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: 2018-06-13 12:34:02 +0200
From: @kutsurak
To: SQL devs <<bugs-sql>>
Version: -- development
Last updated: 2018-08-31 13:23:24 +0200
## Comment 26482
Date: 2018-06-13 12:34:02 +0200
From: @kutsurak
User-Agent: Mozilla/5.0 (X11; Fedora; Linux x86_64; rv:60.0) Gecko/20100101 Firefox/60.0
Build Identifier:
BLOB columns are not handled by python UDFs whereas they should be. Florestan De Moor was kind enough to provide a patch for this.
I am opening this issue to handle any communication about this patch.
Reproducible: Always
## Comment 26483
Date: 2018-06-13 12:35:56 +0200
From: @kutsurak
Created attachment 603
Patch to handle blobs in Python UDFs
> Attached file: [pyobjectsToBlob.patch](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_6612_pyobjectsToBlob.patch_603) (text/plain, 3855 bytes)
> Description: Patch to handle blobs in Python UDFs
## Comment 26555
Date: 2018-07-19 10:43:45 +0200
From: @kutsurak
Patch applied in commit: [7aaeaa80867ff14a736a2edce512aca2ac77d5d1](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=7aaeaa80867ff14a736a2edce512aca2ac77d5d1)
| Implement BLOB handling in python UDFs | https://api.github.com/repos/MonetDB/MonetDB/issues/6612/comments | 0 | 2020-11-30T16:28:56Z | 2024-06-27T13:04:28Z | https://github.com/MonetDB/MonetDB/issues/6612 | 753,621,657 | 6,612 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-06-11 12:56:48 +0200
From: Ingo Müller <<bugzilla.monetdb.org>>
To: GDK devs <<bugs-common>>
Version: 11.29.7 (Mar2018-SP1)
CC: arkr17997, daryt, fosema4365, hexaba9787, merijo9251, pirat, vypal
Last updated: 2020-11-05 14:25:35 +0100
## Comment 26481
Date: 2018-06-11 12:56:48 +0200
From: Ingo Müller <<bugzilla.monetdb.org>>
I am trying to compile the Mar2018_SP1_release branch like so:
./bootstrap && rm -rf build && mkdir build && cd build && CC=gcc-8.1 ../configure --enable-debug=no && make
With:
$ gcc-8.1 --version
gcc-8.1 (GCC) 8.1.0
Copyright (C) 2018 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
But fail with the following error:
libtool: compile: gcc-8.1 -DHAVE_CONFIG_H -I. -I../../gdk -I.. -I../../gdk -I../common/options -I../../gdk/../common/options -I../common/stream -I../../gdk/../common/stream -I../common/utils -I../../gdk/../common/utils -DLIBGDK -g -O2 -Werror -Wall -Wextra -W -Werror-implicit-function-declaration -Wpointer-arith -Wundef -Wformat=2 -Wformat-overflow=1 -Wno-format-truncation -Wno-format-nonliteral -Wno-cast-function-type -Winit-self -Winvalid-pch -Wmissing-declarations -Wmissing-format-attribute -Wmissing-prototypes -Wold-style-definition -Wpacked -Wunknown-pragmas -Wvariadic-macros -fstack-protector-all -Wstack-protector -Wpacked-bitfield-compat -Wsync-nand -Wjump-misses-init -Wmissing-include-dirs -Wlogical-op -Wduplicated-cond -Wduplicated-branches -Wrestrict -Wnested-externs -Wno-char-subscripts -Wunreachable-code -D_REENTRANT -c ../../gdk/gdk_calc.c -fPIC -DPIC -o .libs/libbat_la-gdk_calc.o
In file included from ../../gdk/gdk.h:2271,
from ../../gdk/gdk_calc.c:10:
../../gdk/gdk_calc.c: In function 'addstr_loop':
../../gdk/gdk_atoms.h:266:4: error: 'memcpy' forming offset [3, 65535] is out of the bounds [0, 2] of object 'str_nil' with type 'const char[2]' [-Werror=array-bounds]
memcpy(d_, s_, ATOMsize(t_)); \
^~~~~~~~~~~~~~~~~~~~~~~~~~~~
../../gdk/gdk.h:1050:4: note: in expansion of macro 'ATOMputFIX'
ATOMputFIX((b)->ttype, (p), v); \
^~~~~~~~~~
../../gdk/gdk.h:1173:3: note: in expansion of macro 'Tputvalue'
Tputvalue((b), Tloc((b), (p)), (v), 0); \
^~~~~~~~~
../../gdk/gdk_calc.c:3415:3: note: in expansion of macro 'tfastins_nocheck'
tfastins_nocheck(bn, i, str_nil, Tsize(bn));
^~~~~~~~~~~~~~~~
In file included from ../../gdk/gdk.h:2271,
from ../../gdk/gdk_calc.c:10:
../../gdk/gdk_atoms.h:132:23: note: 'str_nil' declared here
gdk_export const char str_nil[2];
^~~~~~~
In file included from ../../gdk/gdk.h:2271,
from ../../gdk/gdk_calc.c:10:
(A similar error repeats a couple of times more.)
Compiling with --enable-debug=yes (or no option set) works fine. The only difference in the failing command is "-O2" (and the commands succeed/fail if I manually remove/add that option). Compiling with clang 5.0 works fine with an otherwise unchanged setup.
## Comment 26499
Date: 2018-06-26 13:07:28 +0200
From: @sjoerdmullender
Configure using --disable-strict and it will compile (although you will get warnings).
This is actually fixed in the default branch. I don't think this will be fixed in the Mar2018 branch since there are too many changes for a release branch.
## Comment 26754
Date: 2018-12-25 07:24:27 +0100
From: Thanos <<arkr17997>>
Must be thanks for the share this nice post here it is amazing way to look here http://passwordsinmicrosoftedge.com/ and access the all windows setting to see remembered passwords in microsoft edge to secure data.
| Cannot compile with GCC 8.1 and --enable-debug=no | https://api.github.com/repos/MonetDB/MonetDB/issues/6611/comments | 0 | 2020-11-30T16:28:52Z | 2024-06-27T13:04:27Z | https://github.com/MonetDB/MonetDB/issues/6611 | 753,621,600 | 6,611 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-06-10 10:32:53 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: 11.29.7 (Mar2018-SP1)
CC: @njnes
Last updated: 2018-08-31 13:23:10 +0200
## Comment 26479
Date: 2018-06-10 10:32:53 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.117 Safari/537.36
Build Identifier:
Running the query bellow, results in an undefined MAL function call. I guess it's another exp_convert is missing :)
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab1(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. CREATE TABLE tab2(col0 INTEGER, col1 INTEGER, col2 INTEGER);
3. INSERT INTO tab1 VALUES (64,10,57), (3,26,54), (80,13,96);
4. INSERT INTO tab2 VALUES (7,31,27), (79,17,38), (78,59,26);
5. SELECT DISTINCT * FROM tab2, tab1 AS cor0 CROSS JOIN tab1 WHERE cor0.col1 NOT BETWEEN tab2.col0 AND ( NULL );
### Actual Results:
ERROR = !'algebra.rangejoin' undefined in: algebra.rangejoin(X_23:bat[:int], X_38:bat[:int], X_53:int, nil:BAT, nil:BAT, true:bit, true:bit, nil:lng);
### Expected Results:
PostgreSQL gets this results, however our behavior with BETWEEN NULL is different.
col0 | col1 | col2 | col0 | col1 | col2 | col0 | col1 | col2
------+------+------+------+------+------+------+------+------
78 | 59 | 26 | 3 | 26 | 54 | 3 | 26 | 54
78 | 59 | 26 | 3 | 26 | 54 | 64 | 10 | 57
78 | 59 | 26 | 3 | 26 | 54 | 80 | 13 | 96
78 | 59 | 26 | 64 | 10 | 57 | 3 | 26 | 54
78 | 59 | 26 | 64 | 10 | 57 | 64 | 10 | 57
78 | 59 | 26 | 64 | 10 | 57 | 80 | 13 | 96
78 | 59 | 26 | 80 | 13 | 96 | 3 | 26 | 54
78 | 59 | 26 | 80 | 13 | 96 | 64 | 10 | 57
78 | 59 | 26 | 80 | 13 | 96 | 80 | 13 | 96
79 | 17 | 38 | 3 | 26 | 54 | 3 | 26 | 54
79 | 17 | 38 | 3 | 26 | 54 | 64 | 10 | 57
79 | 17 | 38 | 3 | 26 | 54 | 80 | 13 | 96
79 | 17 | 38 | 64 | 10 | 57 | 3 | 26 | 54
79 | 17 | 38 | 64 | 10 | 57 | 64 | 10 | 57
79 | 17 | 38 | 64 | 10 | 57 | 80 | 13 | 96
79 | 17 | 38 | 80 | 13 | 96 | 3 | 26 | 54
79 | 17 | 38 | 80 | 13 | 96 | 64 | 10 | 57
79 | 17 | 38 | 80 | 13 | 96 | 80 | 13 | 96
The following two queries give the same error as well:
1. CREATE TABLE tab0(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab0 VALUES (89,91,82), (35,97,1), (24,86,33);
3. SELECT ALL * FROM tab1, tab2, tab2 AS cor0 WHERE + tab2.col2 BETWEEN ( tab1.col0 ) AND NULL;
4. SELECT * FROM tab0, tab2 AS cor0 WHERE ( - tab0.col0 ) BETWEEN ( cor0.col2 ) AND ( NULL );
The sqlite output is empty for these two queries, but again our behavior with BETWEEN NULL is different.
## Comment 26480
Date: 2018-06-10 10:33:30 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [d79ebbbaf60f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d79ebbbaf60f) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=d79ebbbaf60f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=d79ebbbaf60f)
Changeset description:
Added test for Bug #6610
## Comment 26512
Date: 2018-06-27 14:27:12 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [b85418f83f4e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=b85418f83f4e) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=b85418f83f4e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=b85418f83f4e)
Changeset description:
fix bug #6610
## Comment 26513
Date: 2018-06-27 14:27:42 +0200
From: @njnes
added missing constant 2 full bat conversion
| Sqlitelogictest algebra.rangejoin undefined | https://api.github.com/repos/MonetDB/MonetDB/issues/6610/comments | 0 | 2020-11-30T16:28:48Z | 2024-06-27T13:04:26Z | https://github.com/MonetDB/MonetDB/issues/6610 | 753,621,546 | 6,610 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-06-05 15:53:33 +0200
From: Harman DBA <<HCS-DL-INBE-IRI-DSGDB>>
To: SQL devs <<bugs-sql>>
Version: 11.29.7 (Mar2018-SP1)
CC: kravchenko.anton86, martin.van.dinther, @PedroTadim
Last updated: 2019-04-30 12:36:03 +0200
## Comment 26476
Date: 2018-06-05 15:53:33 +0200
From: Harman DBA <<HCS-DL-INBE-IRI-DSGDB>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.181 Safari/537.36
Build Identifier:
We need to rename schema as we have challenges exporting database objects and importing it to a new database with a different schema name if the size of the database is huge(500GB +).
Reproducible: Always
### Steps to Reproduce:
1.need to make a copy of the database with a different schema name
2.exporting database is taking longer time, to avoid, We plan to make a cold backup and rename schema
3.this can resolve if the schema can be renamed
### Actual Results:
NA
### Expected Results:
rename schema on a database
Ability to rename objects - need option to rename tables,rename schemas.
## Comment 26485
Date: 2018-06-14 15:24:44 +0200
From: Martin van Dinther <<martin.van.dinther>>
Changed it to an enhancement request.
Suggested syntax: ALTER SCHEMA name RENAME TO new_name
See also: https://www.postgresql.org/docs/current/static/sql-alterschema.html
Renaming a table name is already recorded in:
https://www.monetdb.org/bugzilla/show_bug.cgi?id=3567
Renaming a column name is already recorded in:
https://www.monetdb.org/bugzilla/show_bug.cgi?id=2478
## Comment 26671
Date: 2018-11-07 11:48:07 +0100
From: @PedroTadim
A new SQL statement will be available in the next feature release of MonetDB for renaming schemas:
ALTER SCHEMA [IF EXISTS] oldname RENAME TO newname;
## Comment 26844
Date: 2019-01-23 13:25:29 +0100
From: @sjoerdmullender
Will be in next feature release.
## Comment 26959
Date: 2019-04-16 22:04:19 +0200
From: @sjoerdmullender
*** Bug #6698 has been marked as a duplicate of this bug. ***
| Rename Schema in MonetDB | https://api.github.com/repos/MonetDB/MonetDB/issues/6609/comments | 0 | 2020-11-30T16:28:45Z | 2024-06-27T13:04:25Z | https://github.com/MonetDB/MonetDB/issues/6609 | 753,621,509 | 6,609 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-05-24 09:43:20 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: 11.29.7 (Mar2018-SP1)
CC: @njnes
Last updated: 2019-04-30 12:36:04 +0200
## Comment 26466
Date: 2018-05-24 09:43:20 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.117 Safari/537.36
Build Identifier:
When using a COALESCE function call in a non-aggregated column in a aggregation query, MonetDB generates an error message difficult to understand. It should generate a more user friendly message like: cannot use non GROUP BY column '...' in query results without an aggregate function.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab0(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. SELECT DISTINCT - COALESCE ( - 86, + cor0.col1, cor0.col1, - cor0.col0 ) AS col2 FROM tab0 AS cor0 GROUP BY cor0.col2, cor0.col0;
### Actual Results:
The error message:
SELECT: no such aggregate 'sql_neg'
### Expected Results:
The error message:
cannot use non GROUP BY column 'cor0.col1' in query results without an aggregate function.
## Comment 26467
Date: 2018-05-24 09:49:07 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [232e9b446772](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=232e9b446772) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=232e9b446772](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=232e9b446772)
Changeset description:
Added test for Bug #6608
## Comment 26782
Date: 2019-01-09 11:49:51 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [fa366cd68200](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=fa366cd68200) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=fa366cd68200](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=fa366cd68200)
Changeset description:
We will approve bug #6608 error messages (same issue as in bug #3340).
| Sqlitelogictest error message in group by with coalesce | https://api.github.com/repos/MonetDB/MonetDB/issues/6608/comments | 0 | 2020-11-30T16:28:41Z | 2024-06-27T13:04:24Z | https://github.com/MonetDB/MonetDB/issues/6608 | 753,621,468 | 6,608 |
[
"MonetDB",
"MonetDB"
] | # Deleted Bugzilla Bug
Date: 2020-11-30T16:22:43Z
This bug was created during the import from Bugzilla.
