably a server-side exec problem
## Comment 12170
Date: 2009-08-04 19:35:14 +0200
From: @njnes
added test to BugTracker-2009/Tests/prepare_decimal_bug.SF-2831994.sql
0.0 got stripped to '.' having no digits. Asking for a decimal with length 0, automagically switches to maximum (19) digits length, which is larger as 5). So now we detect to low number of digits (<= 0) and set it to atleast 1.
## Comment 12171
Date: 2009-08-11 13:59:08 +0200
From: @drstmane
The test (using mclient, not JDBC) seems to work fine:
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.64.d.1-Fedora10/src_test_BugTracker-2009/prepare_decimal_bug.SF-2831994.out.00.html
Could this one hence be closed?
## Comment 12172
Date: 2009-08-11 14:18:18 +0200
From: @drstmane
Fixed by Niels.
Closing.
## Comment 12173
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2831994 at http://sourceforge.net/support/tracker.php?aid=2831994
| Prepared statement problem | https://api.github.com/repos/MonetDB/MonetDB/issues/2149/comments | 0 | 2020-11-30T10:24:20Z | 2024-06-27T11:40:57Z | https://github.com/MonetDB/MonetDB/issues/2149 | 753,351,361 | 2,149 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-03 06:20:46 +0200
From: Wouter Alink <<alink>>
To: @njnes
Version: -- development
Last updated: 2009-11-11 07:50:44 +0100
## Comment 12162
Date: 2009-08-03 18:20:46 +0200
From: Wouter Alink <<alink>>
The following query
START TRANSACTION;
CREATE TABLE b (value int);
INSERT INTO b VALUES (25);
SELECT (rel.value / 10) AS value, value as x FROM b as rel;
ROLLBACK;
Gives perhaps strange results:
$ mclient -lsql -dtest1 temp.sql
[ 1 ]
% lhm., lhm.rel table_name
% value, x name
% int, int type
% 1, 2 length
[ 2, 25 ]
Expected output would either be:
[ 2, 2 ]
Or an error which says that value is ambiguous.
My mserver5 says:
MonetDB server v5.12.1, based on kernel v1.30.1
Serving database 'test1', using 4 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs dynamically linked
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2009 MonetDB B.V., all rights reserved
Visit http://monetdb.cwi.nl/ for further information
warning: please don't forget to set your vault key!
(see /home/alink/opt/MonetDB-stable/etc/monetdb5.conf)
Listening for connection requests on mapi:monetdb://127.0.0.1:50100/
MonetDB/SQL module v2.30.1 loaded
## Comment 12163
Date: 2009-08-04 00:33:45 +0200
From: @njnes
selection doesn't look at aliases in the current selection list, it looks at column results from the inner query, ie in your example columns from 'rel'.
## Comment 12164
Date: 2009-08-10 17:12:34 +0200
From: @njnes
reopened on request from wouter
## Comment 12165
Date: 2009-08-10 17:29:42 +0200
From: Wouter Alink <<alink>>
Apologies, I reduced the issue too much (I was already afraid to do so).
The initial problem was:
SELECT EXTRACT(YEAR FROM rel.value) as value
FROM x as rel
GROUP BY value;
It was grouping on rel.value, and not on year (which I was expecting).
The work-around was, using something like:
SELECT EXTRACT(YEAR FROM rel.value) as avalue
FROM x as rel
GROUP BY avalue;
Which worked like a charm.
## Comment 12166
Date: 2009-08-16 11:14:14 +0200
From: @njnes
GROUP by works on the pre-selection result normaly, ie that the group by isn't using the extract etc is to be expected. The work around for your problem is an extra feature, ie we support fallback onto the selection when we cannot find column expressions in the pre-selection result.
## Comment 12167
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2831582 at http://sourceforge.net/support/tracker.php?aid=2831582
| SQL: ambiguity not detected | https://api.github.com/repos/MonetDB/MonetDB/issues/2148/comments | 0 | 2020-11-30T10:24:16Z | 2024-06-28T13:23:47Z | https://github.com/MonetDB/MonetDB/issues/2148 | 753,351,325 | 2,148 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-03 03:17:58 +0200
From: @yzchang
To: Henning Rode <<hrode>>
Version: -- development
Last updated: 2009-09-15 05:17:31 +0200
## Comment 12157
Date: 2009-08-03 15:17:58 +0200
From: @yzchang
On FC10, 64bits, installed using ' monetdb-install.sh --enable-xquery --nightly=stable' about 3 days ago, executing a tijah:queryall function results in segmentation fault in Mserver:
================
In server side terminal
================
$ Mserver --dbinit="module(pathfinder);"
MonetDB Server v4.30.4
based on GDK v1.30.4
Copyright (c) 1993-July 2008, CWI. All rights reserved.
Copyright (c) August 2008-2009, MonetDB B.V.. All rights reserved.
Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs; dynamically linked.
Visit http://monetdb.cwi.nl/ for further information.
PF/Tijah module v0.13.0 loaded. http://dbappl.cs.utwente.nl/pftijah
MonetDB/XQuery module v0.28.3 loaded (default back-end is 'algebra')
XRPC administrative console at http://127.0.0.1:50001/admin
MonetDB>Segmentation fault
================
In client side terminal
================
$ cat debates.xml
<?xml version="1.0" standalone="yes"?>
<proceedings>
<speech speaker="Roy Walter">
<p>Wind farms at Heathrow are a good thing.</p>
</speech>
</proceedings>
$ MonetDB/bin/mclient -lx
xquery>pf:add-doc("/tmp/debates.xml", "debates.xml")
more><>
xquery>pf:documents()
more><>
<document updatable="false" url="/tmp/debates.xml" collection="debates.xml">debates.xml</document>
xquery>tijah:create-ft-index()
more><>
xquery>tijah:queryall("//speech[about(., Heathrow)]")
more><>
MAPI = monetdb@localhost:50000
ACTION= read_line
QUERY = tijah:queryall("//speech[about(., Heathrow)]")
ERROR = Connection terminated
The same crash happens on the same machine, if the document was shredded as an updatable document with 10% hole.
The query also crashes Mserver on Windows Vista 64bits, installed using the latest released windows installer, with document shredded as both read-only and updatable.
I'm not sure if this is a 64bits related problem, as I got the query from a user who is using this tijah function on a 32-bits XP machine, and did not mention any crash.
Jennie
## Comment 12158
Date: 2009-08-21 10:53:54 +0200
From: Henning Rode <<hrode>>
my last check-in seems to fix the problem.
## Comment 12159
Date: 2009-08-21 11:46:22 +0200
From: @drstmane
Is there a test for this one?
If so, please state where;
otherwise, please (try to) add one, or explain why that isn't (easily) possible.
After that, please close, again.
Thanks!
## Comment 12160
Date: 2009-09-15 17:17:31 +0200
From: Henning Rode <<hrode>>
test case added. bug was fixed already earlier.
## Comment 12161
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2831472 at http://sourceforge.net/support/tracker.php?aid=2831472
| tijah:queryall crashes Mserver | https://api.github.com/repos/MonetDB/MonetDB/issues/2147/comments | 0 | 2020-11-30T10:24:13Z | 2024-06-27T11:40:55Z | https://github.com/MonetDB/MonetDB/issues/2147 | 753,351,280 | 2,147 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-01 10:13:53 +0200
From: @grobian
To: @njnes
Version: -- development
Last updated: 2009-08-03 11:02:55 +0200
## Comment 12152
Date: 2009-08-01 22:13:53 +0200
From: @grobian
CREATE TABLE CHAR_TBL(f1 char);
SELECT '' AS seven, CHAR_TBL.*;
!Column expression Table 'char_tbl' unknown
Might be that MonetDB doesn't do this because we want CHAR_TBL in a WHERE-clause, in that case we better just fix the test.
pg_regress/char
## Comment 12153
Date: 2009-08-01 22:27:47 +0200
From: @grobian
related and perhaps more of an issue is the following query:
QUERY = SELECT '' AS two, * FROM TEXT_TBL;
ERROR = !syntax error, unexpected '*' in: "select '' as two, *"
aparently MonetDB expects text_tbl.* here
pg_regress/text
## Comment 12154
Date: 2009-08-03 11:09:27 +0200
From: @grobian
ok, the first issue (select from table without table in FROM-clause) is fixed in the tests (we don't suppor this), so this bug is about the remaining issue SELECT '' AS two, * FROM TEXT_TBL; not working.
## Comment 12155
Date: 2009-08-03 23:02:55 +0200
From: @njnes
fixed parser to handle '*' any were in the selection
## Comment 12156
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2830905 at http://sourceforge.net/support/tracker.php?aid=2830905
| SQL: unexpected '*' in: "select '' as two, * | https://api.github.com/repos/MonetDB/MonetDB/issues/2146/comments | 0 | 2020-11-30T10:24:09Z | 2024-06-27T11:40:55Z | https://github.com/MonetDB/MonetDB/issues/2146 | 753,351,218 | 2,146 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-01 09:49:09 +0200
From: @grobian
To: Fabian Groffen <<fabian>>
Version: -- development
Last updated: 2009-08-02 11:25:05 +0200
## Comment 12149
Date: 2009-08-01 21:49:09 +0200
From: @grobian
SELECT 'be_r' LIKE 'b_e__r' ESCAPE '_' AS "true";
returns false, because it doesn't ignore the invalid '_e' (compare to '\e') escape.
test in pg_regress/strings_like
## Comment 12150
Date: 2009-08-02 11:25:05 +0200
From: @grobian
fixed in Stable, will first appear in next feature release
## Comment 12151
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2830900 at http://sourceforge.net/support/tracker.php?aid=2830900
| SQL: LIKE doesn't ignore invalid escapes | https://api.github.com/repos/MonetDB/MonetDB/issues/2145/comments | 0 | 2020-11-30T10:24:06Z | 2024-06-27T11:40:54Z | https://github.com/MonetDB/MonetDB/issues/2145 | 753,351,179 | 2,145 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-08-01 09:38:30 +0200
From: @grobian
To: @njnes
Version: -- development
Last updated: 2009-08-03 11:03:33 +0200
## Comment 12145
Date: 2009-08-01 21:38:30 +0200
From: @grobian
sql>SELECT 'indio' NOT LIKE 'in_o' AS "true";
!SELECT: no such binary operator 'not_like(char,char)'
0 tuples
pg_regress/strings_like is full of it
## Comment 12146
Date: 2009-08-03 11:06:19 +0200
From: @grobian
implementing the not_like functions in pcre.mx doesn't solve the problem
## Comment 12147
Date: 2009-08-03 23:03:33 +0200
From: @njnes
fixed implementation of not_(i)like
## Comment 12148
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2830895 at http://sourceforge.net/support/tracker.php?aid=2830895
| SQL: NOT LIKE -> no such binary operator | https://api.github.com/repos/MonetDB/MonetDB/issues/2144/comments | 0 | 2020-11-30T10:24:02Z | 2024-06-27T11:40:53Z | https://github.com/MonetDB/MonetDB/issues/2144 | 753,351,137 | 2,144 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-31 12:02:57 +0200
From: @swingbit
To: @njnes
Version: -- development
Last updated: 2009-09-20 09:39:14 +0200
## Comment 12140
Date: 2009-07-31 12:02:57 +0200
From: @swingbit
Starting mserver5 in read-only mode, i.e. with 'gdk_readonly=yes' in monetdb5.conf, and then trying a CREATE statement, crashes the server:
sql>create table x(a int);
MAPI = monetdb@localhost:50151
ACTION= read_line
QUERY = create table x(a int);
ERROR = Connection terminated
## Comment 12141
Date: 2009-08-06 15:58:47 +0200
From: @mlkersten
See report 2808453
## Comment 12142
Date: 2009-08-14 14:10:02 +0200
From: @drstmane
tagged subject
## Comment 12143
Date: 2009-09-20 21:39:14 +0200
From: @njnes
added checks to disallow schema statements on readonly db.
added test to src/test/BugTracker-2009/create_on_ro_db_crash.SF-2830238.py
## Comment 12144
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2830238 at http://sourceforge.net/support/tracker.php?aid=2830238
| SQL: Crash when trying to write on read-only DB | https://api.github.com/repos/MonetDB/MonetDB/issues/2143/comments | 0 | 2020-11-30T10:23:58Z | 2024-06-27T11:40:52Z | https://github.com/MonetDB/MonetDB/issues/2143 | 753,351,096 | 2,143 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-29 02:43:45 +0200
From: Roy Walter <<gooseflight>>
To: @sjoerdmullender
Version: -- development
Last updated: 2009-08-04 11:43:54 +0200
## Comment 12136
Date: 2009-07-29 14:43:45 +0200
From: Roy Walter <<gooseflight>>
The page at http://monetdb.cwi.nl/downloads/Ubuntu/ provides instructions for downloading MonetDB packges for Jaunty. Before the line "After this you can use apt-get install monetdb\* to install the MonetDB packages." there should be an instruction to run "apt-get update". Without this I was unable to install the packages.
## Comment 12137
Date: 2009-07-29 15:35:00 +0200
From: @sjoerdmullender
I added apt-get updte to the instructions. What do you think?
## Comment 12138
Date: 2009-08-04 11:43:54 +0200
From: @sjoerdmullender
Closing.
## Comment 12139
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2828959 at http://sourceforge.net/support/tracker.php?aid=2828959
| MonetDB Ubuntu download/install | https://api.github.com/repos/MonetDB/MonetDB/issues/2142/comments | 0 | 2020-11-30T10:23:56Z | 2024-06-27T11:40:51Z | https://github.com/MonetDB/MonetDB/issues/2142 | 753,351,054 | 2,142 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-27 06:19:35 +0200
From: @swingbit
To: @njnes
Version: -- development
Last updated: 2009-07-31 11:56:43 +0200
## Comment 12133
Date: 2009-07-27 18:19:35 +0200
From: @swingbit
The following wrong SQL is not caught correctly:
sql>select *;
MAPI = monetdb@localhost:50151
ACTION= read_line
QUERY = select *;
ERROR = Connection terminated
## Comment 12134
Date: 2009-07-31 11:56:43 +0200
From: @swingbit
I see only now that my bug report is a duplicate of bug report 2825974 (although this was reported in the development branch)
Closing this bug report as duplicate (it is also fixed now).