GitHub Issues Tracker allows to assign consecutive
ids only. So this dummy issue was created in place of
a deleted Bugzilla Bug.
Please ignore this report.
| Deleted Bugzilla Bug | https://api.github.com/repos/MonetDB/MonetDB/issues/6607/comments | 0 | 2020-11-30T16:28:38Z | 2020-11-30T16:28:40Z | https://github.com/MonetDB/MonetDB/issues/6607 | 753,621,424 | 6,607 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-05-22 10:26:37 +0200
From: @swingbit
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.27.13 (Jul2017-SP4)
CC: martin.van.dinther, @njnes
Last updated: 2018-08-31 13:23:09 +0200
## Comment 26461
Date: 2018-05-22 10:26:37 +0200
From: @swingbit
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
The general form of the generate_series function (in all the integer-size variants) is:
create function sys.generate_series(first int, last int, stepsize int)
Parameters "first" and "last" obviously suggest that the series is from "first" to "last" included.
But it actually is from "first" to ("last" - 1)
sql>select * from sys.generate_series(1,5,1);
+-------+
| value |
+=======+
| 1 |
| 2 |
| 3 |
| 4 |
+-------+
4 tuples (1.241ms)
It looks to me like actually generating all numbers from "first" to "last" was the initially intended behaviour - I can't think why one would want to complicate function parameters this way intentionally. A "=" probably slipped off the condition in the loop during implementation. However changing that now would break backward compatibility.
Reproducible: Always
## Comment 26475
Date: 2018-06-01 19:30:53 +0200
From: Martin van Dinther <<martin.van.dinther>>
I agree it should behave similar to Postgresql Series Generating Functions:
https://www.postgresql.org/docs/9.6/static/functions-srf.html
## Comment 26594
Date: 2018-08-08 15:17:38 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [e63096367379](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e63096367379) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=e63096367379](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=e63096367379)
Changeset description:
Changed parameter name from "last" to "limit".
This should alleviate bug #6606.
## Comment 26606
Date: 2018-08-28 09:42:37 +0200
From: @sjoerdmullender
The confusion was (hopefully) taken away by changing the name of the parameter.
Many such generators go from start up to, but not including finish, so it's not so strange to that here as well. Although I also see the merit of aligning with PostgreSQL.
Let's make it a different bug if we really want to change the semantics.
| Misleading parameter name in generate_series function | https://api.github.com/repos/MonetDB/MonetDB/issues/6606/comments | 0 | 2020-11-30T16:28:35Z | 2024-06-27T13:04:23Z | https://github.com/MonetDB/MonetDB/issues/6606 | 753,621,382 | 6,606 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-05-21 18:07:38 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: 11.29.3 (Mar2018)
CC: @kutsurak, @njnes
Last updated: 2019-09-10 20:18:26 +0200
## Comment 26459
Date: 2018-05-21 18:07:38 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.117 Safari/537.36
Build Identifier:
Run the queries bellow for wrong results. The NOT IN operator doesn't handle null values properly.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab0(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab0 VALUES (97,1,99), (15,81,47), (87,21,10);
3. SELECT * FROM tab0 cor0 WHERE + col2 + + + 89 + - CAST ( NULL AS REAL ) NOT IN ( 28 * + 37 - ( + col1 ) + + 63, 4, ( ( + - col0 ) ), + col2 );
### Actual Results:
+------+------+------+
| col0 | col1 | col2 |
+======+======+======+
| 97 | 1 | 99 |
| 15 | 81 | 47 |
| 87 | 21 | 10 |
+------+------+------+
### Expected Results:
An empty result set.
I added other similar queries in the test file, which all of them should return an empty result set. I'm sure that all of them are related.
## Comment 26460
Date: 2018-05-21 18:15:10 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [170e20019960](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=170e20019960) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=170e20019960](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=170e20019960)
Changeset description:
Added test for Bug #6605
## Comment 26463
Date: 2018-05-23 09:52:39 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [bee93cbacfa0](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=bee93cbacfa0) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=bee93cbacfa0](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=bee93cbacfa0)
Changeset description:
partial fixes for bug #6605, ie handle again NULLs on the left side.
## Comment 26464
Date: 2018-05-23 09:53:46 +0200
From: @njnes
nan == nan -> 0 ie the current BATselect(nil, true,true,false) fails.
## Comment 26469
Date: 2018-05-30 11:13:21 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [437ba3315182](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=437ba3315182) made by Panagiotis Koutsourakis <kutsurak@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=437ba3315182](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=437ba3315182)
Changeset description:
Bug #6605 has not been fixed in Mar2018
## Comment 26999
Date: 2019-05-06 11:22:27 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [49fdc1899528](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=49fdc1899528) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=49fdc1899528](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=49fdc1899528)
Changeset description:
Adding more sqlite tests failing queries:
Bug #6605 outputs only 3 columns out of 6 on default.
Bug #6566 crashes on both Apr2019 and default.
Bug #6579 crashes on default.
## Comment 27003
Date: 2019-05-08 13:04:32 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [ba2b6178ae8d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ba2b6178ae8d) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ba2b6178ae8d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ba2b6178ae8d)
Changeset description:
fix for problem with in without subquery (new output of bug #6605)
## Comment 27274
Date: 2019-09-10 20:18:26 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [890a6d8e9e25](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=890a6d8e9e25) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=890a6d8e9e25](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=890a6d8e9e25)
Changeset description:
fix new crash in sqlitelogictest-cast-null-not-in.Bug-6605 (ie bug #6605)
| Sqlitelogictest set queries with wrong results | https://api.github.com/repos/MonetDB/MonetDB/issues/6605/comments | 0 | 2020-11-30T16:28:32Z | 2024-06-27T13:04:22Z | https://github.com/MonetDB/MonetDB/issues/6605 | 753,621,350 | 6,605 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-05-17 12:57:57 +0200
From: @swingbit
To: SQL devs <<bugs-sql>>
Version: 11.27.13 (Jul2017-SP4)
CC: @kutsurak, @njnes
Last updated: 2018-11-14 14:53:32 +0100
## Comment 26454
Date: 2018-05-17 12:57:57 +0200
From: @swingbit
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
This succeeds:
sql>create procedure test()
more>begin
more> create table x(i int);
more>end;
operation successful (4.570ms)
This gives no error, but also no success (indeed, it can be repeated and nothing happens):
sql>create procedure test()
more>begin
more> create table x as select * from sys.functions with data;
more>end;
sql>create procedure test()
more>begin
more> create table x as select * from sys.functions with data;
more>end;
sql>
Reproducible: Always
## Comment 26455
Date: 2018-05-18 10:48:37 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [a9ef7be95524](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=a9ef7be95524) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=a9ef7be95524](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=a9ef7be95524)
Changeset description:
Steps to fix bug #6604. When creating a table inside a sequential block (UDF) check for the rel_insert case when the table is created with data.
However even in the scenario without data, the table doesn't get created. Another bug :(
## Comment 26456
Date: 2018-05-18 11:29:11 +0200
From: @kutsurak
Hi Roberto,
This seems to be fixed in the Mar2018 version:
$ mclient -d supervisor
Welcome to mclient, the MonetDB/SQL interactive terminal (Mar2018)
Database: MonetDB v11.29.3 (Mar2018), 'mapi:monetdb://localhost:50000/supervisor'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>create procedure test()
more>begin
more> create table foo (i int);
more>end;
operation successful
sql>create procedure test()
more>begin
more> create table x as select * from sys.functions with data;
more>end;
CREATE PROCEDURE: name 'test' already in use
sql>
Apparently Pedro stumbled upon a different issue while trying to reproduce the problem you are having.
Panos.
## Comment 26458
Date: 2018-05-21 13:14:46 +0200
From: @njnes
currently the 'create' within a procedure (or function) is handled as a
declare table.
## Comment 26462
Date: 2018-05-22 11:00:19 +0200
From: @swingbit
Niels, I see the status changed to NEEDINFO.
Do you need info from me?
## Comment 26677
Date: 2018-11-14 14:53:32 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [2048aab70e74](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=2048aab70e74) made by Martin van Dinther <martin.van.dinther@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=2048aab70e74](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=2048aab70e74)
Changeset description:
Adapt test to match issue reported in bug #6604
Added stable output and error files
The original issue reported in 6604 (multiple create proceudre test() succeed) is not reproducable in Aug2018 release
| create procedure fails silently | https://api.github.com/repos/MonetDB/MonetDB/issues/6604/comments | 0 | 2020-11-30T16:28:29Z | 2024-06-27T13:04:21Z | https://github.com/MonetDB/MonetDB/issues/6604 | 753,621,319 | 6,604 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-05-12 15:20:56 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: 11.29.3 (Mar2018)
CC: @njnes
Last updated: 2018-06-20 11:02:16 +0200
## Comment 26445
Date: 2018-05-12 15:20:56 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.117 Safari/537.36
Build Identifier:
The aggregation query bellow produces two identical rows in the result set despite the use of DISTINCT qualifier. The having clause has no effect in the result set.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab0(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab0 VALUES(83,0,38),(26,0,79),(43,81,24);
3. SELECT DISTINCT + + col1, col1 FROM tab0 AS cor0 GROUP BY col0, col1 HAVING NOT NULL IS NOT NULL;
### Actual Results:
+------+------+
| col1 | col1 |
+======+======+
| 0 | 0 |
| 0 | 0 |
| 81 | 81 |
+------+------+
### Expected Results:
+------+------+
| col1 | col1 |
+======+======+
| 0 | 0 |
| 81 | 81 |
+------+------+
## Comment 26446
Date: 2018-05-12 15:24:29 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [ce9b9c9ea18b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ce9b9c9ea18b) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ce9b9c9ea18b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ce9b9c9ea18b)
Changeset description:
Added test for bug #6603
## Comment 26447
Date: 2018-05-12 17:19:58 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [e8375b6f4bce](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e8375b6f4bce) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=e8375b6f4bce](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=e8375b6f4bce)
Changeset description:
fixed bug #6603 (ony reduce groupby/distinct if all groupby cols
are in the distinct list)
| Sqlitelogictest: Aggregation query with distinct clause produces duplicated rows | https://api.github.com/repos/MonetDB/MonetDB/issues/6603/comments | 0 | 2020-11-30T16:28:25Z | 2024-06-27T13:04:20Z | https://github.com/MonetDB/MonetDB/issues/6603 | 753,621,289 | 6,603 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-05-11 13:40:09 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: 11.29.3 (Mar2018)
CC: @njnes
Last updated: 2018-06-20 11:02:15 +0200
## Comment 26441
Date: 2018-05-11 13:40:09 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.117 Safari/537.36
Build Identifier:
The query bellow produces wrong results in Mar2018 branch.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab2(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab2 VALUES(64,77,40),(75,67,58),(46,51,23);
3. SELECT * FROM tab2 WHERE + col2 NOT IN ( + - 59 + + ( 76 ), col1, + CAST ( NULL AS INTEGER ), col1, - 19, col1 );
### Actual Results:
[ 64, 77, 40 ]
[ 75, 77, 40 ]
[ 46, 77, 40 ]
### Expected Results:
An empty result set.
## Comment 26442
Date: 2018-05-11 13:40:34 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [265c2bacd276](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=265c2bacd276) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=265c2bacd276](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=265c2bacd276)
Changeset description:
Added test for bug #6602
## Comment 26451
Date: 2018-05-17 08:06:01 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [1d3a15ef3aee](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=1d3a15ef3aee) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=1d3a15ef3aee](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=1d3a15ef3aee)
Changeset description:
added stable output for bug #6602 (and correct for bugs 6530 and 6594)
## Comment 26457
Date: 2018-05-20 23:16:59 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [19f676b55173](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=19f676b55173) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=19f676b55173](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=19f676b55173)
Changeset description:
fixes for bug #6602, ie make sure we handle null's in NOT in (left and right
side are handled in the rel2bin code now)
| Sqlitelogictest wrong results in IN query | https://api.github.com/repos/MonetDB/MonetDB/issues/6602/comments | 0 | 2020-11-30T16:28:22Z | 2024-06-27T13:04:19Z | https://github.com/MonetDB/MonetDB/issues/6602 | 753,621,245 | 6,602 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-05-10 20:31:12 +0200
From: @kutsurak
To: SQL devs <<bugs-sql>>
Version: 11.29.3 (Mar2018)
CC: marcell.bernat, @njnes
Last updated: 2018-06-20 11:02:18 +0200
## Comment 26433
Date: 2018-05-10 20:31:12 +0200
From: @kutsurak
User-Agent: Mozilla/5.0 (X11; Fedora; Linux x86_64; rv:59.0) Gecko/20100101 Firefox/59.0
Build Identifier:
Consider the query
(1) SELECT * FROM tbl WHERE name IS NULL
When tbl is a remote table query (1) produces the correct results, but the next query, say
(2) SELECT * FROM tbl
returns the same result as query (1).
Reproducible: Always
### Steps to Reproduce:
1. Create a remote table mirroring a base table that contains null values
2. SELECT * FROM tbl WHERE name IS NULL
3. SELECT * FROM tbl
## Comment 26434
Date: 2018-05-10 20:39:18 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [5dc1a714f475](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=5dc1a714f475) made by Panagiotis Koutsourakis <kutsurak@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=5dc1a714f475](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=5dc1a714f475)
Changeset description:
Add test for Bug #6601
## Comment 26437
Date: 2018-05-11 13:12:49 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [99b2bffa6931](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=99b2bffa6931) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=99b2bffa6931](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=99b2bffa6931)
Changeset description:
fixed bug #6601
## Comment 26438
Date: 2018-05-11 13:13:38 +0200
From: @njnes
fixed (ie made remote queries (names) unique)
| "where is null" clause on remote table causes problem with next query | https://api.github.com/repos/MonetDB/MonetDB/issues/6601/comments | 0 | 2020-11-30T16:28:20Z | 2024-06-27T13:04:18Z | https://github.com/MonetDB/MonetDB/issues/6601 | 753,621,216 | 6,601 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-05-10 15:41:48 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: 11.29.3 (Mar2018)
CC: @njnes
Last updated: 2018-06-20 11:02:09 +0200
## Comment 26431
Date: 2018-05-10 15:41:48 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.117 Safari/537.36
Build Identifier:
Run the queries bellow for compilation errors in MonetDB. The same errors occur in default branch.
Reproducible: Always
### Steps to Reproduce:
1. SELECT ALL + 60 * - NULLIF ( - CAST ( NULL AS INTEGER ), 51 * 70 * 62 / + 52 * + 58 / 90 - CAST ( + 44 AS INTEGER ) ) AS col2, 35;
2. SELECT ALL 11 * COUNT ( * ) * + 2 * 6;
3. SELECT ALL - 9 * NULLIF ( + 30, + 64 ) * + 46 * 1;
### Actual Results:
1. division by zero
2. overflow in calculation 11*12
3. !Semantic errors TypeException:user.sql.init[24]:'calc.-' undefined in: X_31503:int := calc.-(X_31502:sht).
### Expected Results:
1. NULL, 35
2. 132
3. -12420
## Comment 26432
Date: 2018-05-10 15:43:23 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [9428a9b2be0a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9428a9b2be0a) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=9428a9b2be0a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=9428a9b2be0a)
Changeset description:
Added test for bug #6600
## Comment 26435
Date: 2018-05-11 10:18:52 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [f3fcfcad3fbc](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=f3fcfcad3fbc) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=f3fcfcad3fbc](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=f3fcfcad3fbc)
Changeset description:
fixes for bug #6577 and 6600
## Comment 26439
Date: 2018-05-11 13:14:32 +0200
From: @njnes
fixed the 3 problem (ie compilation issue)
## Comment 26443
Date: 2018-05-11 19:36:32 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [dbc08d32a038](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=dbc08d32a038) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=dbc08d32a038](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=dbc08d32a038)
Changeset description:
fixed precedence of +- unary operator, fixes part 1 of bug #6600
## Comment 26444
Date: 2018-05-12 13:11:49 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [ffdcc0d6f2fb](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ffdcc0d6f2fb) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ffdcc0d6f2fb](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ffdcc0d6f2fb)
Changeset description:
fixed last part of bug #6600, ie correct digits (bits) in atom_mul
| Sqlitelogictest queries fail to execute | https://api.github.com/repos/MonetDB/MonetDB/issues/6600/comments | 0 | 2020-11-30T16:28:17Z | 2024-06-27T13:04:17Z | https://github.com/MonetDB/MonetDB/issues/6600 | 753,621,192 | 6,600 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-05-09 14:33:23 +0200
From: Manuel <<manuel>>
To: SQL devs <<bugs-sql>>
Version: 11.27.11 (Jul2017-SP3)
Last updated: 2018-05-09 15:38:53 +0200
## Comment 26426
Date: 2018-05-09 14:33:23 +0200
From: Manuel <<manuel>>
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.139 Safari/537.36
Build Identifier:
I found this behaviour in monetdb:
The value of a count(*) measure depends on whether or not there are other measures in the select list:
To illustrate the issue, consider the following table:
CREATE TABLE "sys"."unitTestDontDelete" (
"A" VARCHAR(255),
"B" BIGINT,
"C" DOUBLE,
"D" TIMESTAMP
);
INSERT INTO "sys"."unitTestDontDelete" VALUES (NULL, NULL, NULL, NULL, 0);
INSERT INTO "sys"."unitTestDontDelete" VALUES ('Cat1', 0, 0.5, '2013-06-10 11:10:10.000000');
INSERT INTO "sys"."unitTestDontDelete" VALUES ('Cat2', 1, 1.5, '2013-06-11 12:11:11.000000');
INSERT INTO "sys"."unitTestDontDelete" VALUES ('Cat1', 2, 2.5, '2013-06-12 13:12:12.000000');
INSERT INTO "sys"."unitTestDontDelete" VALUES ('Cat2', 3, 3.5, '2013-06-13 14:13:13.000000');
INSERT INTO "sys"."unitTestDontDelete" VALUES ('Cat1', 4, 4.5, '2013-06-14 15:14:14.000000');
INSERT INTO "sys"."unitTestDontDelete" VALUES ('Cat2', 5, 5.5, '2013-06-15 16:15:15.000000');
INSERT INTO "sys"."unitTestDontDelete" VALUES ('Cat1', 6, 6.5, '2013-06-16 17:16:16.000000');
INSERT INTO "sys"."unitTestDontDelete" VALUES ('Cat2', 7, 7.5, '2013-06-17 18:17:17.000000');
INSERT INTO "sys"."unitTestDontDelete" VALUES ('Cat1', 8, 8.5, '2013-06-18 19:18:18.000000');
The query:
select count(*) as "c3_recordcount" from "unitTestDontDelete" as "t2" where "t2"."A" = 'Nope'
returns a single record with value 0 as expected, however the query
select count(*) as "c3_recordcount", count(distinct "t2"."B") as "c7_b__unique_count_" from "unitTestDontDelete" as "t2" where "t2"."A" = 'Nope'
returns
c3_recordcount c7_b__unique_count_
-------------------------------------
<null> 0
Reproducible: Always
### Steps to Reproduce:
1. create the table in the description
2. Execute the query select count(*) as "c3_recordcount", count(distinct "t2"."B") as "c7_b__unique_count_" from "unitTestDontDelete" as "t2" where "t2"."A" = 'Nope'
### Actual Results:
c3_recordcount c7_b__unique_count_
-------------------------------------
<null> 0
### Expected Results:
c3_recordcount c7_b__unique_count_
-------------------------------------
0 0
## Comment 26427
Date: 2018-05-09 15:35:30 +0200
From: @sjoerdmullender
This was already fixed in the Jul2017-SP4 (11.27.13) release.
## Comment 26428
Date: 2018-05-09 15:38:53 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [5e821c5bed12](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=5e821c5bed12) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=5e821c5bed12](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=5e821c5bed12)
Changeset description:
Test for bug #6599.
| Broken measure on empty record set | https://api.github.com/repos/MonetDB/MonetDB/issues/6599/comments | 0 | 2020-11-30T16:28:14Z | 2024-06-27T13:04:16Z | https://github.com/MonetDB/MonetDB/issues/6599 | 753,621,145 | 6,599 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-05-08 11:53:13 +0200
From: Ingo Müller <<bugzilla.monetdb.org>>
To: Documentation maintainers <<bugs-docs>>
Version: unspecified
CC: webhostinguk
Last updated: 2019-01-22 16:44:16 +0100
## Comment 26421
Date: 2018-05-08 11:53:13 +0200
From: Ingo Müller <<bugzilla.monetdb.org>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:59.0) Gecko/20100101 Firefox/59.0
Build Identifier:
When compiling with Python 3.4., I get the following error:
In file included from ../../../../../../sql/backends/monet5/UDF/pyapi3/pyapi3.c:13:0:
../../../../../../sql/backends/monet5/UDF/pyapi3/../pyapi/pyapi.c: In function 'PYAPI3PyAPIprelude':
../../../../../../sql/backends/monet5/UDF/pyapi3/../pyapi/pyapi.c:1313:3: error: implicit declaration of function 'Py_DecodeLocale' [-Werror=implicit-function-declaration]
wchar_t* program = Py_DecodeLocale("mserver5", NULL);
^
../../../../../../sql/backends/monet5/UDF/pyapi3/../pyapi/pyapi.c:1313:3: warning: nested extern declaration of 'Py_DecodeLocale' [-Wnested-externs]
../../../../../../sql/backends/monet5/UDF/pyapi3/../pyapi/pyapi.c:1313:22: warning: initialization makes pointer from integer without a cast
wchar_t* program = Py_DecodeLocale("mserver5", NULL);
^
I am compiling under Debian Jessie and I have python3-dev version 3.4.2-2 installed. The directory /usr/include/python3.4m/, which contains Python.h, indeed does not contain the string Py_DecodeLocale. Also, some googleing suggests that this function was introduced in Python 3.5.
I guess that this is OK, but should be documented. Better yet, the configure script should test for that and disable Python integration if only Python 3.4 was found. (If I do that, compilation works fine.)
Reproducible: Always
## Comment 26422
Date: 2018-05-08 11:54:41 +0200
From: Ingo Müller <<bugzilla.monetdb.org>>
I tried to compile Mar2018_release.
## Comment 26424
Date: 2018-05-09 10:37:29 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [3103947f85fa](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=3103947f85fa) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=3103947f85fa](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=3103947f85fa)
Changeset description:
Add check for Python 3.5 version in Python 3 UDFs. This fixes bug #6598
## Comment 26425
Date: 2018-05-09 11:31:14 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [d9a1bee0ff29](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d9a1bee0ff29) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=d9a1bee0ff29](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=d9a1bee0ff29)
Changeset description:
Properly fix bug #6598.
## Comment 26448
Date: 2018-05-16 11:56:12 +0200
From: Ingo Müller <<bugzilla.monetdb.org>>
(Please let me know if I am expected to do something here. The change looks like it solves my problem, but I haven tested.)
## Comment 26449
Date: 2018-05-16 12:39:44 +0200
From: @sjoerdmullender
I did try it out, and it worked for me.
I'm closing this bug. If it doesn't work after all, please reopen.
## Comment 26841
Date: 2019-01-22 16:44:16 +0100
From: Morin Mote <<webhostinguk>>
Not working, please share specified in file.
--
Regards
http://www.webhostingreviewsx.co.uk/ipage-uk-hosting-reviews/
| Python 3.4 not supported (due to usage of Py_DecodeLocale) | https://api.github.com/repos/MonetDB/MonetDB/issues/6598/comments | 0 | 2020-11-30T16:28:11Z | 2024-06-27T13:04:15Z | https://github.com/MonetDB/MonetDB/issues/6598 | 753,621,102 | 6,598 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-05-04 23:23:47 +0200
From: @eyalroz
To: GDK devs <<bugs-common>>
Version: 11.29.7 (Mar2018-SP1)
Last updated: 2018-06-27 11:21:13 +0200
## Comment 26419
Date: 2018-05-04 23:23:47 +0200
From: @eyalroz
User-Agent: Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:59.0) Gecko/20100101 Firefox/59.0
Build Identifier:
GDKfree makes some problematic assumptions:
1.
an assumption that s > sizeof(size_t)
is assumed by the line
asize = ((size_t *) s)[-1];
if an invalid but non-null pointer is provided, there would be underflow. An assert for avoiding this could be nice.
2.
an assumption that i is not too different from asize
is assumed by the loop
for (; i < asize - MALLOC_EXTRA_SPACE; i++)
assert(((char *) s)[i] == '\xBD');
but if an invalid pointer was provided and for some reason we have a large stretch of memory filled with '\xBF', this will run for quite a long time. It would be better to first assert that asize = (i + 7) & ~7 + MALLOC_EXTRA_SPACE + DEBUG_SPACE (according to the logic in GDKfree()). If that doesn't hold, there's some sort of corruption for sure.
3.
An assumption that asize and i are both non-zero. Not a problematic assumption per se, but worth an assert in my opinion.
(In case you were wondering - I've noticed these when looking into an assert firing on the final GDKfree() when BBPexit()ing)
Reproducible: Always
## Comment 26511
Date: 2018-06-27 11:21:13 +0200
From: @sjoerdmullender
If GDKfree is called with an invalid pointer you have a bigger problem. That should *never* happen.
I don't see the need to try to catch these kinds of errors in GDKfree, plus it's hard to do. There is no pattern to the pointers that we can check, so the only thing we could do is check wheter a pointer was handed out earlier. This is nice for a debugging malloc library, but GDKmalloc/GDKfree is not such a library.
The loop with asize is inside ifndef NDEBUG and thus only used in a debug build. Debug builds don't have to be extremely efficient. And in any case, the difference between how much was asked for in GDKmalloc and how much was returned is pretty small (except potentially with invalid pointers which is of no concern (see above)).
| GDKfree makes a couple of problematic assumptions | https://api.github.com/repos/MonetDB/MonetDB/issues/6597/comments | 0 | 2020-11-30T16:28:07Z | 2024-06-28T06:42:25Z | https://github.com/MonetDB/MonetDB/issues/6597 | 753,621,043 | 6,597 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-05-02 17:48:58 +0200
From: @swingbit
To: SQL devs <<bugs-sql>>
Version: 11.27.13 (Jul2017-SP4)
Last updated: 2018-08-31 13:23:23 +0200
## Comment 26417
Date: 2018-05-02 17:48:58 +0200
From: @swingbit
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
With the table in attachment, I get the following timings on the same query:
- Before ANALYZE, default pipe: 122ms
- Before ANALYZE, sequential pipe: 40ms
- After ANALYZE, default pipe: 11s
- After ANALYZE, sequential pipe: 7m45s
This is only 1M tuples for 18MB data, so I expect the problem is about data distribution, not data size (indeed, the same query on different data works just fine).
The question is why it becomes so much slower after ANALYZE.