## Comment 12135
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2827917 at http://sourceforge.net/support/tracker.php?aid=2827917
This bug was marked DUPLICATE in the database it was moved from.
Changing resolution to "MOVED"
| select * with no FROM clause crashes | https://api.github.com/repos/MonetDB/MonetDB/issues/2141/comments | 0 | 2020-11-30T10:23:52Z | 2024-06-28T13:23:46Z | https://github.com/MonetDB/MonetDB/issues/2141 | 753,351,016 | 2,141 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-27 06:15:23 +0200
From: @swingbit
To: @njnes
Version: -- development
Last updated: 2009-08-07 04:01:51 +0200
## Comment 12129
Date: 2009-07-27 18:15:23 +0200
From: @swingbit
In the following statement, the order by clause is ambiguous.
I guess the current behaviour is to order on the first of the two ids, rather than issueing an error, as it is done for the target list.
sql>select args.id, types.id from args, types order by id limit 10;
+------+----+
| id | id |
+======+====+
| 1210 | 9 |
| 1210 | 8 |
| 1210 | 6 |
| 1210 | 7 |
| 1210 | 3 |
| 1210 | 5 |
| 1210 | 2 |
| 1210 | 0 |
| 1210 | 4 |
| 1210 | 1 |
+------+----+
10 tuples
## Comment 12130
Date: 2009-08-06 15:34:14 +0200
From: @mlkersten
An error message would be welcome
## Comment 12131
Date: 2009-08-07 16:01:51 +0200
From: @njnes
added test to src/test/BugTracker-2009/Tests/ambiguous_order_by.SF-2827916.sql
fixed, ie we now strictly check for ambiguous column names
## Comment 12132
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2827916 at http://sourceforge.net/support/tracker.php?aid=2827916
| ambiguous ORDER BY clause not checked | https://api.github.com/repos/MonetDB/MonetDB/issues/2140/comments | 0 | 2020-11-30T10:23:49Z | 2024-06-27T11:40:49Z | https://github.com/MonetDB/MonetDB/issues/2140 | 753,350,972 | 2,140 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-27 02:13:11 +0200
From: Franck Routier <<alci>>
To: @sjoerdmullender
Version: -- development
Last updated: 2009-07-27 02:41:41 +0200
## Comment 12125
Date: 2009-07-27 14:13:11 +0200
From: Franck Routier <<alci>>
Monetdb is segfaulting, probably while interpreting a query or planning
it.
Error in syslog says:
[ 1230.812708] mserver5[5369]: segfault at 7fab4c2b5ff8 ip
00007fab4f894285 sp 00007fab4c2b6000 error 6 in
lib_sql.so.2.30.2[7fab4f7e3000+116000]
Following is the faulty ddl / query.
The ddl is the following:
CREATE TABLE cabact
(
cabactcsu integer NOT NULL,
cabact___rforefide character varying(32) NOT NULL,
cabact___rteprcide character varying(32) NOT NULL,
cabact___rtestdide character varying(32) NOT NULL,
cabact___rfovsnide character varying(32) NOT NULL,
cabactlil character varying(100) NOT NULL,
cabactlic character varying(32) NOT NULL,
cabacttxt character varying(1500),
cabact___rfontride character varying(32),
cabact_f1rfodstide character varying(32),
cabact_f2rfodstide character varying(32),
cabact_f3rfodstide character varying(32),
cabact_f4rfodstide character varying(32),
cabact_f5rfodstide character varying(32),
cabact_f6rfodstide character varying(32),
cabact_f7rfodstide character varying(32),
cabact_f8rfodstide character varying(32),
cabact_f9rfodstide character varying(32),
cabact_c1rfodstide character varying(32),
cabact_c2rfodstide character varying(32),
cabact_c3rfodstide character varying(32),
cabact_c4rfodstide character varying(32),
cabact_c5rfodstide character varying(32),
cabact_c6rfodstide character varying(32),
cabact_c7rfodstide character varying(32),
cabact_c8rfodstide character varying(32),
cabact_c9rfodstide character varying(32),
cabactqte double,
cabact___rfountide character varying(32),
cabactdtd timestamp with time zone NOT NULL,
cabactdtf timestamp with time zone NOT NULL,
cabactax1 integer NOT NULL DEFAULT 1,
cabactax2 integer NOT NULL DEFAULT 2,
cabactax3 integer NOT NULL DEFAULT 3,
cabactax4 integer NOT NULL DEFAULT 4,
cabactax5 integer NOT NULL DEFAULT 5,
cabactax6 integer NOT NULL DEFAULT 6,
cabactax7 integer NOT NULL DEFAULT 7,
cabactax8 integer NOT NULL DEFAULT 8,
cabactax9 integer NOT NULL DEFAULT 9
);
CREATE TABLE rfoade
(
rfoade___rforefide character varying(50) NOT NULL,
rfoade___rfovdeide character varying(50) NOT NULL,
rfoade_i_rfodstide character varying(50) NOT NULL,
rfoadeaxe integer DEFAULT 0 NOT NULL,
rfoadervs integer NOT NULL,
rfoadenpm integer DEFAULT 1,
rfoade_s_rfodstide character varying(32) NOT NULL,
rfoadegch character varying(120) DEFAULT 'AAAAA' NOT NULL,
rfoadedrt character varying(120) DEFAULT 'ZZZZZ' NOT NULL,
rfoadeniv integer DEFAULT 0 NOT NULL,
rfoadetxt character varying(1800),
rfoadenum integer DEFAULT 99999 NOT NULL,
rfoadeden integer DEFAULT 999 NOT NULL,
rfoadechm character varying(5500) DEFAULT 'INVALID' NOT NULL,
rfoadeord integer DEFAULT 999999 NOT NULL
);
And the query is:
select * from cabact where cabact___rforefide = 'FHSJ' and
cabact___rteprcide = 'CPTANA' and
cabact___rtestdide = '100' and
cabact___rfovsnide = '200805_001' and
(cabact_f1rfodstide IS NULL or cabact_f1rfodstide IN (select
rfoade_i_rfodstide from rfoade where rfoade___rfovdeide='SECA' and
rfoade___rforefide = 'FHSJ' and rfoadervs='1')) and
(cabact_f2rfodstide IS NULL or cabact_f2rfodstide IN
(select rfoade_i_rfodstide from rfoade where rfoade___rfovdeide='SECA'
and rfoade___rforefide = 'FHSJ' and rfoadervs='1')) and
(cabact_f3rfodstide IS NULL or cabact_f3rfodstide IN
(select rfoade_i_rfodstide from rfoade where rfoade___rfovdeide='SECA'
and rfoade___rforefide = 'FHSJ' and rfoadervs='1')) and
(cabact_f4rfodstide IS NULL or cabact_f4rfodstide IN
(select rfoade_i_rfodstide from rfoade where rfoade___rfovdeide='SECA'
and rfoade___rforefide = 'FHSJ' and rfoadervs='1')) and
(cabact_f5rfodstide IS NULL or cabact_f5rfodstide IN
(select rfoade_i_rfodstide from rfoade where rfoade___rfovdeide='SECA'
and rfoade___rforefide = 'FHSJ' and rfoadervs='1')) and
(cabact_f6rfodstide IS NULL or cabact_f6rfodstide IN
(select rfoade_i_rfodstide from rfoade where rfoade___rfovdeide='SECA'
and rfoade___rforefide = 'FHSJ' and rfoadervs='1')) and
(cabact_f7rfodstide IS NULL or cabact_f7rfodstide IN
(select rfoade_i_rfodstide from rfoade where rfoade___rfovdeide='SECA'
and rfoade___rforefide = 'FHSJ' and rfoadervs='1')) and
(cabact_f8rfodstide IS NULL or cabact_f8rfodstide IN
(select rfoade_i_rfodstide from rfoade where rfoade___rfovdeide='SECA'
and rfoade___rforefide = 'FHSJ' and rfoadervs='1')) and
(cabact_f9rfodstide IS NULL or cabact_f9rfodstide IN
(select rfoade_i_rfodstide from rfoade where rfoade___rfovdeide='SECA'
and rfoade___rforefide = 'FHSJ' and rfoadervs='1')) and
(cabact_c1rfodstide IS NULL or cabact_c1rfodstide IN
(select rfoade_i_rfodstide from rfoade where rfoade___rfovdeide='SECA'
and rfoade___rforefide = 'FHSJ' and rfoadervs='1')) and
(cabact_c2rfodstide IS NULL or cabact_c2rfodstide IN
(select rfoade_i_rfodstide from rfoade where rfoade___rfovdeide='SECA'
and rfoade___rforefide = 'FHSJ' and rfoadervs='1')) and
(cabact_c3rfodstide IS NULL or cabact_c3rfodstide IN
(select rfoade_i_rfodstide from rfoade where rfoade___rfovdeide='SECA'
and rfoade___rforefide = 'FHSJ' and rfoadervs='1')) and
(cabact_c4rfodstide IS NULL or cabact_c4rfodstide IN
(select rfoade_i_rfodstide from rfoade where rfoade___rfovdeide='SECA'
and rfoade___rforefide = 'FHSJ' and rfoadervs='1')) and
(cabact_c5rfodstide IS NULL or cabact_c5rfodstide IN
(select rfoade_i_rfodstide from rfoade where rfoade___rfovdeide='SECA'
and rfoade___rforefide = 'FHSJ' and rfoadervs='1')) and
(cabact_c6rfodstide IS NULL or cabact_c6rfodstide IN
(select rfoade_i_rfodstide from rfoade where rfoade___rfovdeide='SECA'
and rfoade___rforefide = 'FHSJ' and rfoadervs='1')) and
(cabact_c7rfodstide IS NULL or cabact_c7rfodstide IN
(select rfoade_i_rfodstide from rfoade where rfoade___rfovdeide='SECA'
and rfoade___rforefide = 'FHSJ' and rfoadervs='1')) and
(cabact_c8rfodstide IS NULL or cabact_c8rfodstide IN
(select rfoade_i_rfodstide from rfoade where rfoade___rfovdeide='SECA'
and rfoade___rforefide = 'FHSJ' and rfoadervs='1'))
Discussions on the mailing list (initial post on 24/07/2009, title 'Segfault interpreting sql query') suggest it might have to do with recursion in :
65292 0x00007f2e7aa09d92 in rel_destroy_ (rel=0x2372608)
at ../../../src/server/rel_select.mx:125
125 prop_destroy(rel->p);
(gdb) list
120 exp_destroy(rel->l);
121 if (rel->p)
122 prop_destroy(rel->p);
123 } else if (rel->op == op_basetable) {
124 if (rel->p)
125 prop_destroy(rel->p);
126 } else if (is_modify(rel->op)) {
127 if (rel->r)
128 rel_destroy(rel->r);
129 }
## Comment 12126
Date: 2009-07-27 14:14:27 +0200
From: Franck Routier <<alci>>
Initial report was done using the following versions:
monetdb5-sql 2.30.2-20090629,
monetdb5-server 5.12.2-20090629
monetdb-client 1.30.2-20090629
## Comment 12127
Date: 2009-07-27 14:41:41 +0200
From: @sjoerdmullender
By changing the recursive function prop_destroy into an iterative one I was able to fix this bug.
I added the test as sql/src/test/BugTracker-2009 recursion-problem.SF-2827784.
The fix will be in the upcoming bug fix release.
## Comment 12128
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2827784 at http://sourceforge.net/support/tracker.php?aid=2827784
| Segfault evaluating query | https://api.github.com/repos/MonetDB/MonetDB/issues/2139/comments | 0 | 2020-11-30T10:23:45Z | 2024-06-27T11:40:48Z | https://github.com/MonetDB/MonetDB/issues/2139 | 753,350,933 | 2,139 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-23 04:44:02 +0200
From: Matthew W. Jones <<matburt>>
To: @njnes
Version: -- development
Last updated: 2009-08-07 09:39:00 +0200
## Comment 12118
Date: 2009-07-23 16:44:02 +0200
From: Matthew W. Jones <<matburt>>
With the latest May 2009 SP1 release I see this:
create table testdistinct (testcol varchar(10));
... inserts ...
sql>select * from testdistinct ;
+---------+
| testcol |
+=========+
| test |
| test |
| test |
| other |
| other |
| other |
+---------+
6 tuples
sql>select distinct(testcol) from testdistinct ;
+---------+
| testcol |
+=========+
| test |
| other |
+---------+
2 tuples
sql>select count(distinct(testcol)) from testdistinct ;
+----+
| L2 |
+====+
| 6 |
+----+
1 tuple
This seems incorrect, I would expect count() to return just 2.
## Comment 12119
Date: 2009-07-23 17:24:37 +0200
From: @grobian
I would expect the same.
I added the test script to CVS: sql/src/test/BugTracker-2009/count_distinct.SF-2826015.sql
## Comment 12120
Date: 2009-07-23 20:13:50 +0200
From: @njnes
select count ( distinct testcol) from testdistinct; seems to work just fine.
## Comment 12121
Date: 2009-07-23 20:36:07 +0200
From: Matthew W. Jones <<matburt>>
It does indeed work... but there should be a disparity between the two.
sql>select count(distinct testcol) from testdistinct ;
+----+
| L2 |
+====+
| 2 |
+----+
1 tuple
sql>select count(distinct(testcol)) from testdistinct ;
+----+
| L3 |
+====+
| 6 |
+----+
1 tuple
## Comment 12122
Date: 2009-07-23 20:36:48 +0200
From: Matthew W. Jones <<matburt>>
should not, rather.