sql>select count(distinct id) from pim_0_obj_type;
+---------+
| L3 |
+=========+
| 1072284 |
+---------+
1 tuple (13.760ms)
sql>select count(distinct type) from pim_0_obj_type;
+------+
| L3 |
+======+
| 10 |
+------+
1 tuple (15.949ms)
sql>select count(distinct typestr) from pim_0_obj_type;
+------+
| L3 |
+======+
| 10 |
+------+
1 tuple (9.249ms)
-- Before ANALYZE
sql>set optimizer='default_pipe';
operation successful (0.608ms)
sql>select count(*) from (select id as a1,type as a2,typestr as a3,max(prob) from pim_0_obj_type group by a1,a2,a3) as x;
+---------+
| L11 |
+=========+
| 1072446 |
+---------+
1 tuple (122.836ms)
sql>set optimizer='sequential_pipe';
operation successful (0.717ms)
sql>select count(*) from (select id as a1,type as a2,typestr as a3,max(prob) from pim_0_obj_type group by a1,a2,a3) as x;
+---------+
| L11 |
+=========+
| 1072446 |
+---------+
1 tuple (40.347ms)
sql>select column,type,count,columnsize,heapsize,hashes,sorted from sys.storage('sys','pim_0_obj_type');
+---------+--------+---------+------------+----------+--------+--------+
| column | type | count | columnsize | heapsize | hashes | sorted |
+=========+========+=========+============+==========+========+========+
| id | int | 1072446 | 4289784 | 0 | 0 | null |
| type | int | 1072446 | 4289784 | 0 | 0 | null |
| typestr | clob | 1072446 | 2144892 | 8585 | 0 | null |
| prob | double | 1072446 | 8579568 | 0 | 0 | null |
+---------+--------+---------+------------+----------+--------+--------+
4 tuples (1.335ms)
sql>analyze sys.pim_0_obj_type;
-- After ANALYZE
sql>select column,type,count,columnsize,heapsize,hashes,sorted from sys.storage('sys','pim_0_obj_type');
+---------+--------+---------+------------+----------+---------+--------+
| column | type | count | columnsize | heapsize | hashes | sorted |
+=========+========+=========+============+==========+=========+========+
| id | int | 1072446 | 4289784 | 0 | 0 | true |
| type | int | 1072446 | 4289784 | 0 | 4390952 | false |
| typestr | clob | 1072446 | 2144892 | 8585 | 4522024 | false |
| prob | double | 1072446 | 8579568 | 0 | 0 | true |
+---------+--------+---------+------------+----------+---------+--------+
4 tuples (2.588ms)
sql>set optimizer='default_pipe';
operation successful (0.977ms)
sql>select count(*) from (select id as a1,type as a2,typestr as a3,max(prob) from pim_0_obj_type group by a1,a2,a3) as x;
+---------+
| L11 |
+=========+
| 1072446 |
+---------+
1 tuple (11.3s)
sql>set optimizer='sequential_pipe';
operation successful (0.760ms)
sql>select count(*) from (select id as a1,type as a2,typestr as a3,max(prob) from pim_0_obj_type group by a1,a2,a3) as x;
+---------+
| L11 |
+=========+
| 1072446 |
+---------+
1 tuple (7m 45s)
Reproducible: Always
$ mserver5 --version
MonetDB 5 server v11.27.14 (64-bit, 128-bit integers)
This is an unreleased version
Copyright (c) 1993 - July 2008 CWI
Copyright (c) August 2008 - 2018 MonetDB B.V., all rights reserved
Visit https://www.monetdb.org/ for further information
Found 15.5GiB available memory, 8 available cpu cores
Libraries:
libpcre: 8.41 2017-07-05 (compiled with 8.41)
openssl: OpenSSL 1.1.0h 27 Mar 2018 (compiled with OpenSSL 1.1.0h-fips 27 Mar 2018)
libxml2: 2.9.7 (compiled with 2.9.7)
Compiled by: roberto@photon.hq.spinque.com (x86_64-pc-linux-gnu)
Compilation: gcc -O3 -fomit-frame-pointer -pipe -Werror -Wall -Wextra -W -Werror-implicit-function-declaration -Wpointer-arith -Wdeclaration-after-statement -Wundef -Wformat=2 -Wno-format-nonliteral -Winit-self -Winvalid-pch -Wmissing-declarations -Wmissing-format-attribute -Wmissing-prototypes -Wold-style-definition -Wpacked -Wunknown-pragmas -Wvariadic-macros -fstack-protector-all -Wstack-protector -Wpacked-bitfield-compat -Wsync-nand -Wjump-misses-init -Wmissing-include-dirs -Wlogical-op -Wduplicated-cond -Wduplicated-branches -Wrestrict -Wnested-externs -Wunreachable-code -D_FORTIFY_SOURCE=2
Linking : /usr/bin/ld -m elf_x86_64
## Comment 26418
Date: 2018-05-02 17:51:02 +0200
From: @swingbit
The data file is too large to be attached, here it is:
https://drive.google.com/open?id=1YcuWJpbQ8_-ozE-1t9TDaFtajDutErGp
## Comment 26420
Date: 2018-05-07 11:13:07 +0200
From: @swingbit
UPDATE: This only happens with PERSISTENTHASH and PARENT_HASH enabled.
More details of what then happens:
The problematic grouping (8 minutes) is on the string attribute, which normally takes a few milliseconds:
| 48 | X_11=<tmp_2750>[1072446]:bat[:int] := algebra.projection(C_5=<tmp_2766>[1072446]:bat[:oid], X_8=<tmp_2746>[1072446]:bat[:int]); |
| 9688 | (X_18=<tmp_2766>[1072446]:bat[:oid], C_19=<tmp_2770>[1072284]:bat[:oid], X_20=<tmp_2721>[1072284]:bat[:lng]) := group.group(X_11=<tmp_2750>[1072446]:bat[:int]); |
| 482097975 | (X_21=<tmp_2721>[1072446]:bat[:oid], C_22=<tmp_2770>[1072446]:bat[:oid], X_23=<tmp_1656>[1072446]:bat[:lng]) := group.subgroup(X_17=<tmp_516>[1072446]:bat[:str], X_18=<tmp_2766>[1072446]:bat[:oid]); |
| 806 | (X_24=<tmp_2766>[1072446]:bat[:oid], C_25=<tmp_516>[1072446]:bat[:oid], X_26=<tmp_1656>[1072446]:bat[:lng]) := group.subgroupdone(X_14=<tmp_2767>[1072446]:bat[:int], X_21=<tmp_2721>[1072446]:bat[:oid]); |
Relevant part of --algorithms output:
BATproject(l=tmp_31641072446-sorted-key,r=tmp_27451072446[int])
BATproject(l=tmp_3164,r=tmp_2745)=tmp_31751072446
BATproject(l=tmp_31641072446-sorted-key,r=tmp_27441072446[str])
BATproject(l=tmp_3164,r=tmp_2744)=tmp_32301072446
BATproject(l=tmp_31641072446-sorted-key,r=tmp_27461072446[int]-sorted)
BATproject(l=tmp_3164,r=tmp_2746)=tmp_31731072446-sorted
BATgroup(b=tmp_31731072446[int],s=NULL0,g=NULL0,e=NULL0,h=NULL0,subsorted=0): compare consecutive values
BATordered: fixed nosorted(51) for tmp_32301072446 (8 usec)
BATordered_rev: fixed norevsorted(1) for tmp_31641072446 (1 usec)
BATcheckhash: reusing persisted hash tmp_2744
BATgroup(b=tmp_32301072446[str],s=NULL0,g=tmp_31641072446,e=NULL0,h=NULL0,subsorted=0): use existing hash table
...
(stopped manually)
-- algorithms output without persistent hashes:
BATproject(l=tmp_14501072446-sorted-key,r=tmp_4721072446[int])
BATproject(l=tmp_1450,r=tmp_472)=tmp_14611072446
BATproject(l=tmp_14501072446-sorted-key,r=tmp_4111072446[int])
BATproject(l=tmp_1450,r=tmp_411)=tmp_15141072446
BATproject(l=tmp_14501072446-sorted-key,r=tmp_4711072446[str])
BATproject(l=tmp_1450,r=tmp_471)=tmp_14571072446
BATordered: fixed norevsorted(47) for tmp_14571072446
BATordered: fixed nosorted(51) for tmp_14571072446 (20 usec)
BATordered: fixed norevsorted(47) for tmp_15141072446
BATordered: fixed nosorted(51) for tmp_15141072446 (19 usec)
BATordered: fixed norevsorted(47) for tmp_14501072446
BATordered: fixed nosorted(51) for tmp_14501072446 (8 usec)
BATgroup(b=tmp_15141072446[int],s=NULL0,g=tmp_14501072446,e=NULL0,h=NULL0,subsorted=0): create partial hash table
HASHnew: create hash(size 1072446, mask 1048576, width 4, nil 4294967295, total 8484088 bytes);
BATordered: fixed norevsorted(1) for tmp_14611072446
BATordered: fixed sorted for tmp_14611072446 (1422 usec)
BATordered: fixed norevsorted(47) for tmp_14571072446
BATordered: fixed nosorted(51) for tmp_14571072446 (4 usec)
BATgroup(b=tmp_14611072446[int],s=NULL0,g=tmp_14571072446,e=NULL0,h=NULL0,subsorted=0): subscan old groups
virtualize(bn=tmp_15141072446,seq=0)
BATproject(l=tmp_15141072446-sorted-key,r=tmp_14611072446[int]-sorted)
BATproject(l=tmp_1514,r=tmp_1461)=tmp_14571072446-sorted
## Comment 26470
Date: 2018-05-31 13:23:31 +0200
From: @swingbit
Any luck in reproducing this?
## Comment 26473
Date: 2018-06-01 11:27:12 +0200
From: @swingbit
Checking with gdb, I see that all the time is spent in GRP_use_existing_hash_table_tpe(sht) (sht because strings are dictionary-compressed). I guess this is the culprit?
Using persistent hashes and running ANALYZE is the only case when this macro is used.
Which to me is also surprising, should the hash not be persisted after it is created upon the first query call, and then found on the second call, without the need for ANALYZE?
## Comment 26500
Date: 2018-06-26 13:08:34 +0200
From: @sjoerdmullender
I can reproduce this, and I think the reason for the slowness is the large number of duplicate items in the data together with mitosis and the enabling of parent hashes. The latter is not a supported configuration for a reason, and this is it.
I'm not sure I want to spend a lot of time on this problem right now, although I have some ideas.
## Comment 26501
Date: 2018-06-26 13:10:58 +0200
From: @swingbit
One comment on that: I doubt it is about mitosis. It actually is so unreasonably slower with sequential pipe, not with default pipe.
## Comment 26503
Date: 2018-06-26 14:59:56 +0200
From: @sjoerdmullender
(In reply to Roberto Cornacchia from comment 4)
> Which to me is also surprising, should the hash not be persisted after it is
> created upon the first query call, and then found on the second call,
> without the need for ANALYZE?
When there is no hash table, BATgroup creates a temporary hash table that is not compatible with a full hash table (duplicates are not stored). Therefore this hash is not persisted.
## Comment 26504
Date: 2018-06-26 15:12:20 +0200
From: @swingbit
Thanks Sjoerd. So only by calling ANALYZE the full hash table is created (and persisted, if enabled)?
## Comment 26506
Date: 2018-06-26 15:20:21 +0200
From: @swingbit
In general: I do understand if you say it is not a priority for you.
Just to cast my vote: persistent hashes are quite essential for us. The performance difference for joins is such that doing without is simply not an option. We talk about 5-6ms vs. 200ms for a single join, when we need to deliver much more complex queries that stay below 100ms.
For now, I have left persistent hashes enabled, but commented out their usage for aggregations. Of course it would be awesome if the could work without commenting any code, but for now this will do.
I do hope persistent hashes will keep being supported, even though disabled by default.
## Comment 26507
Date: 2018-06-26 15:22:28 +0200
From: @sjoerdmullender
(In reply to Roberto Cornacchia from comment 8)
> Thanks Sjoerd. So only by calling ANALYZE the full hash table is created
> (and persisted, if enabled)?
My comment was for BATgroup (and BATunique). In other places where a hash is created it will be persisted (e.g. point select or join).
## Comment 26508
Date: 2018-06-26 15:23:46 +0200
From: @sjoerdmullender
(In reply to Roberto Cornacchia from comment 9)
> In general: I do understand if you say it is not a priority for you.
I actually already have a solution. I'm just running tests before committing it.
The solution is to not use the preexisting hash at all when doing subgrouping.
## Comment 26509
Date: 2018-06-26 15:58:38 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [a8c356ffefb4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=a8c356ffefb4) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=a8c356ffefb4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=a8c356ffefb4)
Changeset description:
When subgrouping, don't use a preexisting hash table.
This fixes bug #6596.
The rationale is that we need to go through the hash chain to find a
preexisting group for the value we're dealing with. If there are many
distinct groups, but lots of duplicates in the column being
subgrouped (is that a word?) this becomes very expensive. We're
better off using the code that builds a new temporary hash table that
is based on the value being subgrouped *and* its group id. This does
cost extra memory during subgrouping.
## Comment 26510
Date: 2018-06-26 17:02:50 +0200
From: @sjoerdmullender
I am assuming my patch adequately fixes the problem.
## Comment 26516
Date: 2018-06-27 16:34:53 +0200
From: @swingbit
It does, thanks!
## Comment 26521
Date: 2018-06-28 14:08:51 +0200
From: @swingbit
One issue still (I'm not sure whether generic or related to this bug).
With the patch applied, and after ANALYZE:
Default pipeline has not improved much (5s).
Sequential pipeline has improved (50ms).
However, they both perform very bad if called twice at short distance (less than 1s):
- Default, call 1: 5ms
- Sequential, call 2: 11s
- Sequential, call 1: 51ms
- Sequential, call 2: 4s
If I let a few seconds in between the 2 calls, they all perform good (except default pipe)
Can this still be related to this bug?
## Comment 26523
Date: 2018-06-29 09:53:53 +0200
From: @sjoerdmullender
I'm wondering whether this is due to persisting the hash. I still have to check, but perhaps what's happening is that while the hash is being persisted (which can take a while), the hash lock is taken for the BAT, which means that the next query must wait when it is checking whether a hash exists.
If this is the case, then query 3 and later should all be quick.
## Comment 26524
Date: 2018-06-29 10:10:55 +0200
From: @sjoerdmullender
(In reply to Sjoerd Mullender from comment 16)
> I'm wondering whether this is due to persisting the hash. I still have to
> check, but perhaps what's happening is that while the hash is being
> persisted (which can take a while), the hash lock is taken for the BAT,
> which means that the next query must wait when it is checking whether a hash
> exists.
>
> If this is the case, then query 3 and later should all be quick.
Ignore this comment. The analyze should persist any hashes and the subgrouping which made the query slow originally doesn't use those hashes anymore anyway.
## Comment 26525
Date: 2018-06-29 10:17:17 +0200
From: @swingbit
Actually, ignore my comment, too.
What I wrote definitely happened, but I no longer see it after dropping and recreating the table (with or without analyze).
If I see it again and can frame it better I'll open a new report.
## Comment 26526
Date: 2018-06-29 10:23:04 +0200
From: @sjoerdmullender
I haven't been able to reproduce this dramatic slow down.
Here are the timings I got on my machine (debug build) using your dataset (which I assume is a lot smaller than the one you're testing with):
sql:0.000 opt:3.717 run:276.375 clk:282.063 ms
sql:0.000 opt:3.743 run:265.037 clk:270.372 ms
sql:0.000 opt:3.790 run:265.595 clk:271.028 ms
analyze sys.pim_0_obj_type;
sql:0.000 opt:3.861 run:255.614 clk:260.748 ms
sql:0.000 opt:4.043 run:253.165 clk:258.431 ms
sql:0.000 opt:3.826 run:251.523 clk:256.568 ms
sql:0.000 opt:3.900 run:250.855 clk:256.025 ms
sql:0.000 opt:3.870 run:249.873 clk:255.019 ms
set optimizer='sequential_pipe';
sql:0.000 opt:0.805 run:66.705 clk:70.104 ms
sql:0.000 opt:0.812 run:77.599 clk:79.572 ms
sql:0.000 opt:0.797 run:92.253 clk:94.234 ms
sql:0.000 opt:0.814 run:90.597 clk:92.676 ms
sql:0.000 opt:0.786 run:93.923 clk:95.948 ms
sql:0.000 opt:0.796 run:66.223 clk:68.747 ms
sql:0.000 opt:0.782 run:66.024 clk:68.015 ms
sql:0.000 opt:1.084 run:91.666 clk:93.979 ms
sql:0.000 opt:0.771 run:91.746 clk:93.780 ms
sql:0.000 opt:0.786 run:65.728 clk:67.739 ms
In this last sequence the faster times are after waits, so slow downs do occur, but not nearly as extreme as in your case.
As you can infer, in this test I did not try sequential_pipe on unanalyzed data.
## Comment 26527
Date: 2018-06-29 10:23:47 +0200
From: @sjoerdmullender
(In reply to Roberto Cornacchia from comment 18)
> Actually, ignore my comment, too.
>
> What I wrote definitely happened, but I no longer see it after dropping and
> recreating the table (with or without analyze).
>
> If I see it again and can frame it better I'll open a new report.
Ok, this report stays resolved.
| Multicolumn aggregation very slow after ANALYZE when persistent hashes are enabled | https://api.github.com/repos/MonetDB/MonetDB/issues/6596/comments | 0 | 2020-11-30T16:28:02Z | 2024-06-27T13:04:13Z | https://github.com/MonetDB/MonetDB/issues/6596 | 753,620,972 | 6,596 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-05-01 13:47:45 +0200
From: @joerivanruth
To: SQL devs <<bugs-sql>>
Version: 11.29.3 (Mar2018)
CC: @njnes
Last updated: 2018-06-20 11:02:13 +0200
## Comment 26416
Date: 2018-05-01 13:47:45 +0200
From: @joerivanruth
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.13; rv:60.0) Gecko/20100101 Firefox/60.0
Build Identifier:
Given a table t with a single DECIMAL value in it:
CREATE TABLE t (
n DECIMAL(8,2)
);
INSERT INTO t VALUES (2.00);
I expect the query
SELECT n/1 FROM t
to return 2.00.
However, if on another server I create a REMOTE TABLE
CREATE REMOTE TABLE t (
n DECIMAL(8,2)
)
ON '$SLAVE';
running the query against the remote table yields an assertion
failure:
Assertion failed: (bid == 0 || is_bat_nil(bid) || t == TYPE_any || ATOMtype(_b->ttype) == ATOMtype(t)), function runMALsequence, file /Users/joeri/monets/Mar2018_release/src/monetdb5/mal/mal_interpreter.c, line 661.
This is on the Mar2018_release tag. Interestingly, on the current tip
of the Mar2018 branch (06b77742de12c1cb2ca998f4b04a87cefcf5c1a1) there
is no assertion but an incorrect result:
sql>SELECT n/1 FROM t;
+---------------+
| L2 |
+===============+
| 100.000 |
+---------------+
Reproducible: Always
### Steps to Reproduce:
!/bin/sh
FARM="$PWD/FARM"
MASTER=mapi:monetdb://localhost:5501/master
SLAVE=mapi:monetdb://localhost:5502/slave
set -e -x
killall -9 monetdbd mserver5 || true
rm -rf "$FARM"
mkdir "$FARM"
start() {
name="${1*/}"
tmp="${1*:}"
port="${tmp%%/*}"
mserver5 --daemon=yes --dbpath="$FARM/$name" --set mapi_port="$port" \
2>&1 >"$FARM/$name.log" &
sleep 1
}
start "$MASTER"
start "$SLAVE"
mclient -d "$SLAVE" <<END
CREATE TABLE t (
n DECIMAL(8,2)
);
INSERT INTO t VALUES (2.00);
END
mclient -d "$MASTER" <<END
CREATE REMOTE TABLE t (
n DECIMAL(8,2)
)
ON '$SLAVE';
END
query() {
mclient -d "$1" <<END
SELECT n/1 FROM t;
END
}
query "$SLAVE"
query "$MASTER"
### Actual Results:
Assertion failure on Mar2018_release (see above).
100.000 on [06b77742de12c1cb2ca998f4b04a87cefcf5c1a1](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=06b77742de12c1cb2ca998f4b04a87cefcf5c1a1)
### Expected Results:
2.00
## Comment 26430
Date: 2018-05-10 12:01:20 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [439cd2cddb4c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=439cd2cddb4c) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=439cd2cddb4c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=439cd2cddb4c)
Changeset description:
fixes for bugs 6594 and 6595
6595, disabled optimizers based on arguments (A...) as
these are passed after the optimizer is run
Also approved output after recent changes on system functions
| Remote decimal division triggers assertion / returns wrong answer | https://api.github.com/repos/MonetDB/MonetDB/issues/6595/comments | 0 | 2020-11-30T16:27:57Z | 2024-06-27T13:04:12Z | https://github.com/MonetDB/MonetDB/issues/6595 | 753,620,919 | 6,595 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-29 11:33:57 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: 11.29.3 (Mar2018)
CC: @njnes
Last updated: 2018-06-20 11:02:16 +0200
## Comment 26413
Date: 2018-04-29 11:33:57 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.117 Safari/537.36
Build Identifier:
Run the query bellow for an assertion error during compilation.
Reproducible: Always
### Steps to Reproduce:
1. SELECT ALL + 62 * + - ( CASE WHEN NULL NOT IN ( CAST ( CAST ( - 77 AS INTEGER ) AS INTEGER ) ) THEN NULL WHEN ( NULL ) NOT IN ( 94 * COUNT ( * ) ) THEN + ( NULLIF ( 45, 80 + 90 ) ) - COUNT ( * ) ELSE NULL END ) AS col2;
### Actual Results:
Assertion error
mserver5: rel_bin.c:576: exp_bin: Assertion `s' failed.
### Expected Results:
A single NULL value.
The backtrace is already known from previous bugs :)
## Comment 26415
Date: 2018-04-29 11:39:20 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [dede692aec19](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=dede692aec19) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=dede692aec19](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=dede692aec19)
Changeset description:
Added tests for bugs 6577 and 6594
## Comment 26429
Date: 2018-05-10 12:01:19 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [439cd2cddb4c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=439cd2cddb4c) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=439cd2cddb4c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=439cd2cddb4c)
Changeset description:
fixes for bugs 6594 and 6595
6595, disabled optimizers based on arguments (A...) as
these are passed after the optimizer is run
Also approved output after recent changes on system functions
## Comment 26450
Date: 2018-05-17 08:06:00 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [1d3a15ef3aee](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=1d3a15ef3aee) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=1d3a15ef3aee](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=1d3a15ef3aee)
Changeset description:
added stable output for bug #6602 (and correct for bugs 6530 and 6594)
| Sqlitelogictest crash on complex case statement | https://api.github.com/repos/MonetDB/MonetDB/issues/6594/comments | 0 | 2020-11-30T16:27:54Z | 2024-06-27T13:04:11Z | https://github.com/MonetDB/MonetDB/issues/6594 | 753,620,880 | 6,594 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-27 15:57:13 +0200
From: Manuel <<manuel>>
To: SQL devs <<bugs-sql>>
Version: 11.27.11 (Jul2017-SP3)
Last updated: 2018-08-31 13:23:22 +0200
## Comment 26412
Date: 2018-04-27 15:57:13 +0200
From: Manuel <<manuel>>
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
I have recently is much faster when the escape clause is omitted fromt eh like operator. I can provide you the dataset used in my experiments, but it should not be hard to reproduce the issue on any dataset.
In my use case, submitting a query like:
where "country" like '%Aus%'
is almost ten time faster than
where "country" like '%Aus%' escape '!'
while in this case there is no need to escape anything, hence the escape clause can be removed, I found this behaviour quite strange.
On the opposite, when using ilike, the behaviour seems the reversed, and the query runs faster when escape clause is present.
Reproducible: Always
### Steps to Reproduce:
1. Create containing a varchar column, and a significant amount of records (I tried with 200 million records)
2. submit a query involving the like operator and the escape clause
3. submit a query involving the like operator without the escape clause
4. repeat step 2 and 3 using the ilike operator.
### Actual Results:
When using the like operator performance is much better when the escape keyword is omitted. When using the ilike operator performance is much better when the escape keyword is present
### Expected Results:
Performance not to be affected by the escape keyword
## Comment 26517
Date: 2018-06-27 16:54:06 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [29ae281a5ed3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=29ae281a5ed3) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=29ae281a5ed3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=29ae281a5ed3)
Changeset description:
If ESCAPE character does not occur in LIKE pattern, treat as if no ESCAPE.
This fixes bug #6593.
| Poor performance with like operator and escape clause | https://api.github.com/repos/MonetDB/MonetDB/issues/6593/comments | 0 | 2020-11-30T16:27:50Z | 2024-06-27T13:04:10Z | https://github.com/MonetDB/MonetDB/issues/6593 | 753,620,838 | 6,593 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-26 11:05:59 +0200
From: Harman DBA <<HCS-DL-INBE-IRI-DSGDB>>
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.29.3 (Mar2018)
CC: @yzchang
Last updated: 2018-06-05 20:41:50 +0200
## Comment 26410
Date: 2018-04-26 11:05:59 +0200
From: Harman DBA <<HCS-DL-INBE-IRI-DSGDB>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.117 Safari/537.36
Build Identifier:
Doing the followings at the same time:
1. create table xyz as select * from abc; ==> From Session 1
2. insert into abc select * from mno; ==> from Session 2
==> Hitting : COMMIT: transaction is aborted because of DDL concurrency conflicts, will ROLLBACK instead
Reproducible: Always
### Steps to Reproduce:
Doing the followings at the same time:
1. create table xyz as select * from abc; ==> From Session 1
2. insert into abc select * from mno; ==> from Session 2
==> Hitting : COMMIT: transaction is aborted because of DDL concurrency conflicts, will ROLLBACK instead
## Comment 26477
Date: 2018-06-05 20:41:50 +0200
From: @yzchang
Hai this is not a bug, it's how the MonetDB transaction manager normally works. For more information, please see: https://www.monetdb.org/blog/monetdb-sql-transaction-management-scheme
To resolve the situation, just repeat the aborted transaction until it succeeds. You can introduce a rondom waiting time before each retry to avoid conflict. Even better: serialise such concurrent transactions.
| COMMIT: transaction is aborted because of DDL concurrency conflicts, will ROLLBACK instead | https://api.github.com/repos/MonetDB/MonetDB/issues/6592/comments | 0 | 2020-11-30T16:27:46Z | 2024-06-28T06:46:31Z | https://github.com/MonetDB/MonetDB/issues/6592 | 753,620,784 | 6,592 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-25 16:30:44 +0200
From: @eyalroz
To: SQL devs <<bugs-sql>>
Version: 11.29.7 (Mar2018-SP1)
CC: @aris-koning, hknief, james, @kutsurak, m.raasveldt, martin.van.dinther, @njnes
Last updated: 2019-04-30 12:36:02 +0200
## Comment 26408
Date: 2018-04-25 16:30:44 +0200
From: @eyalroz
User-Agent: Mozilla/5.0 (X11; Fedora; Linux x86_64; rv:59.0) Gecko/20100101 Firefox/59.0
Build Identifier:
tl;dr: We should use an invalid character to concatenate names within sql_catalog_nme.
MonetDB persists an "SQL catalog", which maps SQL schema-table-column triplets to BAT ids within the pool. It's implemented using two BATs with fixed logical names sql_catalog_nme and sql_catalog_bid - a name column and a BAT id column.
Since only one string column is used, the triplet of strings we actually want to search for is mangled into a single string, by concatenating the names with underscore separators.
Example: If we're looking for schema "sys", table "ontime", column "flightdata", we'll actually look for "sys_ontime_flightdata" within "sql_catalog_nme", find it at a certain index in sql_catalog_nme, then get the BAT id from the corresponding index in sql_catalog_bid.
This approach is problematic, since schema, table and column names can themselves contain underscores. There is currently no escaping or underscore or special treatment for name collisions.
Example mclient session:
sql>create table ab ( cd_ef INTEGER );
operation successful
sql>create table ab_cd ( ef INTEGER );
CREATE TABLE: sys_ab_cd_ef conflicts
It is, of course, perfectly valid to create an SQL table "ab" with column "cd_ef" and table ab_cd with column "ef".
The problem is even more severe when trying to perform the _reverse_ lookup, i.e. trying to determine the SQL name triplet (schema, table, column) for a BAT (whose ID appears in sql_catalog_bid). Any presence of underscores in any of the three names creates an ambiguity when we wish to demangle. While MonetDB doesn't do this itself, it is not inconceivable to want to do this in forensic analysis.
Proposed solution:
Make the mangling unambiguous, by using a separator which cannot appear in any of the names. AFAICT, schema, table and column names all use the following grammar rules::
SQL language identifier> ::=
<SQL language identifier start>
[ { <underscore> | <SQL language identifier part> }... ]
<SQL language identifier start> ::= <simple Latin letter>
<SQL language identifier part> ::=
<simple Latin letter>
| <digit>
We could use something like a printing character such as ., , @ or %. I prefer the former - since that's what's often used intuitively, anyway.
Reproducible: Always
## Comment 26409
Date: 2018-04-25 16:49:32 +0200
From: @sjoerdmullender
Removed my name: I don't need two copies.
## Comment 26423
Date: 2018-05-09 08:57:12 +0200
From: @njnes
quoted identifiers allow all your alternatives, ie not that easy to fix
## Comment 26622
Date: 2018-09-24 02:14:25 +0200
From: James Watmuff <<james>>
>> quoted identifiers allow all your alternatives, ie not that easy to fix
Instead of using a different separator character, could we just escape the separator character ('_') in the filename if it occurs within an identifier? Escaping could be achieved by repeating the separator character twice.
Examples:
(sys, ab, cd) => sys_ab_cd
(sys, ab, cd_ef) => sys_ab_cd__ef
(sys, ab_cd, ef) => sys_ab__cd_ef
(sys, ab__cd, ef) => sys_ab____cd_ef
## Comment 26623
Date: 2018-09-24 10:29:33 +0200
From: @sjoerdmullender
(In reply to James Watmuff from comment 3)
> >> quoted identifiers allow all your alternatives, ie not that easy to fix
>
> Instead of using a different separator character, could we just escape the
> separator character ('_') in the filename if it occurs within an identifier?
> Escaping could be achieved by repeating the separator character twice.
>
> Examples:
> (sys, ab, cd) => sys_ab_cd
> (sys, ab, cd_ef) => sys_ab_cd__ef
> (sys, ab_cd, ef) => sys_ab__cd_ef
> (sys, ab__cd, ef) => sys_ab____cd_ef
Sorry, that's still ambiguous. If you have a sequence of three underscores (because one of the constituent names starts or ends with an underscore) how would you group them?
We did come up with a potential solution to this problem, though. It will take time to implement (not because it's particularly difficult, just because there is plenty of other work): use various control characters instead of underscore (and in some other internally generated names). I'm assuming that control characters cannot be specified at the SQL level, so you cannot have names that contain those, so they should be good and unambiguous for internal use.