## Comment 12123
Date: 2009-08-07 09:39:00 +0200
From: @njnes
added test to src/test/BugTracker-2009/Tests/count_distinct_bug.SF-2826015.sql
bug was fixed by cleanup of parser code (function handling)
## Comment 12124
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2826015 at http://sourceforge.net/support/tracker.php?aid=2826015
| count distinct incorrect | https://api.github.com/repos/MonetDB/MonetDB/issues/2138/comments | 0 | 2020-11-30T10:23:42Z | 2024-06-27T11:40:47Z | https://github.com/MonetDB/MonetDB/issues/2138 | 753,350,889 | 2,138 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-23 03:15:30 +0200
From: @skinkie
To: @njnes
Version: -- development
Last updated: 2009-08-03 11:28:33 +0200
## Comment 12105
Date: 2009-07-23 15:15:30 +0200
From: @skinkie
Some of our 'friendly' webservice users tried MonetDB in a somewhat disturbing way, resulting in a segmentation fault of the server. The query in the subject will reproduce it from mclient as well.
## Comment 12106
Date: 2009-07-23 16:35:50 +0200
From: @skinkie
The attached file basically prevents the crash, but I wonder if we can solve it higher in the tree. Output now is:
sql>select * where bla = "whatever";
!SELECT: identifier 'bla' unknown
0 tuples
So that looks good :)
## Comment 12107
Date: 2009-07-31 05:59:56 +0200
From: @drstmane
Would it be possible to provide a context diff to ensure correct automatic application (or prevent incorrect application automatic application) even in case the original code base has changed in the meantime?
Would it be possible to provide the context diff in unified format to improve human readability?
(admittedly personal taste)
Would it be possible to explain/document the details of the patch, and hence enlighten us what the origin of the actual problem was, and how it has been solved?
Thank you very much in advance!
## Comment 12108
Date: 2009-07-31 07:08:32 +0200
From: @skinkie
selection->h is not available with broken queries, must be explicitly checked before use. Removes void cast of where because the variable is in use.
The origin of the problem was a user entering a malformed query on my webinterface that segfaulted monet. The patch implements a null check to prevent n->* to be used.
## Comment 12109
Date: 2009-07-31 07:13:24 +0200
From: @drstmane
I meant origin as "what was wrong in the code".
## Comment 12110
Date: 2009-07-31 07:46:00 +0200
From: @drstmane
Just for the records:
In the development trunk, the following checkin appears to avoid the crash:
===================================================================
2009/07/25 - nielsnes: sql/src/server/rel_select.mx,1.148
added support for 'create cluster name on tname ( column list )'
which caused lots of changes all over the place.
also fixed (moved code out of sql -> mkey)
moved finally copy into over to a function and into relational alg.
===================================================================
(cf. `cvs diff -r1.14{7,8} sql/src/server/rel_select.mx` for details)
sql>select * where bla = "whatever";
!SELECT: a query should atleast have a selection and/or a from part
0 tuples
(While this problem has been reported for the development trunk (CVS HEAD) and the provided patch represents a diff with the development trunk (revision 1.147 of sql/src/server/rel_select.mx, i.e., before the above patch), the problem seems occur also in the May2009 release branch, and the patch appears to be applicable there, too.)
## Comment 12111
Date: 2009-07-31 07:54:19 +0200
From: @drstmane
ps:
1)
On the May2009 release branch, the provided patch results in
sql>select * where bla = "whatever";
!SELECT: a query should atleast have a selection and/or a from part
0 tuples
2)
On the May2009 release branch, the provided patch does not appear to "harm" any of the tests in the sql repository,
i.e., after applying the patch, `Mtest.py -r` reports the same as it did before.
## Comment 12112
Date: 2009-07-31 07:56:06 +0200
From: @drstmane
Correction:
1)
On the May2009 release branch, the provided patch results in
sql>select * where bla = "whatever";
!SELECT: identifier 'bla' unknown
0 tuples
## Comment 12113
Date: 2009-07-31 07:58:42 +0200
From: @skinkie
I see that the provided patch should not check for selection->h but selection like Niels did. Both work probably because ->h is the first element of the struct.
## Comment 12114
Date: 2009-07-31 08:59:26 +0200
From: @drstmane
For the records:
===================================================================
2009/07/31 - nielsnes: sql/src/server/rel_select.mx,1.143.2.5
fix crash with select * from 'whatever' = bla
===================================================================
2009/07/31 - stmane: sql/src/server/rel_select.mx,1.143.2.6
"selection" might be NULL in this case;
hence, we postpone dereferencing it
until we are sure it is not NULL.
NOTE:
this and Niels' previous checkin might (will)
conflict with
"
2009/07/25 - nielsnes: sql/src/server/rel_select.mx,1.148
added support for 'create cluster name on tname ( column list )'
which caused lots of changes all over the place.
also fixed (moved code out of sql -> mkey)
moved finally copy into over to a function and into relational alg.
"
during propagation;
in fact, the above mentioned checkin on the development trunk
might (does?) subsume these two checkin on the May2009 branch
(needs to be verified!)
===================================================================
## Comment 12115
Date: 2009-07-31 09:46:57 +0200
From: @drstmane
added test in May2009 release branch (sql/src/test/BugTracker-2009/Tests/select_X_where_bla_EQ_whatever.SF-2825974.sql
The report can be closed, once the test has been propagated to the development trunk,
and the code fix has been propagated to the development trunk, or the alternative fix on the development trunk has been verified (see comments below),
and nightly testing has confirmed the fix(es) for both the May2009 release branch (or its successor Aug2009) and the development trunk.
## Comment 12116
Date: 2009-08-03 11:28:33 +0200
From: @drstmane
fix and test have been propagated to the development trunk
and testing confirmed the fix.
closing.
## Comment 12117
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2825974 at http://sourceforge.net/support/tracker.php?aid=2825974
| select * where bla = "whatever" crashes | https://api.github.com/repos/MonetDB/MonetDB/issues/2137/comments | 0 | 2020-11-30T10:23:39Z | 2024-06-27T11:40:46Z | https://github.com/MonetDB/MonetDB/issues/2137 | 753,350,847 | 2,137 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-22 11:01:59 +0200
From: heangu <<heangu>>
To: @njnes
Version: -- development
Last updated: 2009-09-17 10:36:47 +0200
## Comment 12091
Date: 2009-07-22 11:01:59 +0200
From: heangu <<heangu>>
I opened two sessions at the same time, use session one to insert data, session two execute START TRANSACTION but without COMMIT. When restart, I found the data just inserted was lost!! MY test case is as follows:
1. Start the server.
2. Open one session. mclient -lsql -dtest. (Denote it as S1)
3. Open the second session as step 2. (Denote it as S2)
4. IN S1:
CREATE TABLE t(a int);
CREATE TABLE x(a int);
INSERT INTO t VALUES(1);
INSERT INTO t SELECT * FROM t;
INSERT INTO t SELECT * FROM t;
INSERT INTO t SELECT * FROM t;
INSERT INTO t SELECT * FROM t;
INSERT INTO t SELECT * FROM t;
INSERT INTO t SELECT * FROM t;
INSERT INTO t SELECT * FROM t;
INSERT INTO t SELECT * FROM t;
INSERT INTO t SELECT * FROM t;
INSERT INTO t SELECT * FROM t;
5. IN S2:
START TRANSACTION;
6. IN S1:
INSERT INTO x VALUES(1);
INSERT INTO x VALUES(2);
7. IN S2:
COMMIT;
UPDATE t SET a = 8;
8. WAIT a few minutes.
9. KILL server and start it.
10. Open session S1. (mclient -lsql -dtest)
11. IN S1:
SELECT * FROM x;
Here is the problem. I should have got two lines (1 and 2) in STEP 11, but in fact I only got an empty set.
Where has my data gone?
It can't be other person who deleted it. And I can have it occur many times.
## Comment 12092
Date: 2009-07-22 11:09:50 +0200
From: @grobian
I can reproduce the problem
## Comment 12093
Date: 2009-07-23 05:38:32 +0200
From: heangu <<heangu>>
I think the http://sourceforge.net/support/tracker.php?aid=2819966 is possible because of this data lost. This bug is found while trying to reproduce the http://sourceforge.net/support/tracker.php?aid=2819966 that I reported a few days ago.
In this case, if I execute an update on x before restarting server, older versions of MONET will complain ASSERT failed in void_inplace. But new versions of MONET seems OK but data lost.
## Comment 12094
Date: 2009-07-27 16:15:19 +0200
From: Matthew W. Jones <<matburt>>
I am experiencing widespread data corruption when restarting the May2009-SP1 MonetDB release also, but mine is much simpler.
On 64-bit ubuntu:
Linux dsvm01 2.6.24-23-server 1 SMP Wed Apr 1 22:14:30 UTC 2009 x86_64 GNU/Linux
With a table defined as follows:
CREATE TABLE "tablespace"."test_table" (
"col1" varchar(50),
"col2" varchar(50),
"col3" varchar(50),
"col4" varchar(50),
"col5" varchar(50),
"col6" int,
"col7" timestamp(7),
"col8" int,
"col9" int,
"col10" int
);
I imported 20998687 records into this table with all columns populated and ran a couple of queries to check the data:
select count(*) from raw_plays ;
[ 20998687 ]
select count(*) from raw_plays where col1 = '' or col2 = '' or col3 = '' ;
[ 0 ]
select max(col7), min(col7) from test_table ;
[ 2009-07-25 14:17:15.538000, 2009-07-24 15:29:12.524000 ]
During the time I loaded the data and ran these queries I did not restart the database, it had been up for about 3 days I then issued:
monetdb stop db
and
monetdb start db
and re-ran the queries:
select count(*) from raw_plays ;
[ 20998687 ]
select count(*) from raw_plays where col1 = '' or col2 = '' or col3 = '' ;
[ 20852061 ]
select max(col7), min(col7) from test_table ;
[ 2009-07-25 14:17:15.538000, -1-01-01 [000000000000](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=000000000000) ]
This is some serious data corruption, I've lost data in the vast majority of these records!
I wasn't sure if this is the same issue that is seen here, if need-be I can open a new bug report.
## Comment 12095
Date: 2009-07-27 16:19:47 +0200
From: Matthew W. Jones <<matburt>>
"test_table" and "raw_plays" are the same in this case, I did not do a good enough job cleaning up the names before posting here.
## Comment 12096
Date: 2009-07-29 16:03:42 +0200
From: Matthew W. Jones <<matburt>>
Adding gdk_single_user=yes to my configuration file fixed my database corruption issue.
## Comment 12097
Date: 2009-07-30 07:22:04 +0200
From: heangu <<heangu>>
I can fix it by adding some code in store_manager to prevent it from syncing data to disk if any IBATs in gtrans is not empty. Thus kept the log in disk to let it recover from it on next startup.
## Comment 12098
Date: 2009-07-30 07:34:10 +0200
From: @drstmane
heangu,
could you please share your patch with us,
e.g., by attaching it to this bug report?
Thanks!
Stefan
## Comment 12099
Date: 2009-08-03 23:21:19 +0200
From: @njnes
indeed the non empty ibats is a problem.
## Comment 12100
Date: 2009-08-06 21:20:33 +0200
From: @njnes
fixed by first aplying all the ibat inserts on to the central bats, before a logger_restart.
Not closing jet as we need a test for this
## Comment 12101
Date: 2009-08-11 13:53:09 +0200
From: @drstmane
heangu,
could you possibly test whether the latest Aug2009 release candidate of MonetDB --- available from CVS as Aug2009 branches or as nightly built super source ball from http://monetdb.cwi.nl/testing/projects/monetdb/Stable/.DailyBuilds./0_latest/SourceSuperBall/ --- fixes your problem?
Thanks!
Stefan
## Comment 12102
Date: 2009-08-14 14:09:53 +0200
From: @drstmane
tagged subject
## Comment 12103
Date: 2009-08-15 12:41:54 +0200
From: heangu <<heangu>>
Ok. I've tested it. The bug has been fixed.
## Comment 12104
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2825243 at http://sourceforge.net/support/tracker.php?aid=2825243
| SQL: Data lost when restart | https://api.github.com/repos/MonetDB/MonetDB/issues/2136/comments | 0 | 2020-11-30T10:23:35Z | 2024-06-27T11:40:45Z | https://github.com/MonetDB/MonetDB/issues/2136 | 753,350,804 | 2,136 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-17 04:38:10 +0200
From: @skinkie
To: @njnes
Version: -- development
Last updated: 2009-07-20 08:41:40 +0200
## Comment 12084
Date: 2009-07-17 04:38:10 +0200
From: @skinkie
ML quote:
---
The second point is that my string is 99 characters long, and utf-8 with
accented characters. So its length is less than 100 characters but more
than 100 bytes.
---
Since we already have an implementation of UTF8 checking in atoms/str.mx; we ought to use this one upon insert. Never the less I think that the proposed 'faster check' might give some benefit. http://www.daemonology.net/blog/2008-06-05-faster-utf8-strlen.html
Test:
create table utf8len(a varchar(1));
copy 1 records into utf8len from '/tmp/emptyaccount.utf8';
copy 1 records into utf8len from '/tmp/valueformoney.utf8';
insert into utf8len values ('€');
select a, length(a) AS len from utf8len;
select 'Liever €uro' as "Liever euro";
drop table utf8len;
emptyaccount.utf8:
0
valueformoney.utf8:
€
Excepted output:
sql>create table utf8len(a varchar(1));
0 tuples
sql>copy 1 records into utf8len from '/tmp/emptyaccount.utf8';
Rows affected 1
sql>copy 1 records into utf8len from '/tmp/valueformoney.utf8';
Rows affected 1
sql>insert into utf8len values ('€');
Rows affected 1
sql>select a, length(a) AS len from utf8len;
+---+-----+
| a | len |
+===+=====+
| 0 | 1 |
| € | 1 |
| € | 1 |
+---+-----+
3 tuples
sql>select 'Liever €uro' as "Liever euro";
+-------------+
| Liever euro |
+=============+
| Liever €uro |
+-------------+
1 tuple
sql>drop table utf8len;
0 tuples
I have ran testing, 4 extra test came out differently. If someone could take a peak if the difference might be good, it would be helpful.