## Comment 26624
Date: 2018-09-24 11:13:15 +0200
From: @eyalroz
(In reply to Sjoerd Mullender from comment 4)
Indeed, James Watmuff's suggestion is ambiguous
A control-character-based solution would work, obviously; but an unambiguous and possibly more readable escaping scheme can also be chosen. For example: The sequence _X_ could be used, with X representing characters which would not used directly, such as a dot, an underscore and possibly others. This has the benefit of being printable, readable, and safe for copy-pasting as text; and it would not clash with backslash-based or brackets-based schemes. Of course, whatever works is fine.
I should also mention that, even for forensics, the inability to determine exact column names is not critical and can be worked around: One can locate the system tables in sql_catalog_nme as-is; and using these tables, determine the full names. This is what MonetDB itself does, I suppose, and it's what I now do with the MonetDB BBP reader library (https://bitbucket.org/eyalroz/monetdb-bbp-reader).
## Comment 26625
Date: 2018-09-24 11:43:08 +0200
From: @sjoerdmullender
(In reply to Eyal Rozenberg from comment 5)
> (In reply to Sjoerd Mullender from comment 4)
>
> Indeed, James Watmuff's suggestion is ambiguous
>
> A control-character-based solution would work, obviously; but an unambiguous
> and possibly more readable escaping scheme can also be chosen. For example:
> The sequence _X_ could be used, with X representing characters which would
> not used directly, such as a dot, an underscore and possibly others. This
> has the benefit of being printable, readable, and safe for copy-pasting as
> text; and it would not clash with backslash-based or brackets-based schemes.
> Of course, whatever works is fine.
Sorry, doesn't work either. As long as you can have a schema/table/column name that can contain the sequence that is used to glue those bits together, things are ambiguous. What *could* work if whenever such a sequence occurs inside one of those part, that sequence is somehow escaped. For instance, use _. as glue, and whenever there is an underscore in the schema/table/column name, replace it with a double _ (__). Then when you're reading the glued-together name from left to right, whenever you encounter an underscore, the following character determines whether it's a separator (.) or part of the name (_).
## Comment 26626
Date: 2018-09-24 12:14:45 +0200
From: @eyalroz
(In reply to Sjoerd Mullender from comment 6)
> Sorry, doesn't work either. As long as you can have a schema/table/column
> name that can contain the sequence that is used to glue those bits together,
> things are ambiguous.
It would work, I believe, since you also escape all underscores. So underscores would only ever be used as an escape character.
## Comment 26627
Date: 2018-09-25 01:19:08 +0200
From: James Watmuff <<james>>
Oops - yes my original suggestion was obviously flawed.
It just occurred to me that escaping SQL identifiers is already a solved problem using double quotes.
So what about just "schema"."table"."column", where (") inside an identifier is escaped as ("").
## Comment 26701
Date: 2018-11-28 17:54:29 +0100
From: Herman Knief <<hknief>>
I have recently been evaluating a number of engines for use with an existing platform/application. I've really liked what I have seen with MonetDB so far, but have run into this issue. Given the pre-existence of the code I have to work with and number of deploy instances, changing the schema is not a good option. Would love to see this get a little higher priority for a fix so that I could continue. Thanks.
## Comment 26859
Date: 2019-01-30 11:00:56 +0100
From: @njnes
The bats within the logger are now found using an id. This solves any name
conflict introduced by the mangling. Solution will be part of the April
release.
## Comment 26860
Date: 2019-01-30 11:04:23 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [eb19fe85e464](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=eb19fe85e464) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=eb19fe85e464](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=eb19fe85e464)
Changeset description:
fix bug #6591, bats within the logger are found using the sqlid, which solves
any naming conflict introduced by the name mangling.
| Name mangling in the SQL catalog results in ambiguity | https://api.github.com/repos/MonetDB/MonetDB/issues/6591/comments | 0 | 2020-11-30T16:27:42Z | 2024-06-27T13:04:08Z | https://github.com/MonetDB/MonetDB/issues/6591 | 753,620,736 | 6,591 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-25 14:28:36 +0200
From: Harman DBA <<HCS-DL-INBE-IRI-DSGDB>>
To: Merovingian devs <<bugs-merovingian>>
Version: 11.29.7 (Mar2018-SP1)
CC: HCS-DL-INBE-IRI-DSGDB, martin.van.dinther
Last updated: 2020-07-22 17:07:22 +0200
## Comment 26407
Date: 2018-04-25 14:28:36 +0200
From: Harman DBA <<HCS-DL-INBE-IRI-DSGDB>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.117 Safari/537.36
Build Identifier:
As per business requirement - our database must be available 24X7. For cloning the database - we need to have read consistent version of all the bat files .
The read consistency is feasible if we shutdown the database . The challenge is - our database must be available 24X7 . My concern is - how we can have read consistent version of the bat files(for cloning) without shutting down the database
Reproducible: Always
### Steps to Reproduce:
1.NA
2.
3.
### Actual Results:
NA
### Expected Results:
NA
NA
## Comment 26411
Date: 2018-04-26 14:41:02 +0200
From: Martin van Dinther <<martin.van.dinther>>
What you are looking for is called a Hot Backup.
MonetDB does not support Hot Backups.
In order to make an active database behave read only, you first need to make all user tables read only via SQL: ALTER TABLE <full qualified table name> SET READ ONLY;
However this does not prevent users to issue DDL commands, which change the database and SQL catalog. To prevent that also you must lock the database. See: https://www.monetdb.org/Documentation/monetdb-man-page
Next you can use msqldump program to dump the locked database.
See: https://www.monetdb.org/Documentation/msqldump-man-page
I changed the Importance from "critical" to "enhancement" as it is not a bug, but an enhancement request.
## Comment 26809
Date: 2019-01-18 15:57:51 +0100
From: @sjoerdmullender
Note that msqldump runs a single transaction to dump the complete database. This means that the result is a consistent version of the database, even if other clients make changes while the database is being dumped.
## Comment 27932
Date: 2020-07-22 17:06:43 +0200
From: Martin van Dinther <<martin.van.dinther>>
FYI: In Jun2020 release we have introduced a new procedure:
call hot_snapshot('full path to tar file name.tar');
which allows you to create a concistent copy of all the bat files into one tar file optionally compressed (when adding .lz4 or .gz or .bzip2 or .xv to the end of the tar file name) of a running monetdb server.
This new procedure may fulfil your business requirement to create a clone of the database.
Let us know if we may set this issue to resolved?
| How to place database in readonly mode without shutdown the database . | https://api.github.com/repos/MonetDB/MonetDB/issues/6590/comments | 0 | 2020-11-30T16:27:39Z | 2024-06-28T13:39:45Z | https://github.com/MonetDB/MonetDB/issues/6590 | 753,620,684 | 6,590 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-21 16:47:01 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: 11.29.3 (Mar2018)
CC: @njnes
Last updated: 2018-06-20 11:02:14 +0200
## Comment 26400
Date: 2018-04-21 16:47:01 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.117 Safari/537.36
Build Identifier:
Run the query bellow for an assertion error in rel_bin.c.
Quick note: This query is also crashing on Mar2018 branch.
Reproducible: Always
### Steps to Reproduce:
1. SELECT DISTINCT + - 82 * + 26 - 12 * + - 27 * + + CASE WHEN NOT NULL IN ( - 69, ( + SUM ( DISTINCT CAST ( NULL AS INTEGER ) ) ) + + MAX ( - 6 ) + + - COUNT ( * ) / - - ( + 77 ) + 37, + 44, - - 0 * + + 23, - - 86, 3, 12 ) THEN + 41 WHEN NOT - 21 / - 95 IS NOT NULL THEN 63 END AS col0;
### Actual Results:
Assertion error:
mserver5: rel_bin.c:576: exp_bin: Assertion `s' failed
### Expected Results:
A single NULL value.
Backtrace:
0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:51
1 0x00007ffff4781c41 in __GI_abort () at abort.c:79
2 0x00007ffff4778f7a in __assert_fail_base (fmt=0x7ffff48c9260 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", assertion=assertion@entry=0x7fffe84fdb78 "s", file=file@entry=0x7fffe84fdb08 "rel_bin.c",
line=line@entry=576, function=function@entry=0x7fffe84fe348 <__PRETTY_FUNCTION__.18608> "exp_bin") at assert.c:92
3 0x00007ffff4778ff2 in __GI___assert_fail (assertion=0x7fffe84fdb78 "s", file=0x7fffe84fdb08 "rel_bin.c", line=576, function=0x7fffe84fe348 <__PRETTY_FUNCTION__.18608> "exp_bin") at assert.c:101
4 0x00007fffe8383d0e in exp_bin (be=0x7fffb0002b60, e=0x7fffb00f19c0, left=0x7fffb010e080, right=0x7fffb0142470, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:576
5 0x00007fffe83831a2 in exp_bin (be=0x7fffb0002b60, e=0x7fffb00ff440, left=0x7fffb010e080, right=0x7fffb0142470, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:450
6 0x00007fffe838d003 in rel2bin_project (be=0x7fffb0002b60, rel=0x7fffb00f9130, refs=0x7fffb010c520, topn=0x0) at rel_bin.c:2481
7 0x00007fffe8398bed in subrel_bin (be=0x7fffb0002b60, rel=0x7fffb00f9130, refs=0x7fffb010c520) at rel_bin.c:5068
8 0x00007fffe838cf30 in rel2bin_project (be=0x7fffb0002b60, rel=0x7fffb00fe7a0, refs=0x7fffb010c520, topn=0x0) at rel_bin.c:2469
9 0x00007fffe8398bed in subrel_bin (be=0x7fffb0002b60, rel=0x7fffb00fe7a0, refs=0x7fffb010c520) at rel_bin.c:5068
10 0x00007fffe8398ee0 in _subrel_bin (be=0x7fffb0002b60, rel=0x7fffb00fe7a0, refs=0x7fffb010c520) at rel_bin.c:5131
11 0x00007fffe8398ff3 in output_rel_bin (be=0x7fffb0002b60, rel=0x7fffb00fe7a0) at rel_bin.c:5154
12 0x00007fffe83b5423 in sql_relation2stmt (be=0x7fffb0002b60, r=0x7fffb00fe7a0) at sql_gencode.c:542
13 0x00007fffe83b5616 in backend_dumpstmt (be=0x7fffb0002b60, mb=0x7fffb00e9150, r=0x7fffb00fe7a0, top=1, add_end=0,
query=0x7fffb00e9310 "select distinct + - 82 * + 26 - 12 * + - 27 * + + case when not null in ( - 69, ( + sum ( distinct cast ( null as integer ) ) ) + + max ( - 6 ) + + - count ( * ) / - - ( + 77 ) + 37, + 44, - - 0 * + +"...) at sql_gencode.c:582
14 0x00007fffe83775df in SQLparser (c=0x7fffe973c368) at sql_scenario.c:1195
15 0x00007ffff7a0f378 in runPhase (c=0x7fffe973c368, phase=1) at mal_scenario.c:510
16 0x00007ffff7a0f481 in runScenarioBody (c=0x7fffe973c368, once=0) at mal_scenario.c:532
17 0x00007ffff7a0f73a in runScenario (c=0x7fffe973c368, once=0) at mal_scenario.c:569
18 0x00007ffff7a11666 in MSserveClient (dummy=0x7fffe973c368) at mal_session.c:519
19 0x00007ffff7a11161 in MSscheduleClient (command=0x7fffb0000b30 "0", challenge=0x7fffc72ecd9b "cslQC9fFutE", fin=0x1813590, fout=0x7fffc0002d90, protocol=PROTOCOL_9, blocksize=8190, compute_column_widths=0)
at mal_session.c:397
20 0x00007ffff7ab22c8 in doChallenge (data=0x18135a0) at mal_mapi.c:279
21 0x00007ffff76989ec in thread_starter (arg=0x9c3930) at gdk_system.c:476
22 0x00007ffff4b0950b in start_thread (arg=0x7fffc72ed700) at pthread_create.c:465
23 0x00007ffff484116f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95
## Comment 26401
Date: 2018-04-21 16:47:23 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [64f13600ec9e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=64f13600ec9e) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=64f13600ec9e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=64f13600ec9e)
Changeset description:
Added test for bug #6589
## Comment 26402
Date: 2018-04-22 11:54:17 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [354f2d85e58b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=354f2d85e58b) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=354f2d85e58b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=354f2d85e58b)
Changeset description:
fixes for bug #6589
| Sqlitelogictest crash on complex on complex expression | https://api.github.com/repos/MonetDB/MonetDB/issues/6589/comments | 0 | 2020-11-30T16:27:35Z | 2024-06-27T13:04:05Z | https://github.com/MonetDB/MonetDB/issues/6589 | 753,620,627 | 6,589 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-21 09:12:11 +0200
From: Guillaume de GENTILE <<gentile_g>>
To: SQL devs <<bugs-sql>>
Version: 11.29.7 (Mar2018-SP1)
CC: @PedroTadim
Last updated: 2019-04-30 12:36:01 +0200
## Comment 26397
Date: 2018-04-21 09:12:11 +0200
From: Guillaume de GENTILE <<gentile_g>>
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
Table aliasing are supported in SQL select queries:
select a.* from table123 a --> OK
but not in SQL update queries!
update table123 a set a.COL1=0 --> ERROR
update table123 set COL1=0 --> OK
Reproducible: Always
### Steps to Reproduce:
1.create a dummy table table123
2.update table using a table alias in SQL query
3.update table123 a set a.COL1=0 --> ERROR
### Actual Results:
update table123 not successful
An error occurred when executing the SQL command:
update table123 a set COL1=0
ParseException:SQLparser:!42000!syntax error, unexpected IDENT, expecting SET in: "update table123 a" [SQL State=M0M10]
1 statement failed.
Execution time: 0s
### Expected Results:
We need to be able to define table aliasing in SQL update queries.
update table123 a set a.COL1=0
## Comment 26705
Date: 2018-11-29 17:57:37 +0100
From: @PedroTadim
I implemented this feature in merge-statements branch of MonetDB and it should be available in a future release.
| Table aliasing are not supported in SQL update queries | https://api.github.com/repos/MonetDB/MonetDB/issues/6588/comments | 0 | 2020-11-30T16:27:32Z | 2024-06-27T13:04:04Z | https://github.com/MonetDB/MonetDB/issues/6588 | 753,620,583 | 6,588 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-20 16:23:19 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-06-20 11:02:11 +0200
## Comment 26391
Date: 2018-04-20 16:23:19 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.117 Safari/537.36
Build Identifier:
On default branch only, run the query bellow for an assertion error in rel_bin.c
Reproducible: Always
### Steps to Reproduce:
1. SELECT ALL + CASE WHEN NOT + - 10 IS NULL AND NULL IS NULL THEN NULL WHEN - - 39 * - 89 NOT IN ( - + 51 * - - 45, COUNT ( * ) / + 67 ) THEN + COUNT ( ALL 54 ) END;
### Actual Results:
mserver5: rel_bin.c:605: exp_bin: Assertion `s' failed
### Expected Results:
An single NULL value.