## Comment 12085
Date: 2009-07-17 17:13:09 +0200
From: @skinkie
The new patch fixes the issues related to bats or singular values that resulted in str_nil. The patch also improves the rendering (nifty feature). If NULL values are found the minimum column width will be made four. This means the column will never be too small and gets into 'odd vertical rendering'.
This patch will break a lot of test or at least they should be allowed correct mostly 1 -> 4 changes.
## Comment 12086
Date: 2009-07-17 17:17:17 +0200
From: @grobian
that feature is not correct
it is a rendering issue of the client, that needs fixing there
## Comment 12087
Date: 2009-07-17 17:25:59 +0200
From: @skinkie
No it is undoubtably an error in the server observe:
SELECT NULL;
vs
CREATE TABLE test (a int);
INSERT INTO test VALUES (null);
SELECT * FROM TEST;
DROP TABLE test;
sql>SELECT NULL;
% . table_name
% single_value name
% char type
% 0 length
[ NULL ]
sql>SELECT * FROM TEST;
% sys.test table_name
% a name
% int type
% 1 length
[ NULL ]
## Comment 12088
Date: 2009-07-17 17:29:38 +0200
From: @grobian
we agree that the current behaviour is weird
we disagree that we should return a width of 4 for NULL-values, because a client needs not to render this "absent" value as the string "NULL".
## Comment 12089
Date: 2009-07-20 20:41:40 +0200
From: @njnes
a test was added and the patch was applied. Thanks go to stefan for creating it.
## Comment 12090
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2822855 at http://sourceforge.net/support/tracker.php?aid=2822855
| utf8 real size upon insert | https://api.github.com/repos/MonetDB/MonetDB/issues/2135/comments | 0 | 2020-11-30T10:23:31Z | 2024-06-27T11:40:44Z | https://github.com/MonetDB/MonetDB/issues/2135 | 753,350,747 | 2,135 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-16 09:29:12 +0200
From: Tomas <<tomnvc>>
To: @njnes
Version: -- development
Last updated: 2009-11-22 10:05:31 +0100
## Comment 12076
Date: 2009-07-16 09:29:12 +0200
From: Tomas <<tomnvc>>
Database:
MonetDB server v5.12.2, based on kernel v1.30.2
Serving database 'Test1', using 1 thread
Compiled for i686-pc-win32/32bit with 32bit OIDs dynamically linked
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2009 MonetDB B.V., all rights reserved
Visit http://monetdb.cwi.nl/ for further information
warning: please don't forget to set your vault key!
(see C:\MonetDB\engine\etc\monetdb5.conf)
Listening for connection requests on mapi:monetdb://127.0.0.1:50000/
Schema and table:
CREATE SCHEMA "test" AUTHORIZATION "monetdb";
SET SCHEMA "test";
CREATE TABLE "test"."tb1" (
"id" varchar(32),
"pan" varchar(19),
"dt_month" smallint,
"dt_year" int,
"amount" int,
"merchant" varchar(128),
"store" varchar(128),
"dt" timestamp(7)
);
To evaluate MonetDB potential performance we created simple table to simulate possible data. Given table was populated with random data where:
ID is totally unique (Guid in HEX format),
pan is truncated primary account number in format like this 411111******1111
dt_month and dt_year simulate month and year with 6 (2004 to 2009) different values for year
amount is random value from 10 to 981256.
Merchant is 100 different string values.
Store is 10000 different string values.
dt is time.
As you can see we wanted to test different value variance on different columns and insert/select query performance.
Problem is that server starts to crash after about 11-12 million records have been added. We tested it using sql insert and copy records into, both ways result in crash.
Same problem occurred on two different machines:
One is Windows XP with 2Gb ram single core cpu, Other is Windows XP with 1.5Gb ram dual core cpu. In both cases number of records for first crashes is about 11-12 millions. The more records we have the more likely crash to occur. Sometimes it crashes so that server does not accept any sql commands any more. It exits upon client connection...
We execute our tests from .NET thru wrapped native mapi library.
Please fix this… I liked its performance up to 10 mill records, but I want to test it with 500 mill records and more! If it works well... we might even develop native ADO.NET provider for it :)
## Comment 12077
Date: 2009-07-16 13:36:05 +0200
From: Tomas <<tomnvc>>
I have attached c project to reproduce the problem (MonetDbCrash.rar)
Please add DBEngine folder near executable file and add this files to it:
iconv.dll, libbat.dll, libeay32.dll, libmapi.dll, libmonetdb5.dll, libmutils.dll, libstream.dll, libxml2.dll, pcre.dll, pthreadvc2.dll, ssleay32.dll, zlib1.dll
## Comment 12078
Date: 2009-07-16 15:09:37 +0200
From: Tomas <<tomnvc>>
We just reproduced same problem on 64bit version of Windows 2003 Server , Computer has 3Gb of memory. It took 30mil records to crash it thought…
## Comment 12079
Date: 2009-07-19 11:46:57 +0200
From: @njnes
This is very likely a 32bit monetdb problem. Could you run a 64 bit version of monetdb on your 64 bit windows server? For you 500M test case we recommend using a 64 bit version.
## Comment 12080
Date: 2009-07-21 07:58:58 +0200
From: Tomas <<tomnvc>>
It seems that 64 bit version does not have this problem. 200M+ records did not crash server and data size on disk was over 25 Gigs. We will have to set up larger environment for our tests and most likely we’ll go for 1000M to 2000M records, not 500M initially intended. I guess it is ok, we would have to go for 64 bit version any way, but still… it would be nice if 32 bit version wouldn’t crash! We would have more confidence using this very nice product :)
## Comment 12081
Date: 2009-08-14 13:56:22 +0200
From: @drstmane
tagged subject
## Comment 12082
Date: 2009-11-22 22:05:31 +0100
From: @njnes
closing, its a duplicate of all the bugs related to 32 bit. We are working on improvements (ie proper errors).
## Comment 12083
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2822300 at http://sourceforge.net/support/tracker.php?aid=2822300
This bug was marked DUPLICATE in the database it was moved from.
Changing resolution to "MOVED"
| SQL: Server crashes silently | https://api.github.com/repos/MonetDB/MonetDB/issues/2134/comments | 0 | 2020-11-30T10:23:27Z | 2024-06-28T13:23:45Z | https://github.com/MonetDB/MonetDB/issues/2134 | 753,350,705 | 2,134 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-14 07:41:53 +0200
From: @skinkie
To: GDK devs <<bugs-common>>
Version: -- development
CC: mrunal.gawade, @skinkie
Last updated: 2012-11-27 17:18:09 +0100
## Comment 12073
Date: 2009-07-14 19:41:53 +0200
From: @skinkie
select 'bla';
bla
select '"'||'bla';
"\"bla"
Now what would be the way to get my string just surrounded by " "?
## Comment 12074
Date: 2009-07-14 20:30:37 +0200
From: @sjoerdmullender
The CSV formatter is pretty primitive, but it does the job.
The renderer does not look at the type of the columns, only at the content of the fields. If a value contains the separator (i.e. comma or tab, depending on the mode) or if it contains a newline or a double quote, it will surround the value with double quotes, otherwise it won't. This should be enough for most purposes (although I can imagine that any white space should trigger the quotes).
As far as I know, there is no "standard" for CSV data (i.e. there are various variations). If you know of a standard, please let us know.
I'm closing this as Invalid, but feel free to open a Feature Request. If you do, please be clear in what *exactly* you would like to see changed so that we don't have to do any second guessing.
## Comment 12075
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2821456 at http://sourceforge.net/support/tracker.php?aid=2821456
The original assignee of this bug does not have
an account here. Reassigning to the default assignee
for the component, bugs-common@monetdb.org.
Previous assignee was nobody@users.sourceforge.net.
## Comment 15387
Date: 2011-01-03 13:39:45 +0100
From: @skinkie
I was pointed by a guy on 'the standard'.
http://www.ietf.org/rfc/rfc4180.txt
7. If double-quotes are used to enclose fields, then a double-quote
appearing inside a field must be escaped by preceding it with
another double quote. For example:
"aaa","b""bb","ccc"
## Comment 18156
Date: 2012-11-27 17:02:53 +0100
From: mrunal.gawade
Seems to be working fine now. Has a test. strange_escaping_in_csv.Bug-2133
## Comment 18159
Date: 2012-11-27 17:18:09 +0100
From: @skinkie
Interesting indeed! Great it is fixed :)
| Strange " escaping in -f csv | https://api.github.com/repos/MonetDB/MonetDB/issues/2133/comments | 0 | 2020-11-30T10:23:25Z | 2024-06-27T11:40:42Z | https://github.com/MonetDB/MonetDB/issues/2133 | 753,350,663 | 2,133 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-14 12:27:46 +0200
From: Floris Ouwendijk <<floris233>>
To: Jan Rittinger <<jan.rittinger>>
Version: -- development
Last updated: 2009-07-14 06:04:01 +0200
## Comment 12069
Date: 2009-07-14 12:27:46 +0200
From: Floris Ouwendijk <<floris233>>
The following query took extremely long to execute:
doc("bookmarks.xml")/bookmarks/folder[@name='builtin']/folder[@name='Standard Query Blocks']/folder[@name='Files']/folder[@name='File extensions']/bookmark-definition[@name="Microsoft Office files"]/query/*
Running /opt/MonetDB-May2009-SP1/bin/pf -TA produced the following output:
parsing: 000ms
module import: 000ms
normalization: 000ms
semantical analysis: 000ms
XML Schema import: 000ms
core mapping: 000ms
core simplification: 000ms
type checking: 008ms
core tree optimization: 012ms
logical algebra tree generation: 001ms
map to unique column names: 002ms
constant optimization: 000ms
icol optimization: 000ms
key optimization: 000ms
complex optimization: 010ms
general optimization: 000ms
required value optimization: 000ms
general optimization: 000ms
equi-join pushdown: 002ms
icol optimization: 000ms
set optimization: 000ms
icol optimization: 000ms
general optimization: 000ms
structural CSE optimization: 000ms
complex optimization: 001ms
set optimization: 000ms
required value optimization: 000ms
rank optimization: 000ms
constant optimization: 000ms
key optimization: 000ms
required nodes optimization: 000ms
introduce proxy operators: 000ms
mvd optimization: 005ms
thetajoin optimization: 011ms
resolve proxy operators: 000ms
equi-join pushdown: 075ms
icol optimization: 001ms
set optimization: 003ms
step-join push-down: 000ms
icol optimization: 000ms
general optimization: 001ms
projection removal: 001ms
structural CSE optimization: 198ms
complex optimization: 18m 37s 415ms
set optimization: 000ms
required value optimization: 000ms
rank optimization: 000ms
introduce proxy operators: 000ms
mvd optimization: 000ms
thetajoin optimization: 000ms
resolve proxy operators: 000ms
equi-join pushdown: 000ms
icol optimization: 000ms
set optimization: 000ms
step-join push-down: 000ms
icol optimization: 000ms
constant optimization: 000ms
key optimization: 000ms
general optimization: 000ms
projection removal: 000ms
structural CSE optimization: 000ms
complex optimization: 006ms
set optimization: 000ms
required value optimization: 000ms
rank optimization: 000ms
MonetDB specific optimizations: 000ms
complex optimization: 001ms
general optimization: 000ms
complete property inference: 000ms
logical algebra optimization: 18m 37s 785ms
CSE in logical algebra tree: 000ms
compilation to physical algebra: 000ms
introduction of recursion boundaries: 000ms
MIL code generation: 000ms
dead code elimination: 000ms
MIL code serialization: 000ms
----------------------------------------------
overall compilation time: 18m 37s 813ms
18mins is a bit much for a query describing exactly what path to follow. Is there, in the meantime, a better method to execute te query?
## Comment 12070
Date: 2009-07-14 12:30:46 +0200
From: @drstmane
Guess this one is for Jan ...
## Comment 12071
Date: 2009-07-14 18:04:01 +0200
From: Jan Rittinger <<jan.rittinger>>
Fixed in CVS -- now compiles within 170 msec.
(The problem was that we didn't remove unused constant expressions---attach operators---timely, which resulted in hundreds of cross products appearing in the plan. The property inference in the complex optimization phase than took quite some time before the rewrites threw away the superfluous cross products.)
## Comment 12072
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2821251 at http://sourceforge.net/support/tracker.php?aid=2821251
| PF: Complex optimization overly complex | https://api.github.com/repos/MonetDB/MonetDB/issues/2132/comments | 0 | 2020-11-30T10:23:21Z | 2024-06-27T11:40:41Z | https://github.com/MonetDB/MonetDB/issues/2132 | 753,350,609 | 2,132 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-11 07:41:27 +0200
From: z lh <<zlh>>
To: @mlkersten
Version: -- development
Last updated: 2009-07-20 04:39:59 +0200
## Comment 12065
Date: 2009-07-11 07:41:27 +0200
From: z lh <<zlh>>
create function test(a int) returns int
begin
if a>3
then return 1;
end if;
return 4;
end;
select test(4);
problem: can't return function, in if else statement, the expected result is 1 not 4
note: optimizer inline problem
create function te(a int) returns int
begin
declare b int;
set b=0;
if a>3
then set b= 1;
elseif a>2
then set b=2;
end if;
return b;
end;
select te(3);
+-----------------+
| te_single_value |
+=================+
| 0 |
+-----------------+
problem: return result not correct, the expected result is 2 not 0
## Comment 12066
Date: 2009-07-19 13:34:30 +0200
From: @njnes
seems to be caused by the inline optimizer. It rewrites the return statements incorrectly.