Backtrace:
0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:51
1 0x00007ffff4781c41 in __GI_abort () at abort.c:79
2 0x00007ffff4778f7a in __assert_fail_base (fmt=0x7ffff48c9260 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", assertion=assertion@entry=0x7fffe84fdb98 "s", file=file@entry=0x7fffe84fdb28 "rel_bin.c",
line=line@entry=605, function=function@entry=0x7fffe84fe368 <__PRETTY_FUNCTION__.18631> "exp_bin") at assert.c:92
3 0x00007ffff4778ff2 in __GI___assert_fail (assertion=0x7fffe84fdb98 "s", file=0x7fffe84fdb28 "rel_bin.c", line=605, function=0x7fffe84fe368 <__PRETTY_FUNCTION__.18631> "exp_bin") at assert.c:101
4 0x00007fffe8382226 in exp_bin (be=0x7fffb0002b60, e=0x7fffb00f6df0, left=0x7fffb00febe0, right=0x7fffb00fed10, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:605
5 0x00007fffe8381612 in exp_bin (be=0x7fffb0002b60, e=0x7fffb00f9830, left=0x7fffb00febe0, right=0x7fffb00fed10, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:480
6 0x00007fffe8381612 in exp_bin (be=0x7fffb0002b60, e=0x7fffb00f9b60, left=0x7fffb00febe0, right=0x7fffb00fed10, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:480
7 0x00007fffe8381612 in exp_bin (be=0x7fffb0002b60, e=0x7fffb00f9fe0, left=0x7fffb00febe0, right=0x7fffb00fed10, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:480
8 0x00007fffe8381612 in exp_bin (be=0x7fffb0002b60, e=0x7fffb00fa070, left=0x7fffb00febe0, right=0x7fffb00fed10, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:480
9 0x00007fffe838b536 in rel2bin_project (be=0x7fffb0002b60, rel=0x7fffb00f2ee0, refs=0x7fffb00fdfb0, topn=0x0) at rel_bin.c:2510
10 0x00007fffe8397120 in subrel_bin (be=0x7fffb0002b60, rel=0x7fffb00f2ee0, refs=0x7fffb00fdfb0) at rel_bin.c:5097
11 0x00007fffe838b463 in rel2bin_project (be=0x7fffb0002b60, rel=0x7fffb00f8190, refs=0x7fffb00fdfb0, topn=0x0) at rel_bin.c:2498
12 0x00007fffe8397120 in subrel_bin (be=0x7fffb0002b60, rel=0x7fffb00f8190, refs=0x7fffb00fdfb0) at rel_bin.c:5097
13 0x00007fffe8397413 in _subrel_bin (be=0x7fffb0002b60, rel=0x7fffb00f8190, refs=0x7fffb00fdfb0) at rel_bin.c:5160
14 0x00007fffe8397526 in output_rel_bin (be=0x7fffb0002b60, rel=0x7fffb00f8190) at rel_bin.c:5183
15 0x00007fffe83b4359 in sql_relation2stmt (be=0x7fffb0002b60, r=0x7fffb00f8190) at sql_gencode.c:608
16 0x00007fffe83b454c in backend_dumpstmt (be=0x7fffb0002b60, mb=0x7fffb0111ed0, r=0x7fffb00f8190, top=1, add_end=1,
query=0x7fffb0112090 "select all + case when not + - 10 is null and null is null then null when - - 39 * - 89\nnot in ( - + 51 * - - 45, count ( * ) / + 67 ) then + count ( all 54 ) end;")
at sql_gencode.c:648
17 0x00007fffe83b4dc4 in backend_dumpproc (be=0x7fffb0002b60, c=0x7fffe973c368, cq=0x7fffb0116700, r=0x7fffb00f8190) at sql_gencode.c:778
18 0x00007fffe837588d in SQLparser (c=0x7fffe973c368) at sql_scenario.c:1233
19 0x00007ffff7a0d9eb in runPhase (c=0x7fffe973c368, phase=1) at mal_scenario.c:510
20 0x00007ffff7a0daf4 in runScenarioBody (c=0x7fffe973c368, once=0) at mal_scenario.c:532
21 0x00007ffff7a0ddad in runScenario (c=0x7fffe973c368, once=0) at mal_scenario.c:569
22 0x00007ffff7a0fcdf in MSserveClient (dummy=0x7fffe973c368) at mal_session.c:519
23 0x00007ffff7a0f7d4 in MSscheduleClient (command=0x7fffb0000b30 "0", challenge=0x7fffc72ecd9b "o0Tk43R22u", fin=0x1819dd0, fout=0x7fffc0002d90, protocol=PROTOCOL_9, blocksize=8190, compute_column_widths=0)
at mal_session.c:397
24 0x00007ffff7ab150e in doChallenge (data=0x1819de0) at mal_mapi.c:279
25 0x00007ffff769787c in thread_starter (arg=0x9c7cc0) at gdk_system.c:476
26 0x00007ffff4b0950b in start_thread (arg=0x7fffc72ed700) at pthread_create.c:465
27 0x00007ffff484116f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95
## Comment 26392
Date: 2018-04-20 16:24:05 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [98e9ba8965ca](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=98e9ba8965ca) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=98e9ba8965ca](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=98e9ba8965ca)
Changeset description:
Added test for bug #6587
## Comment 26399
Date: 2018-04-21 15:50:53 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [30a19086985a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=30a19086985a) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=30a19086985a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=30a19086985a)
Changeset description:
fixes for bug #6587
| Sqlitelogictest crash on complex case statement | https://api.github.com/repos/MonetDB/MonetDB/issues/6587/comments | 0 | 2020-11-30T16:27:28Z | 2024-06-27T13:04:03Z | https://github.com/MonetDB/MonetDB/issues/6587 | 753,620,534 | 6,587 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-20 10:43:01 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: martin.van.dinther, @njnes
Last updated: 2018-08-31 13:23:15 +0200
## Comment 26387
Date: 2018-04-20 10:43:01 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.117 Safari/537.36
Build Identifier:
Run the query bellow for an assertion error on rel_bin.c. This bug only happens on default branch.
Reproducible: Always
### Steps to Reproduce:
1. SELECT - 59 + - 66, - 66 * + - COUNT ( ALL COALESCE ( NULLIF ( 90, + COALESCE ( - 67, 72 + - 30 ) + 17 ), 90 + 96, + 63 ) ) AS col1;
### Actual Results:
mserver5: rel_bin.c:605: exp_bin: Assertion `s' failed.
### Expected Results:
A single row with values -125 and 66
Backtrace:
0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:51
1 0x00007ffff4781c41 in __GI_abort () at abort.c:79
2 0x00007ffff4778f7a in __assert_fail_base (fmt=0x7ffff48c9260 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", assertion=assertion@entry=0x7fffe84fda58 "s", file=file@entry=0x7fffe84fd9e8 "rel_bin.c",
line=line@entry=605, function=function@entry=0x7fffe84fe228 <__PRETTY_FUNCTION__.18631> "exp_bin") at assert.c:92
3 0x00007ffff4778ff2 in __GI___assert_fail (assertion=0x7fffe84fda58 "s", file=0x7fffe84fd9e8 "rel_bin.c", line=605, function=0x7fffe84fe228 <__PRETTY_FUNCTION__.18631> "exp_bin") at assert.c:101
4 0x00007fffe8382226 in exp_bin (be=0x7fffb0002b60, e=0x7fffb02348b0, left=0x0, right=0x7fffb0375e10, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:605
5 0x00007fffe8381612 in exp_bin (be=0x7fffb0002b60, e=0x7fffb0234c90, left=0x0, right=0x7fffb0375e10, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:480
6 0x00007fffe838b536 in rel2bin_project (be=0x7fffb0002b60, rel=0x7fffb0230670, refs=0x7fffb0375cc0, topn=0x0) at rel_bin.c:2510
7 0x00007fffe8397120 in subrel_bin (be=0x7fffb0002b60, rel=0x7fffb0230670, refs=0x7fffb0375cc0) at rel_bin.c:5097
8 0x00007fffe838b463 in rel2bin_project (be=0x7fffb0002b60, rel=0x7fffb0232a40, refs=0x7fffb0375cc0, topn=0x0) at rel_bin.c:2498
9 0x00007fffe8397120 in subrel_bin (be=0x7fffb0002b60, rel=0x7fffb0232a40, refs=0x7fffb0375cc0) at rel_bin.c:5097
10 0x00007fffe838c3f1 in rel2bin_groupby (be=0x7fffb0002b60, rel=0x7fffb02303c0, refs=0x7fffb0375cc0) at rel_bin.c:2715
11 0x00007fffe839717c in subrel_bin (be=0x7fffb0002b60, rel=0x7fffb02303c0, refs=0x7fffb0375cc0) at rel_bin.c:5105
12 0x00007fffe838b463 in rel2bin_project (be=0x7fffb0002b60, rel=0x7fffb0229d00, refs=0x7fffb0375cc0, topn=0x0) at rel_bin.c:2498
13 0x00007fffe8397120 in subrel_bin (be=0x7fffb0002b60, rel=0x7fffb0229d00, refs=0x7fffb0375cc0) at rel_bin.c:5097
14 0x00007fffe838b463 in rel2bin_project (be=0x7fffb0002b60, rel=0x7fffb022a890, refs=0x7fffb0375cc0, topn=0x0) at rel_bin.c:2498
15 0x00007fffe8397120 in subrel_bin (be=0x7fffb0002b60, rel=0x7fffb022a890, refs=0x7fffb0375cc0) at rel_bin.c:5097
16 0x00007fffe838b463 in rel2bin_project (be=0x7fffb0002b60, rel=0x7fffb02375a0, refs=0x7fffb0375cc0, topn=0x0) at rel_bin.c:2498
17 0x00007fffe8397120 in subrel_bin (be=0x7fffb0002b60, rel=0x7fffb02375a0, refs=0x7fffb0375cc0) at rel_bin.c:5097
18 0x00007fffe8397413 in _subrel_bin (be=0x7fffb0002b60, rel=0x7fffb02375a0, refs=0x7fffb0375cc0) at rel_bin.c:5160
19 0x00007fffe8397526 in output_rel_bin (be=0x7fffb0002b60, rel=0x7fffb02375a0) at rel_bin.c:5183
20 0x00007fffe83b42c0 in sql_relation2stmt (be=0x7fffb0002b60, r=0x7fffb02375a0) at sql_gencode.c:608
21 0x00007fffe83b44b3 in backend_dumpstmt (be=0x7fffb0002b60, mb=0x7fffb00691c0, r=0x7fffb02375a0, top=1, add_end=0,
query=0x1a73a30 "select - 59 + - 66, - 66 * + - count ( all coalesce ( nullif ( 90, + coalesce ( - 67, 72 + - 30 ) + 17 ), 90 + 96, + 63 ) ) as col1;") at sql_gencode.c:648
22 0x00007fffe83754f4 in SQLparser (c=0x7fffe973c368) at sql_scenario.c:1195
23 0x00007ffff7a0d9eb in runPhase (c=0x7fffe973c368, phase=1) at mal_scenario.c:510
24 0x00007ffff7a0daf4 in runScenarioBody (c=0x7fffe973c368, once=0) at mal_scenario.c:532
25 0x00007ffff7a0ddad in runScenario (c=0x7fffe973c368, once=0) at mal_scenario.c:569
26 0x00007ffff7a0fcdf in MSserveClient (dummy=0x7fffe973c368) at mal_session.c:519
27 0x00007ffff7a0f7d4 in MSscheduleClient (command=0x7fffb0000b30 "0", challenge=0x7fffc72ecd9b "6LXsnZvzFMe", fin=0x1801b50, fout=0x7fffc0002d90, protocol=PROTOCOL_9, blocksize=8190, compute_column_widths=0)
at mal_session.c:397
28 0x00007ffff7ab150e in doChallenge (data=0x1801b60) at mal_mapi.c:279
29 0x00007ffff769787c in thread_starter (arg=0x9afa40) at gdk_system.c:476
30 0x00007ffff4b0950b in start_thread (arg=0x7fffc72ed700) at pthread_create.c:465
31 0x00007ffff484116f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95
## Comment 26389
Date: 2018-04-20 11:07:40 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [bddfde338d15](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=bddfde338d15) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=bddfde338d15](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=bddfde338d15)
Changeset description:
Added test for bug #6586
## Comment 26390
Date: 2018-04-20 15:56:08 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [7c2cd5b6f4bf](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=7c2cd5b6f4bf) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=7c2cd5b6f4bf](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=7c2cd5b6f4bf)
Changeset description:
fixes for bug #6586 (also fixed div by zero problems)
| Sqlitelogictest crash on complex aggregation query | https://api.github.com/repos/MonetDB/MonetDB/issues/6586/comments | 0 | 2020-11-30T16:27:25Z | 2024-06-27T13:04:02Z | https://github.com/MonetDB/MonetDB/issues/6586 | 753,620,483 | 6,586 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-19 13:59:53 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: 11.29.3 (Mar2018)
Last updated: 2018-06-20 11:02:11 +0200
## Comment 26383
Date: 2018-04-19 13:59:53 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.117 Safari/537.36
Build Identifier:
Add a merge table 2 into merge table 1, then merge table 1 into merge table 2. Doing a select query into any of these tables will cause an infinite loop in rel_optimizer. However I already fixed it.
Reproducible: Always
### Steps to Reproduce:
1. create merge table test1 (a int);
2. create merge table test2 (a int);
3. alter table test1 add table test2;
4. alter table test2 add table test1;
5. select a from test1;
### Actual Results:
An infinite loop in rel_optimizer.
### Expected Results:
An empty result set.
I already fixed it.
## Comment 26384
Date: 2018-04-19 14:01:08 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [097dd32a0c74](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=097dd32a0c74) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=097dd32a0c74](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=097dd32a0c74)
Changeset description:
Added test and fixes bug #6585
ie check for the table to be added if it is already a parent of the merge table
| Nested Merge tables cause an infinite loop in rel_optimizer | https://api.github.com/repos/MonetDB/MonetDB/issues/6585/comments | 0 | 2020-11-30T16:27:22Z | 2024-06-27T13:04:01Z | https://github.com/MonetDB/MonetDB/issues/6585 | 753,620,434 | 6,585 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-19 13:53:35 +0200
From: @joerivanruth
To: SQL devs <<bugs-sql>>
Version: 11.29.3 (Mar2018)
CC: @njnes
Last updated: 2018-06-20 11:02:12 +0200
## Comment 26382
Date: 2018-04-19 13:53:35 +0200
From: @joerivanruth
I've tested this on Linux and MacOS, and on 'default', 'Mar2018',
'Jul2017' and 'Dec2016'.