## Comment 12067
Date: 2009-07-20 16:39:59 +0200
From: @mlkersten
The code has now been protected against multiple short-cut exits in inlined functions.
Such functions are never inlined.
## Comment 12068
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2819972 at http://sourceforge.net/support/tracker.php?aid=2819972
| function problom ----return and elseif | https://api.github.com/repos/MonetDB/MonetDB/issues/2131/comments | 0 | 2020-11-30T10:23:18Z | 2024-06-27T11:40:40Z | https://github.com/MonetDB/MonetDB/issues/2131 | 753,350,563 | 2,131 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-11 07:30:39 +0200
From: heangu <<heangu>>
To: GDK devs <<bugs-common>>
Version: -- development
Last updated: 2009-07-20 08:38:31 +0200
## Comment 12062
Date: 2009-07-11 07:30:39 +0200
From: heangu <<heangu>>
The server crashed on startup complaining assertion failed in void_inplace. I checked the data in disk, only to find that in one column of my table(not system table), UBAT's count is 40 but the main BAT's disk file is empty!! This happend three months ago and have never occured since until a few days ago. And I tried many time to reproduce it but failed.
## Comment 12063
Date: 2009-07-20 20:38:31 +0200
From: @mlkersten
The information might have been in the log file.
Without further information about repeatability, we keep our eyes open and close this bug report for now.
## Comment 12064
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2819968 at http://sourceforge.net/support/tracker.php?aid=2819968
The original assignee of this bug does not have
an account here. Reassigning to the default assignee
for the component, bugs-common@monetdb.org.
Previous assignee was nobody@users.sourceforge.net.
| Assertion failed when startup | https://api.github.com/repos/MonetDB/MonetDB/issues/2130/comments | 0 | 2020-11-30T10:23:15Z | 2024-06-28T13:23:45Z | https://github.com/MonetDB/MonetDB/issues/2130 | 753,350,525 | 2,130 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-11 07:22:28 +0200
From: z lh <<zlh>>
To: GDK devs <<bugs-common>>
Version: -- development
Last updated: 2009-07-20 08:35:57 +0200
## Comment 12058
Date: 2009-07-11 07:22:28 +0200
From: z lh <<zlh>>
sql>create table t(i int);
Timer 11.531 msec 0 rows
sql>create table x(a int);
Timer 2.792 msec 0 rows
sql>insert into t values(1);
Rows affected 1
Timer 33.536 msec 0 rows
sql>insert into t select i+count(i) from t;
Rows affected 1
Timer 2.659 msec 0 rows
sql>insert into t select i+count(i) from t;
Rows affected 2
Timer 2.461 msec 0 rows
sql>insert into t select i+count(i) from t;
Rows affected 4
Timer 2.266 msec 0 rows
sql>insert into t select i+count(i) from t;
Rows affected 8
Timer 10.562 msec 0 rows
sql>insert into t select i+count(i) from t;
Rows affected 16
Timer 2.193 msec 0 rows
sql>insert into t select i+count(i) from t;
Rows affected 32
Timer 2.254 msec 0 rows
sql>insert into t select i+count(i) from t;
Rows affected 64
Timer 2.396 msec 0 rows
sql>insert into t select i+count(i) from t;
Rows affected 128
Timer 2.265 msec 0 rows
sql>insert into t select i+count(i) from t;
Rows affected 256
Timer 8.952 msec 0 rows
sql>insert into t select i+count(i) from t;
Rows affected 512
Timer 10.674 msec 0 rows
sql>insert into t select i+count(i) from t;
Rows affected 1024
Timer 2.563 msec 0 rows
sql>insert into t select i+count(i) from t;
Rows affected 2048
Timer 2.994 msec 0 rows
sql>insert into t select i+count(i) from t;
Rows affected 4096
Timer 3.305 msec 0 rows
sql>insert into x select * from t;
Rows affected 8192
Timer 8.288 msec 0 rows
restart the server
sql>update x set a = a+1 where a<1000 and a>500;
Rows affected 499
Timer 7.050 msec
sql>update x set a = a+1 where a<5000 and a>4500;
Rows affected 499
Timer 11.554 msec
restart the server
sql>delete from x where a > 800 and a < 2000;
Rows affected 1200
Timer 7.002 msec
restart the server
sql>delete from x;
Rows affected 6992
Timer 4.455 msec
restart the server
sql>insert into x select * from t;
Rows affected 8192
Timer 175010.603 msec
sql>
restart the server
sql>select count(*) from t;
+--------+
| count_ |
+========+
| 8192 |
+--------+
Timer 5849.920 msec 1 rows
sql>update x set a = a+100;
Rows affected 8192
Timer 2000.985 msec 1 rows
sql>delete from x;
restart the server
sql>insert into x values(1);
Rows affected 1
restart the server, but the server crashed on startup
the error message is
mserver5: gdk_bat.c:1856: void_inplace: Assertion `BATcount(b) > (id -b->H->seq)' failed.
(gdb) bt
0 0x0000003a0a430045 in raise () from /lib64/libc.so.6
1 0x0000003a0a431ae0 in abort () from /lib64/libc.so.6
2 0x0000003a0a429756 in __assert_fail () from /lib64/libc.so.6
3 0x00002aaaaba198cf in void_inplace (b=0xa5669a8, id=65553, val=0xafd8928, force=1 '\001') at gdk_bat.c:1856
4 0x00002aaaaba19c2f in void_replace_bat (b=0xa5669a8, u=0xa571a48, force=1 '\001') at gdk_bat.c:1881
5 0x00002aaabdf0d220 in load_bat (bat=0xaf87c48, type=6, sz=1024, cnt=0xaf879b0) at bat_storage.c:470
6 0x00002aaabdf0d6f1 in create_col (tr=0xadbfa48, c=0xaf87b28) at bat_storage.c:586
7 0x00002aaabdef62f0 in load_column (tr=0xadbfa48, t=0xaf87978, rid=172) at store.c:501
8 0x00002aaabdef6871 in load_table (tr=0xadbfa48, s=0xadbfab8, rid=172) at store.c:568
9 0x00002aaabdef7d61 in load_schema (tr=0xadbfa48, id=1310, rid=31) at store.c:835
10 0x00002aaabdef813a in load_trans (tr=0xadbfa48, id=1310) at store.c:891
11 0x00002aaabdefa2d9 in store_init (debug=0, store=store_bat,
logdir=0xa585398 "/home/tzf/opt/MonetDB/var/MonetDB5/sql_logs", dbname=0xa585228 "tan", stk=0)
at store.c:1350
12 0x00002aaabded15c5 in mvc_init (debug=0, store=store_bat, stk=0) at sql_mvc.c:37
13 0x00002aaabde68975 in SQLinit (c=0x6049c0) at sql_scenario.c:163
14 0x00002aaaaab01074 in initScenario (c=0x6049c0, s=0x2aaaaad1e570) at mal_scenario.c:74
15 0x00002aaaaab01c4d in setScenario (c=0x6049c0, nme=0xaadb603 "sql") at mal_scenario.c:295
16 0x00002aaaaaac8718 in MSscheduleClient (command=0xaadb5c8 "LIT", challenge=0x41000f80 "UE7a1Mgwag",
fin=0xa94c9e0, fout=0xaad72c0) at mal_session.c:270
17 0x00002aaabb9acd8b in doChallenge (cmd=0x0, in=0xaa50f90, out=0xa54c720) at mal_mapi.c:108
18 0x00002aaabb9ad045 in SERVERlistenThread (Sock=0xaa4fef8) at mal_mapi.c:190
19 0x0000003a0b0062e7 in start_thread () from /lib64/libpthread.so.0
20 0x0000003a0a4ce3bd in clone () from /lib64/libc.so.6
I could not reproduce the bug, but I encountered many times.
note:I spoted the count of ubat is bigger than the count of primary bat in disk.
## Comment 12059
Date: 2009-07-13 03:12:03 +0200
From: heangu <<heangu>>
It's really a bug. But seems strange.
## Comment 12060
Date: 2009-07-20 20:35:57 +0200
From: @mlkersten
Bug not reproducable, because the correct reaction is :
insert into t select i+count(i) from t;
!SELECT: cannot use non GROUP BY column in query results without an aggregate function
close the report for now as no fix and no test file
## Comment 12061
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2819966 at http://sourceforge.net/support/tracker.php?aid=2819966
The original assignee of this bug does not have
an account here. Reassigning to the default assignee
for the component, bugs-common@monetdb.org.
Previous assignee was nobody@users.sourceforge.net.
| can't start server | https://api.github.com/repos/MonetDB/MonetDB/issues/2129/comments | 0 | 2020-11-30T10:23:11Z | 2024-06-28T13:23:44Z | https://github.com/MonetDB/MonetDB/issues/2129 | 753,350,472 | 2,129 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-09 11:51:00 +0200
From: @grobian
To: @mlkersten
Version: -- development
Last updated: 2009-07-20 06:33:24 +0200
## Comment 12055
Date: 2009-07-09 11:51:00 +0200
From: @grobian
MonetDB server v5.13.0, based on kernel v1.31.0
Serving database 'demo', using 1 thread
Compiled for x86_64-pc-solaris2.11/64bit with 64bit OIDs dynamically linked
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2009 MonetDB B.V., all rights reserved
Visit http://monetdb.cwi.nl/ for further information
warning: please don't forget to set your vault key!
(see /export/scratch/stripe/fabian/monetdb/current/program-i86pc/etc/monetdb5.conf)
[New LWP 3]
Listening for connection requests on mapi:monetdb://127.0.0.1:50000/
>include remote
!SyntaxException:parseError:
!';' expected
>include remote;
!SyntaxException:user.main[0]:RETURN missing
>
Program received signal SIGSEGV, Segmentation fault.
0xfffffd7fa22182f1 in varGetProp (mb=0x6651a8, var=0, prop=0)
at /export/scratch/mirror/fabian/monetdb/current/five/src/mal/mal_instruction.mx:3005
3005 for(i=0; i<v->propc; i++){
(gdb) bt
0 0xfffffd7fa22182f1 in varGetProp (mb=0x6651a8, var=0, prop=0)
at /export/scratch/mirror/fabian/monetdb/current/five/src/mal/mal_instruction.mx:3005
1 0xfffffd7f95c09cca in MALoptimizer (c=0x605f80)
at /export/scratch/mirror/fabian/monetdb/current/five/src/optimizer/opt_support.mx:699
2 0xfffffd7fa226049a in runPhase (c=0x605f80, phase=2)
at /export/scratch/mirror/fabian/monetdb/current/five/src/mal/mal_scenario.mx:602
3 0xfffffd7fa22605c3 in runScenarioBody (c=0x605f80)
at /export/scratch/mirror/fabian/monetdb/current/five/src/mal/mal_scenario.mx:641
4 0xfffffd7fa22607bb in runScenario (c=0x605f80)
at /export/scratch/mirror/fabian/monetdb/current/five/src/mal/mal_scenario.mx:672
5 0xfffffd7fa221effe in MSserveClient (dummy=0x605f80)
at /export/scratch/mirror/fabian/monetdb/current/five/src/mal/mal_session.mx:495
6 0x0000000000403479 in main (argc=1, av=0xfffffd7fffdfe848)
at /export/scratch/mirror/fabian/monetdb/current/five/src/tools/mserver5.mx:622
(gdb)
## Comment 12056
Date: 2009-07-20 18:33:24 +0200
From: @mlkersten
Fixed by patching the mal_session.mx file.
## Comment 12057
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2818931 at http://sourceforge.net/support/tracker.php?aid=2818931
| MAL: segv on mortal soul's syntax error | https://api.github.com/repos/MonetDB/MonetDB/issues/2128/comments | 0 | 2020-11-30T10:23:08Z | 2024-06-27T11:40:37Z | https://github.com/MonetDB/MonetDB/issues/2128 | 753,350,427 | 2,128 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-08 03:04:07 +0200
From: @skinkie
To: @sjoerdmullender
Version: -- development
Last updated: 2009-07-20 09:25:30 +0200
## Comment 12051
Date: 2009-07-08 15:04:07 +0200
From: @skinkie
Using the default output a user would expect the following to work:
mclient -u monetdb -P monetdb
because it is explained as:
-u user | --user=user user id
But the actual implementation requires to drop the space between -u and user. And likewise for -P.
My suggestion would be to change the -u user help output to: -u[user] or allow the space between the option.
## Comment 12052
Date: 2009-07-08 15:13:26 +0200
From: @sjoerdmullender
I changed the help text to this:
-P[passwd] | --passwd[=passwd] password (if no passwd given, ask for one)
-u[user] | --user[=user] user id (if no user id given, ask for one)
Happy?
## Comment 12053
Date: 2009-07-20 21:25:30 +0200
From: @mlkersten
Closed since the requested feature was fullfilled.
## Comment 12054
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2818536 at http://sourceforge.net/support/tracker.php?aid=2818536
| improve mclient -u / -p help | https://api.github.com/repos/MonetDB/MonetDB/issues/2127/comments | 0 | 2020-11-30T10:23:05Z | 2024-06-27T11:40:36Z | https://github.com/MonetDB/MonetDB/issues/2127 | 753,350,385 | 2,127 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-07 09:38:54 +0200
From: Franck Routier <<alci>>
To: @njnes
Version: -- development
Last updated: 2009-08-06 09:33:17 +0200
## Comment 12046
Date: 2009-07-07 21:38:54 +0200
From: Franck Routier <<alci>>
The numeric type in Monetdb has two shortcomings :
- numeric is an arbitrary precision number. Imposing a limitation of 18 digits in the implementation is a bit restrictive (it is ok, but most rdbms will have much higher precision)
- when defining the type NUMERIC(P,S), P+S must be <= 18. But in fact, P should be the total number of digits, so if the limit is 18, we should have P <= 18 (and of course S<P)
## Comment 12047
Date: 2009-07-20 20:44:47 +0200
From: @njnes
could you please add the failing prepare/execute statements (including ddl statements). Its hard to reproduce bugs without test scripts.
## Comment 12048
Date: 2009-07-21 09:27:07 +0200
From: Franck Routier <<alci>>
simply try this :
sql>create table numtest (id integer, val numeric(18,8));
You will see :
!decimal(18,8) isn't supported because P=18 + S=8 > 18 in: "create table numtest (id integer, val numeric(18,8)"
!syntax error, unexpected ')' in: ")"
0 tuples
As you see, you cannot create numeric(18,8), as monetdb checks that P+S <= 18.
But in the sql numeric type, P stands for the TOTAL precision of the number, and S the number of digits for the decimal part. So numeric(18,8) should be valid, meaning a number with 10 digits for the floor, and 8 digits for the decimal part...
## Comment 12049
Date: 2009-08-06 21:33:17 +0200
From: @njnes
fixed the numeric type
added test to BugTracker-2009
## Comment 12050
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2818176 at http://sourceforge.net/support/tracker.php?aid=2818176
| NUMERIC data type not following sql standard | https://api.github.com/repos/MonetDB/MonetDB/issues/2126/comments | 0 | 2020-11-30T10:23:01Z | 2024-06-27T11:40:35Z | https://github.com/MonetDB/MonetDB/issues/2126 | 753,350,336 | 2,126 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-07 02:07:25 +0200
From: John van Schie <<johnvanschie>>
To: Peter Boncz <<boncz>>
Version: -- development
Last updated: 2009-08-06 01:24:33 +0200
## Comment 12041
Date: 2009-07-07 14:07:25 +0200
From: John van Schie <<johnvanschie>>
OS: Ubuntu LTS 8.04
Monet: MonetDB 4.28.4, build from super ball, --enable-xquery --enable-optimise, readline.h found by configure.
When I execute the query in the 'query' attachment, Mserver segfaults and mclient returns the following:
MAPI = monetdb@localhost:20925
ACTION= read_line
QUERY = element { "result" } {
ERROR = Connection terminated
Running Mserver in GDB gives the following stacktrace when it segfaults:
0 0x00007fb30444f084 in reduce () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
1 0x00007fb30444f2da in reduce () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
2 0x00007fb30444f2da in reduce () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
3 0x00007fb30444f2da in reduce () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
<snip a lot of lines equal to the one above>
131 0x00007fb3044505c3 in reduce () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
132 0x00007fb30445bf1c in reduce () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
133 0x00007fb30444f2da in reduce () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
134 0x00007fb30444f2b8 in reduce () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
135 0x00007fb30444f2da in reduce () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
136 0x00007fb30444f2da in reduce () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
137 0x00007fb30444f2da in reduce () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
138 0x00007fb304450760 in reduce () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
139 0x00007fb30445bf1c in reduce () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
140 0x00007fb30444f2da in reduce () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
141 0x00007fb30444f2da in reduce () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
142 0x00007fb30445bbd2 in reduce () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
143 0x00007fb304488768 in PFmilgen () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
144 0x00007fb3042c6a78 in PFcompile_MonetDB () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
145 0x00007fb3042c30a0 in xquery_compile_exec () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
146 0x00007fb3042c44eb in xquery_prepare () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
147 0x00007fb3042c6427 in xquery_client_engine () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_pathfinder.so
148 0x00007fb304050f84 in mapi_client_engine () from /opt/MonetDB-Feb2008-SP2-fast-insert-patch/lib/MonetDB4/lib/lib_mapi.so.4
149 0x00007fb30e8703f7 in start_thread () from /lib/libpthread.so.0
150 0x00007fb30db90b3d in clone () from /lib/libc.so.6
151 0x0000000000000000 in ?? ()
Compiling the query with pf -AT and executing the resulting mil code, yields the requested element.
## Comment 12042
Date: 2009-07-08 14:11:09 +0200
From: @sjoerdmullender
This looks very much like a stack overflow problem.
I can't reproduce the crash with the May2009-SP1 release. Since you seem to be using the Feb2009-SP2 release, can you try with a more recent version? We have done some work in recognizing the condition.
## Comment 12043
Date: 2009-07-08 16:27:40 +0200
From: John van Schie <<johnvanschie>>
Sjoerd,
Indeed, the segfault is gone in May2009-SP1.
I didn't try the May version as I couldn't find a reference to such a fix in the changelog at http://monetdb.cwi.nl/projects/monetdb/Development/Releases/May2009/index.html. Is there a list of the fixed bugs in May2009(-SP1) that I could use to prevent bugreports as this one?
## Comment 12044
Date: 2009-08-06 13:24:33 +0200
From: @sjoerdmullender
There is unfortunately no list of fixes in bug fix releases.
This depends on the fixer supplying the information, and they (1) don't always know what the repercussions are of a fix, and (2) even if they do, don't always write it in the log message. Also, when I prepare a bug fix release and look at the differences between the upcoming and previous releases, I can't always tell what the changes are for.
I'll close the bug with an Out of Date.
## Comment 12045
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2817967 at http://sourceforge.net/support/tracker.php?aid=2817967
| XQuery results in segfault | https://api.github.com/repos/MonetDB/MonetDB/issues/2125/comments | 0 | 2020-11-30T10:22:57Z | 2024-06-27T11:40:34Z | https://github.com/MonetDB/MonetDB/issues/2125 | 753,350,299 | 2,125 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-07 01:42:04 +0200
From: John van Schie <<johnvanschie>>
To: @sjoerdmullender
Version: -- development
Last updated: 2009-10-02 02:30:29 +0200
## Comment 12029
Date: 2009-07-07 13:42:04 +0200
From: John van Schie <<johnvanschie>>
OS: Ubuntu LTs 8.04, x86_64
MonetDB: MonetDB 4.28.4 (Feb2009-SP2)
I have a database with one updateable document, bookmarks.xml. After this document has been updated for a couple of times (max 10 times), sometimes queries do not return the expected answer, but return an empty result.
The document was created with an update percentage of 10.
I could not reproduce the bug by inserting the document in a fresh database and performing updates and queries, but I have a copy of a database that shows the behaviour. I'll send it to Stefan.
To reproduce the bug, run MonetDB with the given database folder (it contains the data and the xquery logs). Then continously run the three attached queries in paralel with mclient. Most queries return the correct result, but some return a empty result.
## Comment 12030
Date: 2009-07-08 14:55:51 +0200
From: @sjoerdmullender
I got the database from Stefan and quickly analyzed it. It is clear, the database is corrupt. The big question is, how did it get that way? Can you reproduce the sequence of updates that got you into this mess? If so, I'd like to see the original documents and the queries to get to this state.
## Comment 12031
Date: 2009-07-08 17:06:31 +0200
From: John van Schie <<johnvanschie>>
Sjoerd,
Do you mean the sequence of updates on the bookmarks document or the total series of updates on all updateable documents in the database, as the latter will take me some more time.
-- John
## Comment 12032
Date: 2009-07-08 18:01:39 +0200
From: @sjoerdmullender
Just for the bookmarks document. For now I'm assuming that the code treats the different document collections that you may have independently.
## Comment 12033
Date: 2009-07-09 13:25:35 +0200
From: John van Schie <<johnvanschie>>
Sjoerd,
The query log is attached to the bug report.
The first update is the creation of a folder element.
The other 6 updates, insert the content of another shredded document to the bookmarks file. This other document is a temporary document, and added before the update and deleted after the update. They are _not_ shredded under the same name.
We chose for this indirect updates as problems where encountered when inserting a large node. This solution was implemented Aug 2008, so I don't know how relevant this workaround still is.
## Comment 12034
Date: 2009-07-09 16:01:56 +0200
From: @sjoerdmullender
I have a feeling this is not the complete log of what happened to this bookmarks.xml document.
When I look at the actual database, I see that there are several holes. Since these are not at page boundaries (the document easily fits in a single page), they can only be the result of deleted nodes (or nodes replaced by smaller ones). In your log there are only insert queries.
I can run the queries you attached to create a document (I used the bookmarks.xml from your http://sourceforge.net/support/tracker.php?aid=2817414 as initial document), but the result is correct. I have tried this with the May2009-SP1, Feb2009-SP2 and Feb2009 releases.
## Comment 12035
Date: 2009-07-10 09:51:32 +0200
From: John van Schie <<johnvanschie>>
You're right, it is not a query log. I tried to recreate the updates by looking at the differences of the original document and the document in the corrupt database, but forgot about possible deletes.
It could be very well that deletes were performed, as a delete is executed when a bookmark is 'overwritten', but I do not have any way to recreate it for you. Sorry.
## Comment 12036
Date: 2009-08-06 13:31:17 +0200
From: @sjoerdmullender
Without a query log it really is hunting in the dark / searching for a needle in a haystack.
Do you know the type of deletes/replaces that might have happened?
## Comment 12037
Date: 2009-08-12 12:39:25 +0200
From: John van Schie <<johnvanschie>>
The deletes that could be performed on the document are the removal of <bookmark-definition> elements, before an <bookmark-definition> of the same name is inserted.
For example, when adding <bookmark-definition name="xyz" />, the already existing definition is removed by deleting doc("bookmarks.xml")/bookmarks/folder[@name="NFI-raoul"]/bookmark-definition[@name='xyz']
I expect that this information does not help much in tracing down the problem. Furthermore, we now store bookmarks in a different way, so I propose that we close this bug and we re-open it when we have an incident that is reproducible.
## Comment 12038
Date: 2009-08-14 14:08:42 +0200
From: @drstmane
tagged subject
## Comment 12039
Date: 2009-10-02 14:30:28 +0200
From: @sjoerdmullender
Unfortunately there is no way to reproduce this problem. Closing until we find a way in which we can reproduce.
## Comment 12040
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2817953 at http://sourceforge.net/support/tracker.php?aid=2817953
| XQ: Empty result returned for updated document | https://api.github.com/repos/MonetDB/MonetDB/issues/2124/comments | 0 | 2020-11-30T10:22:54Z | 2024-06-28T13:23:43Z | https://github.com/MonetDB/MonetDB/issues/2124 | 753,350,257 | 2,124 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-06 02:29:41 +0200
From: John van Schie <<johnvanschie>>
To: Peter Boncz <<boncz>>
Version: -- development
Last updated: 2009-11-11 03:24:37 +0100
## Comment 12021
Date: 2009-07-06 14:29:41 +0200
From: John van Schie <<johnvanschie>>
OS: Both Fedora Core 10 and Ubuntu 8.04
MonetDB: 4.28.4, compiled from superball (--enable-xquery --enable-optimise), 64 bit OIDs
Problem:
I have one small document (bookmarks.xml, attached) in MonetDB.
## Comment 12022
Date: 2009-07-06 14:34:32 +0200
From: John van Schie <<johnvanschie>>
<hit the wrong button, posted too early>
OS: Both Fedora Core 10 and Ubuntu 8.04
MonetDB: 4.28.4, compiled from superball (--enable-xquery
--enable-optimise), 64 bit OIDs
Problem:
I have one small document (bookmarks.xml, attached, 100KB) in MonetDB.
I query it with 3 mclients simultaneously, with the following commands:
while [ 0 -lt 1 ]; do mclient -lxq -p49255 < /tmp/bookmark-query1.xq ;done
while [ 0 -lt 1 ]; do mclient -lxq -p49255 < /tmp/bookmark-query2.xq ;done
while [ 0 -lt 1 ]; do mclient -lxq -p49255 < /tmp/bookmark-query3.xq ;done
This results in a Mserver process taking more and more amounts of memory, as if it is leaking memory. I observed it taking up to 5.8 GB of memory, and then failing with the following error message:
MonetDB>xquery_server: client 6 could not flush prompt
!ERROR: MAPIlisten:mapi_client failed
!OS: Cannot allocate memory
!ERROR: MAPIlisten:mapi_client failed
!OS: Cannot allocate memory
!ERROR: MAPIlisten:mapi_client failed
!OS: Cannot allocate memory
## Comment 12023
Date: 2009-07-20 20:42:24 +0200
From: @mlkersten
Memory allocation by itself is not a problem
## Comment 12024
Date: 2009-08-06 12:55:23 +0200
From: @sjoerdmullender
While running those 3 queries in parallel loops, the memory consumption of Mserver grows slowly but seemingly steadily. When I stop those loops (actually, I run each query 1000 times) then memory consumption of Mserver drops again, but not to the same level as it was before the loops started.
However, at the moment I think this (the not coming down completely) can be explained by memory fragmentation.
Why the memory consumption keeps growing during execution, I don't know. Probably because "stuff" cannot be cleaned up after a query because there are other queries still using the document/collection.
## Comment 12025
Date: 2009-08-12 13:33:20 +0200
From: John van Schie <<johnvanschie>>
If it is expected behaviour, then this bug can be closed. Still as a user, I find it hard to understand why querying a 100k document should result in more than 5GB allocated memory, and a failure to allocate even more memory.
## Comment 12026
Date: 2009-08-14 13:59:31 +0200
From: @drstmane
tagged subject
## Comment 12027
Date: 2009-11-11 15:24:34 +0100
From: @peterboncz
the switch away from using the malloc library towards vm allocs for larger areas has resolved this ragmentation problem (it was also reported by loredana..)
I ran this test for 15 minutes, and Mserver stays at 475MB
## Comment 12028
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2817414 at http://sourceforge.net/support/tracker.php?aid=2817414
| XQ: MonetDB allocates >5GB mem while querying small document | https://api.github.com/repos/MonetDB/MonetDB/issues/2123/comments | 0 | 2020-11-30T10:22:51Z | 2024-06-27T11:40:32Z | https://github.com/MonetDB/MonetDB/issues/2123 | 753,350,216 | 2,123 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-02 04:30:33 +0200
From: Hans <<hans_0_>>
To: Jan Rittinger <<jan.rittinger>>
Version: -- development
Last updated: 2009-07-03 02:22:05 +0200
## Comment 12016
Date: 2009-07-02 16:30:33 +0200
From: Hans <<hans_0_>>
OS: Fedora core 10
Installed rpms version x.28.4
While executing the innocent looking xquery listed below with mclient the result is endless.
The message: "!ERROR: xquery_print_result_main: operation failed." is shown after a while.
When removing one of the two elements within element result the result is small and as expected.
The situation happened in a real life situation and was reduced to this small xquery.
let $result :=
<document xid="5">
<folder xid="6">
<file xid="19">
<slack xid="20">
<properties>
<stream xend="507" xstart="544"/>
<streamname>data</streamname>
<name>aap</name>
</properties>
</slack>
<properties>
<md5>xxx</md5>
<sha1>yyy</sha1>
<stream xend="123" xstart="456"/>
<streamname>data</streamname>
<name>noot</name>
</properties>
</file>
</folder>
<properties>
<type>doc</type>
</properties>
</document>
let $notedobjectsresult := for $notedxid in ("20","5")
return $result[@xid=$notedxid]
let $subresult := $notedobjectsresult
let $nobj := count($subresult)
let $out := (
for $i in (1 to $nobj)
let $raw := exactly-one($subresult[$i])
return element { name($raw) } {
$raw/properties/(name|stream)
}
)
return element result {
element attributes { for $a in distinct-values(for $j in $out/@* return name($j)) return element { $a } {} },
element results { $out }
}
<>
Regards,
Hans.
## Comment 12017
Date: 2009-07-02 17:37:33 +0200
From: Jan Rittinger <<jan.rittinger>>
I cooked down the problem to the following XQuery snippet:
let $out := (
for $a in (<a x="5"/>)[@x=("5","20")]
return <b>{ $a/xxx }</b>
)
return <c>{ $out/@*, $out }</c>
The problem is that '{sum}(empty_bat, bat_with_one_row)' that calculates the size for node b returns a single row with a nil value in the tail. This invalid size is added to the document representation and the serialization interprets it incorrectly.
@MIL experts: Is there a nice solution to use a default value for {sum} or do I need to add the following additional check?
res := {sum}(..., ...);
res := [ifthenelse]([isnil](res), res.project(0), res);
## Comment 12018
Date: 2009-07-02 20:12:43 +0200
From: @drstmane
If I recall correctly from abroad, we recently changed (fixed) the implementation/semantics of grouped aggregations to return NIL (NULL) for empty groups.
Hence, the only alternative to your ifthenelse() work-around would be to (re-)introduce "non-nil" alternative of the grouped aggregates (next to the current "nil"-default ones) that would use a fixed (or parameterized) non-nil "default" value for empty groups ...
Greetings from a "MonetDB-dominated" SIGMOD 2009 in Providence, RI!
(;-))
## Comment 12019
Date: 2009-07-03 14:22:05 +0200
From: Jan Rittinger <<jan.rittinger>>
I fixed the bug in the May2009 stable version. For older Pathfinder version the attached patch may be applied.
## Comment 12020
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2815820 at http://sourceforge.net/support/tracker.php?aid=2815820
| PF: Endless loop in innocent xquery | https://api.github.com/repos/MonetDB/MonetDB/issues/2122/comments | 0 | 2020-11-30T10:22:48Z | 2024-06-27T11:40:31Z | https://github.com/MonetDB/MonetDB/issues/2122 | 753,350,186 | 2,122 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-07-02 01:17:21 +0200
From: Maurice van Keulen <<mvankeulen>>
To: @sjoerdmullender
Version: -- development
Last updated: 2009-07-06 04:55:40 +0200
## Comment 12013
Date: 2009-07-02 13:17:21 +0200
From: Maurice van Keulen <<mvankeulen>>
Just installed May2009 version; algebra backend; oid32
The query below crashes the server:
let $doc := <root><field/></root>
return <foo>{$doc//field/string()}</foo>/*
The pattern basically is to create an element without children and then to do a child-step.
## Comment 12014
Date: 2009-07-06 16:55:39 +0200
From: @sjoerdmullender
I can reproduce the problem on the May2009 release, but not on the latest version from CVS in the May2009 branch. So it seems the bug has already been fixed.
I added this pattern as test: tests/BugTracker child-step-empty-node.SF-2815716.
## Comment 12015
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2815716 at http://sourceforge.net/support/tracker.php?aid=2815716
| Creating element without children crashes server | https://api.github.com/repos/MonetDB/MonetDB/issues/2121/comments | 0 | 2020-11-30T10:22:45Z | 2024-06-28T13:23:42Z | https://github.com/MonetDB/MonetDB/issues/2121 | 753,350,136 | 2,121 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-06-30 04:22:48 +0200
From: @gijzelaerr
To: @njnes
Version: -- development
Last updated: 2009-08-06 09:34:43 +0200
## Comment 12009
Date: 2009-06-30 16:22:48 +0200
From: @gijzelaerr
See example which says all:
Xauto_commit 0
sCREATE TABLE "bliep" (col1 INT, col2 VARCHAR(3));
INSERT INTO "bliep" VALUES (0,\'blablabla\') gives error
sROLLBACK;
INSERT INTO "bliep" VALUES (0,\'blablabla\') mserver hangs
## Comment 12010
Date: 2009-08-06 15:53:32 +0200
From: @mlkersten
Bug file added to BugTracker-2009. named rollback_mistake
## Comment 12011
Date: 2009-08-06 21:34:43 +0200
From: @njnes
fixed we now correctly clear the query cache after each error.
## Comment 12012
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2814622 at http://sourceforge.net/support/tracker.php?aid=2814622
| mserver connection hangs after running query after rollback | https://api.github.com/repos/MonetDB/MonetDB/issues/2120/comments | 0 | 2020-11-30T10:22:41Z | 2024-06-27T11:40:29Z | https://github.com/MonetDB/MonetDB/issues/2120 | 753,350,089 | 2,120 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-06-30 04:05:47 +0200
From: gbalint <<g2balint>>
To: Peter Boncz <<boncz>>
Version: -- development
Last updated: 2009-11-11 03:32:04 +0100
## Comment 12004
Date: 2009-06-30 16:05:47 +0200
From: gbalint <<g2balint>>
MonetDB handles BC timestamps, however it fails on feb 29 dates in leap years.
For a non-leap year, subtracting 1 from a 03-01 timestamps works:
sql>select cast('-4-03-01' as timestamp)-1;
+----------------------------+
| sql_sub_single_value |
+============================+
| -4-02-28 23:59:59.000000 |
+----------------------------+
1 tuple
But for a leap year it returns nil:
sql>select cast('-5-03-01' as timestamp)-1;
+----------------------------+
| sql_sub_single_value |
+============================+
| nil 23:59:59.000000 |
+----------------------------+
1 tuple
You also can not instantiate it explicitly:
sql>select cast('-5-02-29' as timestamp)-1;
!SQLException:timestamp:conversion of string '-5-02-29' failed
## Comment 12005
Date: 2009-08-06 23:08:38 +0200
From: @njnes
this is a problem of the mtime (monettime) module. When converting the string representation into internal values, there is a compensation for the non existing 'year zero'. This somehow changes -4 into year -3 etc. passing this to the original module writer.
## Comment 12006
Date: 2009-08-14 14:11:06 +0200
From: @drstmane
tagged subject
## Comment 12007
Date: 2009-11-11 15:31:46 +0100
From: @peterboncz
BC years are a concept that is hardly definable, as the concept of year and uniform time certainly was unknown in that age.
My take is that BC years are not supported by MonetDB/SQL.
## Comment 12008
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2814614 at http://sourceforge.net/support/tracker.php?aid=2814614
| SQL: BC leap years handled incorrectly | https://api.github.com/repos/MonetDB/MonetDB/issues/2119/comments | 0 | 2020-11-30T10:22:37Z | 2024-06-28T13:25:19Z | https://github.com/MonetDB/MonetDB/issues/2119 | 753,350,051 | 2,119 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-06-26 10:18:19 +0200
From: Gábor <<feherg>>
To: Fabian Groffen <<fabian>>
Version: -- development
Last updated: 2009-10-01 01:03:50 +0200
## Comment 11997
Date: 2009-06-26 22:18:19 +0200
From: Gábor <<feherg>>
Operating System: openSUSE 11.1
Compiler: gcc-4.3-34.243
Parser tools: package lex is not installed; flex-2.5.35-43.23; package yacc is not installed; bison-2.3-127.26
More tools: autoconf-2.63-1.136; automake-1.10.1-4.284; libtool-2.2.6-1.35
MonetDB version: May2009 from source (with a recent head checkout of buildtools, and binary distribution of JDBC)
Loaded MonetDB modules: sql; xml; batxml; geom;
Java 1.5.0 is used.
It seems to me that if a large number of bytes is received through a JDBC connection, a "ResultSet is closed!" exception is thrown:
Exception in thread "main" java.sql.SQLException: ResultSet is closed!
at nl.cwi.monetdb.jdbc.MonetResultSet.absolute(MonetResultSet.java:181)
at nl.cwi.monetdb.jdbc.MonetResultSet.relative(MonetResultSet.java:2024)
at nl.cwi.monetdb.jdbc.MonetResultSet.next(MonetResultSet.java:1991)
at MonetJDBC.query(MonetJDBC.java:76)
at MonetJDBC.main(MonetJDBC.java:31)
I can reproduce this with my atteched java program. First it generates a table with 3000 rows, and then issues a query on it...
## Comment 11998
Date: 2009-09-23 13:49:36 +0200
From: @grobian
Hi Gábor,
I'm sorry not to have noticed this bug (and your program to reproduce it) sooner! I will study your example code soon and hopefully fix this bug.
## Comment 11999
Date: 2009-09-23 20:56:19 +0200
From: @grobian
hmmm, I had a problem to compile your example:
[javac] Compiling 2 source files to /export/scratch/mirror/fabian/monetdb/stable/java/tests/build
[javac] /export/scratch/mirror/fabian/monetdb/stable/java/tests/BugLargeBytes_SF_2812963.java:51: cannot find symbol
[javac] symbol : method format(java.lang.String,int)
[javac] location: class java.lang.String
[javac] String.format("insert into jdbc_test (x) values (%1$d)", i);
[javac] ^
[javac] 1 error
But with some modifications, I can reproduce the error.
## Comment 12000
Date: 2009-09-23 21:38:08 +0200
From: @grobian
ok, for some reason the Java garbage collector thinks the resultset you're accessing is actually collectable, and hence calls finalize() on it, with the obvious result...
I'm still in the dark why the garbage collector thinks the resultset is collectable, as the main thread clearly has a reference to it.
## Comment 12001
Date: 2009-09-23 22:18:10 +0200
From: @grobian
Ok, I "fixed" this in the stable branch, if you could test whether this works for your real application that would be nice. Please keep an eye on memory usage if you can.
## Comment 12002
Date: 2009-10-01 13:03:50 +0200
From: @grobian
fix delivered in Aug2009-SP1 release.
## Comment 12003
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2812963 at http://sourceforge.net/support/tracker.php?aid=2812963
| JDBC connection hangs if result is too long | https://api.github.com/repos/MonetDB/MonetDB/issues/2118/comments | 0 | 2020-11-30T10:22:34Z | 2024-06-27T11:40:27Z | https://github.com/MonetDB/MonetDB/issues/2118 | 753,350,017 | 2,118 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-06-26 04:08:26 +0200
From: Gábor <<feherg>>
To: @sjoerdmullender
Version: -- development
Last updated: 2009-08-13 06:05:32 +0200
## Comment 11991
Date: 2009-06-26 16:08:26 +0200
From: Gábor <<feherg>>
Operating System: openSUSE 11.1
Compiler: gcc-4.3-34.243
Parser tools: package lex is not installed; flex-2.5.35-43.23; package yacc is not installed; bison-2.3-127.26
More tools: autoconf-2.63-1.136; automake-1.10.1-4.284; libtool-2.2.6-1.35
MonetDB version: May2009 from source (with a recent head checkout of buildtools)
Loaded MonetDB modules: sql; xml; batxml; geom;
The following query causes a segmentation fault at the server side:
select xmlelement(name a, xmlagg(xmlelement(name b, x))) from test;
where test is:
CREATE TABLE "test"."test" (
"x" varchar(64)
);
Before the query, funcs.sql from the xml test suite was run on the schema.
## Comment 11992
Date: 2009-08-06 17:10:47 +0200
From: @sjoerdmullender
The query plan for this test (sql/src/test/sql_xml/Tests/xmlelement_segfault.SF-2812767.sql) show
_26 := xml.element("a",nil:xml,nil:xml,_25);
sql.exportValue(1,".","element_a","xml",0,0,13,_26,"");
Since _25 is a BAT[oid,xml], the batxml version of xml.element is called, and it produces another BAT[oid,xml]. This means that I would expect sql.exportResult should be called to produce the result, not sql.exportValue. This latter function produces the segfault since it gets a BAT where it expects a single value.
Assigning to Niels, since he probably knows instantly what's wrong.
## Comment 11993
Date: 2009-08-06 21:36:53 +0200
From: @njnes
the xml.agg(bat) -> xml function is missing it should simple aggregate all xml input from the bat into a single value. As is done by the implemented group versions. This version has a single group, ie all values.
## Comment 11994
Date: 2009-08-10 11:08:23 +0200
From: @sjoerdmullender
There are currently three versions of xml.agg:
command xml.agg(val:bat[:oid,:xml],grp:bat[:oid,:oid],e:bat[:oid,:any_1]):bat[:oid,:xml]
command xml.agg(val:bat[:oid,:xml],grp:bat[:oid,:oid]):bat[:oid,:xml]
command xml.agg(val:bat[:oid,:xml]):bat[:oid,:xml]
Do you want a fourth one, or do you want the last one to be changed to return an :xml instead of a bat[:oid,:xml]?
## Comment 11995
Date: 2009-08-13 18:05:31 +0200
From: @sjoerdmullender
The bug is now fixed in CVS by changing the implementation of xml.agg(val:bat[:oid,:xml]):bat[:oid,:xml] to xml.agg(val:bat[:oid,:xml]):xml.
The provided test is added as src/test/sql_xml xmlelement_segfault.SF-2812767.
I do assume that the expected result is just the value <a/>.
## Comment 11996
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2812767 at http://sourceforge.net/support/tracker.php?aid=2812767
| xmlagg() causes segmentation fault | https://api.github.com/repos/MonetDB/MonetDB/issues/2117/comments | 0 | 2020-11-30T10:22:31Z | 2024-06-27T11:40:26Z | https://github.com/MonetDB/MonetDB/issues/2117 | 753,349,973 | 2,117 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-06-26 02:32:40 +0200
From: @arjenderijke
To: @njnes
Version: -- development
Last updated: 2009-08-06 10:39:59 +0200
## Comment 11987
Date: 2009-06-26 14:32:40 +0200
From: @arjenderijke
If you use NULL as a value, the SQL module give a syntax error. For example, start mclient and than do:
sql>declare a int;
sql>set a = null;
The errormessage is: !syntax error, unexpected sqlNULL in: "set a = null"
This also happens if you use NULL as the value of a funtion parameter, for example:
sql> select call_function(NULL);
This was tested on a checkout of CVS current of June, 23.
## Comment 11988
Date: 2009-08-06 22:34:14 +0200
From: @njnes
added test to src/tests/BugTracker=2009/set_and_arg_null.SF-2812729.sql
## Comment 11989
Date: 2009-08-06 22:39:59 +0200
From: @njnes
we now allow for null's in assignments and as arguments of functions
## Comment 11990
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2812729 at http://sourceforge.net/support/tracker.php?aid=2812729
| using null as a value gives syntax error | https://api.github.com/repos/MonetDB/MonetDB/issues/2116/comments | 0 | 2020-11-30T10:22:27Z | 2024-06-27T11:40:25Z | https://github.com/MonetDB/MonetDB/issues/2116 | 753,349,929 | 2,116 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-06-26 02:24:53 +0200
From: @arjenderijke
To: @njnes
Version: -- development
Last updated: 2009-08-06 11:17:12 +0200
## Comment 11984
Date: 2009-06-26 14:24:53 +0200
From: @arjenderijke
The character ^ is not recognized as an operator anymore in SQL. To test, open a mclient and give the following command:
sql>select 2 ^ 2;
The errormessage is: !unexpected symbol (^)
The code is a checkout of the CVS current branch of June, 23.
## Comment 11985
Date: 2009-08-06 23:17:12 +0200
From: @njnes
added test to src/tests/BugTracker-2009/pow_operator.SF-2812721.sql
added special character to parser.
## Comment 11986
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2812721 at http://sourceforge.net/support/tracker.php?aid=2812721
| operator ^ gives error | https://api.github.com/repos/MonetDB/MonetDB/issues/2115/comments | 0 | 2020-11-30T10:22:24Z | 2024-06-27T11:40:24Z | https://github.com/MonetDB/MonetDB/issues/2115 | 753,349,891 | 2,115 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-06-26 03:44:43 +0200
From: @skinkie
To: @njnes
Version: -- development
Last updated: 2009-08-04 12:22:40 +0200
## Comment 11980
Date: 2009-06-26 03:44:43 +0200
From: @skinkie
CREATE TABLE "sys"."kvk" (
"kvk" bigint,
"bedrijfsnaam" varchar(255),
"adres" varchar(64),
"postcode" varchar(6),
"plaats" varchar(32),
"type" varchar(16)
);
CREATE TABLE "sys"."anbi" (
"naam" varchar(256),
"vestigingsplaats" varchar(32),
"beschikking" date,
"intrekking" date
);
select * from kvk,anbi where lower(naam) = lower(bedrijfsnaam) and (vestigingsplaats = 'DEN HAAG' or vestigingsplaats LIKE '%GRAVE%');
!MALException:opt.remap:Not yet implemented
(first run gives more output)
## Comment 11981
Date: 2009-07-20 21:29:45 +0200
From: @mlkersten
test file added in BUgTracker-2009/Tests/orIssue.SF-2812504.sql
## Comment 11982
Date: 2009-08-04 00:22:40 +0200
From: @njnes
fixed by implementing missing versions of like.
## Comment 11983
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2812504 at http://sourceforge.net/support/tracker.php?aid=2812504
| MALException:opt.remap:Not yet implemented | https://api.github.com/repos/MonetDB/MonetDB/issues/2114/comments | 0 | 2020-11-30T10:22:21Z | 2024-06-27T11:40:23Z | https://github.com/MonetDB/MonetDB/issues/2114 | 753,349,846 | 2,114 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-06-22 05:48:03 +0200
From: francois guerin <<fguerin>>
To: Henning Rode <<hrode>>
Version: -- development
Duplicates: #2112
Last updated: 2010-05-04 15:09:54 +0200
## Comment 11977
Date: 2009-06-22 17:48:03 +0200
From: francois guerin <<fguerin>>
Hi,
the mapi client keeps on telling me two contradictory things:
- when i try to create an index (it seems that indexes are automaticaly created with module(pathfinder), Mapi tells me that an index has already been created.
- when i try to make a full text query, Mapi tells that there's no index created.
$ ./bin/mclient -lx --time
xquery>tijah:create-ft-index()
more><>
MAPI = monetdb@localhost:50000
QUERY = tijah:create-ft-index()
ERROR = !ERROR: tj_init_collection, pftijah collection already exists: DFLT_FT_INDEX
Timer 11.846 msec
xquery>let $c := collection("frwikipedia")
for $res in tijah:query($c, "//T[about(.,marcher)]")
return $res/Tranche/Sentence/T
more><>
MAPI = monetdb@localhost:50000
QUERY =
ERROR = !ERROR: tj_get_ft_index, ft-index "DFLT_FT_INDEX" does not exist. create full text index using tijah:create-ft-index()
Timer 22.598 msec
xquery>
(MacBook - Monetdb MAY release - 32bits - installation: --enable-xquery --enable-optimise)
## Comment 11978
Date: 2009-06-23 10:05:49 +0200
From: @drstmane
Duplicate of ID: 2810280 "full text index exists or not exists."
https://sourceforge.net/tracker/?func=detail&aid=2810280&group_id=56967&atid=482468
## Comment 11979
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2810355 at http://sourceforge.net/support/tracker.php?aid=2810355
Bug has invalid status, setting status to "NEW".
Previous status was "DELETED".
## Comment 13990
Date: 2010-05-04 15:09:54 +0200
From: @grobian
*** This bug has been marked as a duplicate of bug #2112 ***
| full text index exists or not exists. | https://api.github.com/repos/MonetDB/MonetDB/issues/2113/comments | 0 | 2020-11-30T10:22:17Z | 2024-06-28T13:25:18Z | https://github.com/MonetDB/MonetDB/issues/2113 | 753,349,807 | 2,113 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-06-22 03:45:53 +0200
From: francois guerin <<fguerin>>
To: Henning Rode <<hrode>>
Version: -- development
Last updated: 2010-05-04 15:09:54 +0200
## Comment 11964
Date: 2009-06-22 15:45:53 +0200
From: francois guerin <<fguerin>>
Hi,
the mapi client keeps on telling me two contradictory things:
- when i try to create an index (it seems that indexes are automaticaly created with module(pathfinder), Mapi tells me that an index has already been created.
- when i try to make a full text query, Mapi tells that there's no index created.
$ ./bin/mclient -lx --time
xquery>tijah:create-ft-index()
more><>
MAPI = monetdb@localhost:50000
QUERY = tijah:create-ft-index()
ERROR = !ERROR: tj_init_collection, pftijah collection already exists: DFLT_FT_INDEX
Timer 11.846 msec
xquery>let $c := collection("frwikipedia")
for $res in tijah:query($c, "//T[about(.,marcher)]")
return $res/Tranche/Sentence/T
more><>
MAPI = monetdb@localhost:50000
QUERY =
ERROR = !ERROR: tj_get_ft_index, ft-index "DFLT_FT_INDEX" does not exist. create full text index using tijah:create-ft-index()
Timer 22.598 msec
xquery>
(MacBook - Monetdb MAY release - 32bits - installation: --enable-xquery --enable-optimise)
## Comment 11965
Date: 2009-06-23 15:40:08 +0200
From: Henning Rode <<hrode>>
hej francois,
i have seen similar problems before in old versions of pf/tijah, when the index got inconsistent due to commit statements at wrong places in the internal indexing procedure. most people simply emptied the database completely and re-indexed everything, which usually works fine then, but it is obviously not a solution.
now, the only problem is, that i need some way to reconstruct your error here. can you describe in a few steps (xqueries) what you did. i would need some xquery that is reproduceable here and leads to the same error message.
best -henning
## Comment 11966
Date: 2009-06-30 09:33:56 +0200
From: Henning Rode <<hrode>>
i changed a few checks in the code that are applied at indexing time.
if that does not help here, i would really need a script to reproduce the bug.
so, without further input, i will close the bug in a few weeks.
## Comment 11967
Date: 2009-08-05 15:21:50 +0200
From: @sjoerdmullender
What is the status?
## Comment 11968
Date: 2009-08-05 15:31:19 +0200
From: Henning Rode <<hrode>>
I never got any answer from the bug submitter. I left it open at that time, because I had seen similar problems before and was not sure whether they are solved in the completely current version. However, I cannot do anything here without being able to re-produce the problem.
## Comment 11969
Date: 2009-08-11 13:34:07 +0200
From: @drstmane
Francois,
could you please provide us with detailed instructions how to re-produce your problem?
Thanks!
Stefan
## Comment 11970
Date: 2009-08-11 15:15:39 +0200
From: francois guerin <<fguerin>>
how to reproduce the problem:
- dl the archive http://fguerin.inria.perso.neuf.fr/final_wiki.tar.bz2
- unzip it
- store xml files under a collection called "frwikipedia" (read-only)
- then, do what I have described in the first post.
and voila, here comes the error.
## Comment 11971
Date: 2009-08-11 15:19:54 +0200
From: @drstmane
Thanks, Francois!
... the ball is back with Henning ;-) ...
## Comment 11972
Date: 2009-08-14 14:01:09 +0200
From: @drstmane
tagged subject
## Comment 11973
Date: 2009-08-20 16:24:50 +0200
From: Henning Rode <<hrode>>
the above described indexing procedure works completely fine here, even on your large dataset.
the query indeed fails on my machine as well, but with a completely different and reasonable error message asking you to use either a relative path expression in your NEXI query or to use a tijah:query without a set of context nodes.
i suppose your error comes is the result of previous experiments resulting in a corrupt database. so, maybe you should remove the entire database and then restart the experiment. it is also possible that indexing this large dataset does simply not work on your laptop, since you probably need here main memory >4GB
## Comment 11974
Date: 2009-08-21 11:49:35 +0200
From: @drstmane
Is there a test for this one?
If so, please state where;
otherwise, please (try to) add one, or explain why that isn't (easily)
possible.
After that, please close, again.
Thanks!
## Comment 11975
Date: 2009-09-15 17:21:11 +0200
From: Henning Rode <<hrode>>
test case cannot be added. the main problem here is already that the error is not reproducable. the query mentioned in the bug reports works fine for me, and is already represented by other test cases.
## Comment 11976
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2810280 at http://sourceforge.net/support/tracker.php?aid=2810280
## Comment 13991
Date: 2010-05-04 15:09:54 +0200
From: @grobian
*** Bug #2113 has been marked as a duplicate of this bug. ***
| PFtijah: full text index exists or not exists. | https://api.github.com/repos/MonetDB/MonetDB/issues/2112/comments | 0 | 2020-11-30T10:22:14Z | 2024-06-28T13:25:18Z | https://github.com/MonetDB/MonetDB/issues/2112 | 753,349,771 | 2,112 |
[
"MonetDB",
"MonetDB"
] | Date: 2009-06-19 04:09:41 +0200
From: @grobian
To: @sjoerdmullender
Version: -- development
Last updated: 2009-06-24 05:33:49 +0200
## Comment 11961
Date: 2009-06-19 16:09:41 +0200
From: @grobian
from gdk_logger.mx:
if (MT_path_absolute(logdir))
snprintf(filename, BUFSIZ, "%s%c%s%c%s%c", logdir, DIR_SEP, dbname, DIR_SEP, fn, DIR_SEP);
else {
char *dbfarm = GDKgetenv("gdk_dbfarm");
snprintf(filename, BUFSIZ, "%s%c%s%c%s%c%s%c", dbfarm, DIR_SEP, dbname, DIR_SEP, logdir, DIR_SEP, fn, DIR_SEP);
}
problem here is that on the default for SQL of sql_logs, the actual logdir becomes: /dbfarm/dbname/sql_logs/sql/
Now if I set an absolute logdir to /dbfarm/dbname/sql_logs (user visible part), the actual logdir becomes: /dbfarm/dbname/sql_logs/dbname/sql/
this is unexpected, as dbname is magically inserted. I suggest removing the dbname, such that the absolute equivalent of a relative logdir yields in the same path.
## Comment 11962
Date: 2009-06-24 17:33:49 +0200
From: @sjoerdmullender
The absolute pathname option is for backward compatibility. With the previous release, if you used your path /dbfarm/dbname/sql_logs in the logdir option, the result would also be /dbfarm/dbname/sql_logs/dbname/sql/. This cannot be changed retroactively.
I'm closing this with Won't Fix.
## Comment 11963
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 2809053 at http://sourceforge.net/support/tracker.php?aid=2809053
| absolute logdir produces other path than relative | https://api.github.com/repos/MonetDB/MonetDB/issues/2111/comments | 0 | 2020-11-30T10:22:11Z | 2024-06-28T13:25:17Z | https://github.com/MonetDB/MonetDB/issues/2111 | 753,349,727 | 2,111 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.