Here's a shell script that reproduces it. It connects to two
databases, creates a table on one and a REMOTE table on the other,
then runs some queries.
The result of the final query comes out empty while it shouldn't be.
!/bin/bash
set -e
MASTER=mapi:monetdb://localhost:50000/master
SLAVE=mapi:monetdb://localhost:50000/slave
run() {
local url="$1"
shift
echo
echo "Connecting to $url"
mclient -d "$url" -e "$@"
}
run $SLAVE <<END
DROP TABLE IF EXISTS t;
CREATE TABLE t (i INTEGER);
INSERT INTO t VALUES (1), (2), (3);
SELECT * FROM t;
SELECT * FROM t WHERE i IS NOT NULL;
END
run $MASTER <<END
DROP TABLE IF EXISTS t;
CREATE REMOTE TABLE t (i INTEGER) ON '$SLAVE';
SELECT * FROM t;
SELECT * FROM t WHERE i IS NOT NULL;
END
## Comment 26393
Date: 2018-04-20 17:27:01 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [12a61dac9008](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=12a61dac9008) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=12a61dac9008](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=12a61dac9008)
Changeset description:
fixes for bug #6583 and 6584
## Comment 26396
Date: 2018-04-20 17:29:23 +0200
From: @njnes
fixed the relational intermediate language parsing.
| SELECT FROM REMOTE TABLE WHERE IS NOT NULL produces wrong result | https://api.github.com/repos/MonetDB/MonetDB/issues/6584/comments | 0 | 2020-11-30T16:27:18Z | 2024-06-27T13:04:00Z | https://github.com/MonetDB/MonetDB/issues/6584 | 753,620,390 | 6,584 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-19 10:08:06 +0200
From: @swingbit
To: SQL devs <<bugs-sql>>
Version: 11.27.13 (Jul2017-SP4)
CC: @njnes
Last updated: 2018-06-20 11:02:18 +0200
## Comment 26381
Date: 2018-04-19 10:08:06 +0200
From: @swingbit
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
sql>trace select '12345' || 678;
+------+
| L2 |
+======+
| 1234 |
: 5678 :
+------+
1 tuple (3.652ms)
+------+--------------------------------------------------------------------------------------------------------+
| usec | statement |
+======+========================================================================================================+
| 98 | X_3=0@0:void := querylog.define("trace select \\'12345\\' || 678;":str, "sequential_pipe":str, 9:int); |
| 167 | sql.resultSet(".L2":str, "L2":str, "char":str, 0:int, 0:int, 3:int, "12345678":str); |
+------+--------------------------------------------------------------------------------------------------------+
I assume that integer 678 is cast to a (fixed size) string before concatenation.
However, sql.resultSet shows a size 0.
Notice also how the result is oddly formatted.
Reproducible: Always
## Comment 26388
Date: 2018-04-20 11:00:45 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [b3f10488c2e7](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=b3f10488c2e7) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=b3f10488c2e7](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=b3f10488c2e7)
Changeset description:
Added test for bug #6583
## Comment 26394
Date: 2018-04-20 17:27:02 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [12a61dac9008](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=12a61dac9008) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=12a61dac9008](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=12a61dac9008)
Changeset description:
fixes for bug #6583 and 6584
## Comment 26395
Date: 2018-04-20 17:28:23 +0200
From: @njnes
the general charachter types of the concat function turn the result in a
char(0), which is now handled by the output (result set) code using the same code as varchar(0)
| Fixed size string concatenation with integer results in fixed size string of size 0 | https://api.github.com/repos/MonetDB/MonetDB/issues/6583/comments | 0 | 2020-11-30T16:27:15Z | 2024-06-27T13:03:59Z | https://github.com/MonetDB/MonetDB/issues/6583 | 753,620,354 | 6,583 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-18 14:58:23 +0200
From: Guillaume de GENTILE <<gentile_g>>
To: SQL devs <<bugs-sql>>
Version: 11.27.9 (Jul2017-SP2)
CC: martin.van.dinther
Last updated: 2018-04-18 18:09:31 +0200
## Comment 26376
Date: 2018-04-18 14:58:23 +0200
From: Guillaume de GENTILE <<gentile_g>>
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
Is it possible to support other encoding than UTF-8 with Java memory database
Reproducible: Always
### Actual Results:
Process failed to parse special characters if file encoding is not "UTF-8"
### Expected Results:
Requirement is to be able to specify the input file encoding.
I see 2 possibilities:
1) Set default encoding in the Database URL.
Example:
<database url>?charset=ISO-8859-1
or
<database url>?encoding=ISO-8859-1
2) Enhance CSV load:
COPY INTO table123 from 'sdsdsdsd.csv' CHARSET ISO-8859-1
or
COPY INTO table123 from 'sdsdsdsd.csv' ENCODING ISO-8859-1
Thanks and regards,
Guillaume
## Comment 26378
Date: 2018-04-18 18:09:31 +0200
From: Martin van Dinther <<martin.van.dinther>>
Changed the Product label value from Client interfaces/JDBC into SQL
as this is an enhencement request to: extend the "COPY INTO <table> FROM <file>" SQL command, which currently only supports files with data in UTF-8 encoding.
| CSV load only support UTF-8 encoding | https://api.github.com/repos/MonetDB/MonetDB/issues/6582/comments | 0 | 2020-11-30T16:27:12Z | 2024-06-28T13:39:45Z | https://github.com/MonetDB/MonetDB/issues/6582 | 753,620,313 | 6,582 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-18 10:50:55 +0200
From: Manuel <<manuel>>
To: SQL devs <<bugs-sql>>
Version: 11.29.3 (Mar2018)
CC: @njnes
Last updated: 2018-06-20 11:02:17 +0200
## Comment 26375
Date: 2018-04-18 10:50:55 +0200
From: Manuel <<manuel>>
Created attachment 601
The sql to create the table used to reproduce the bug
I have spotted an issue on the join operator. I have attached the sql statements to reproduce the queries reported below.
The query:
SELECT "t9"."c6_Department"
FROM
(
SELECT "t8"."c2_Department" AS "c6_Department"
FROM
(SELECT "t0"."Department" AS "c5_Department" FROM "HumanResources_1272" AS "t0" WHERE "t0"."Department" = 'Technology' GROUP BY "Department") AS "t7"
RIGHT OUTER JOIN
(SELECT "t0"."Department" AS "c2_Department" FROM "HumanResources_1272" AS "t0" GROUP BY "t0"."Department") AS "t8"
ON (
"t8"."c2_Department" = "t7"."c5_Department"OR ( "t7"."c5_Department" IS NULL AND "t8"."c2_Department" IS NULL )
)
) AS "t9"
ORDER BY "c6_Department" ASC
returns
c6_Department
---------------
<null>
Technology
which is wrong, since no restriction on the values of Department is present on the right table.
The equivalent query:
SELECT "t9"."c6_Department"
FROM
(
SELECT "t8"."c2_Department" AS "c6_Department"
FROM
(SELECT "t0"."Department" AS "c5_Department" FROM "HumanResources_1272" AS "t0" WHERE "t0"."Department" = 'Technology' GROUP BY "Department") AS "t7"
RIGHT OUTER JOIN
(SELECT "t0"."Department" AS "c2_Department" FROM "HumanResources_1272" AS "t0" GROUP BY "t0"."Department") AS "t8"
ON (
"t8"."c2_Department" = "t7"."c5_Department"OR ( "t8"."c2_Department" IS NULL AND "t7"."c5_Department" IS NULL )
)
) AS "t9"
ORDER BY "c6_Department" ASC
returns
c6_Department
---------------
<null>
Finance
Marketing
Production
Sales
Technology
which is the correct result.
Similar results can be obtained with a left outer join.
> Attached file: [HumanResources_1272.sql](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_6581_HumanResources_1272.sql_601) (text/plain, 679579 bytes)
> Description: The sql to create the table used to reproduce the bug
## Comment 26398
Date: 2018-04-21 11:10:45 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [68e680001a8f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=68e680001a8f) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=68e680001a8f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=68e680001a8f)
Changeset description:
fixes for bug #6581
| Join condition errors. | https://api.github.com/repos/MonetDB/MonetDB/issues/6581/comments | 0 | 2020-11-30T16:27:08Z | 2024-06-27T13:03:57Z | https://github.com/MonetDB/MonetDB/issues/6581 | 753,620,265 | 6,581 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-17 16:10:12 +0200
From: Manuel <<manuel>>
To: SQL devs <<bugs-sql>>
Version: 11.27.11 (Jul2017-SP3)
Last updated: 2018-04-18 15:01:45 +0200
## Comment 26374
Date: 2018-04-17 16:10:12 +0200
From: Manuel <<manuel>>
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
I had a few issues with double values in monetdb.
In order to reproduce the issue I have created a table called "testDoubleP". The table contains a single double column, and a single record with the value 0.9826432701399579.
The value was inserted using the MAPI interface:
COPY 1 RECORDS INTO "testDoubleP" FROM STDIN USING DELIMITERS ',','\n','"';
0.9826432701399579
It is important to use the Mapi Library for this step. I tried using usual an INSERT statement, but I got different results.
The query select * testDoubleP returns the value 0.9826432701399579, which is expected.
However, the query
select "x" from "testDoubleP" where "x" = 0.9826432701399579;
returns no records. This is an issue as the value used in the where condition is retrieved from the database itself, hence
I would expect the query to find the record.
The queries:
select "x" from "testDoubleP" where "x" = 0.98264327013995790;
select "x" from "testDoubleP" where "x" = cast('0.9826432701399579' as double);
both find the record successfully.
Reproducible: Always
### Steps to Reproduce:
1. Create a table with a single double column
2. populate the table using the Mapi library with the value 0.9826432701399579
### Actual Results:
The query select "x" from "testDoubleP" where "x" = 0.9826432701399579; will return no records
### Expected Results:
The record to be found
I tried to insert the record using mclient, and SQL statements
insert into "doubleP" values(0.9826432701399579);
If I select all records in the table (select * from "doubleP") I get back the value 0.982643270139958 (which is the one the SQL insert statement).
In this case, the record can be found by either:
select "x" from "testDoubleP" where "x" = 0.982643270139958;
and
select "x" from "testDoubleP" where "x" = 0.9826432701399579;
## Comment 26377
Date: 2018-04-18 15:01:45 +0200
From: @sjoerdmullender
This is not a bug.
The problem is, when you write 0.9826432701399579 in SQL, it is a value of type DECIMAL not of type DOUBLE. A DECIMAL is stored as an appropriately sized integer. In this case as the value 9826432701399579 with a scale of -16 (i.e. multiply by 10 to the power -16 or 1e-16).
When this value is converted to DOUBLE, the calculation is basically, convert the integer 9826432701399579 to double, then divide by 1e16. The value 9826432701399579 cannot be exactly represented in a double (a double can precisely represent integers of up to 53 bits, this integer needs 54 bits). This means that instead of 9826432701399579, you get the closest value that *can* be represented, which is 9826432701399580.0. This value is then divided by 1e16, yielding 0.982643270139958.
The solution is to not go through the DECIMAL type. You gave one possibility (cast('0.9826432701399579' as double)), another one is to use scientific notation: 0.9826432701399579e0.
Try out these:
select cast(9826432701399579 as double);
select cast(9826432701399579 as double) / 10000000000000000;
select 0.9826432701399579;
select 0.9826432701399579e0;
The difference with the COPY INTO is that in that case the value is not converted to a DECIMAL first but immediately converted from the string value in the input to DOUBLE.
| Double columns rounding errors | https://api.github.com/repos/MonetDB/MonetDB/issues/6580/comments | 0 | 2020-11-30T16:27:04Z | 2024-06-28T06:47:23Z | https://github.com/MonetDB/MonetDB/issues/6580 | 753,620,210 | 6,580 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-14 21:43:37 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @kutsurak, @njnes
Last updated: 2019-05-06 11:22:26 +0200
## Comment 26368
Date: 2018-04-14 21:43:37 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
On default branch, run the query bellow for an infinite loop in exp_bin function call.
Reproducible: Always
### Steps to Reproduce:
1. SELECT ALL - 70 + + - 89 - + ( SUM ( + CASE - NULLIF ( + 87, CASE CAST ( NULL AS INTEGER ) WHEN 13 / + ( 68 * - ( 64 ) )
THEN + 39 WHEN 70 * - 95 THEN 60 ELSE 15 + + 25 END * 5 ) WHEN + 36 THEN NULL WHEN 24 THEN NULL ELSE 66 END ) ) AS col2;
### Actual Results:
Infinite loop in exp_bin call in rel_bin.c, leaving infinite time to compile the query.
### Expected Results:
-225
Backtrace:
0 findVariable (mb=0x7f72680518f0, name=0x7f727e59ea00 "AL14034") at mal_instruction.c:593
1 0x00007f72af1076e8 in pushArgumentId (mb=0x7f72680518f0, p=0x7f723b8463e0, name=0x7f727e59ea00 "AL14034") at mal_instruction.c:1352
2 0x00007f729b8636f1 in stmt_var (be=0x7f7268002b60, varname=0x7f727e59ed49 "L14034", t=0x7f7268044c80, declare=1, level=2) at sql_statement.c:327
3 0x00007f729b84297f in exp_bin (be=0x7f7268002b60, e=0x7f7268032a00, left=0x0, right=0x0, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:521
4 0x00007f729b8423e4 in exp_bin (be=0x7f7268002b60, e=0x7f7268032a90, left=0x0, right=0x0, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:450
5 0x00007f729b842612 in exp_bin (be=0x7f7268002b60, e=0x7f7268032c10, left=0x0, right=0x0, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:480
6 0x00007f729b842612 in exp_bin (be=0x7f7268002b60, e=0x7f7268032ca0, left=0x0, right=0x0, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:480
7 0x00007f729b842612 in exp_bin (be=0x7f7268002b60, e=0x7f7268033200, left=0x0, right=0x0, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:480
8 0x00007f729b842612 in exp_bin (be=0x7f7268002b60, e=0x7f72680334e0, left=0x0, right=0x0, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:480
9 0x00007f729b842612 in exp_bin (be=0x7f7268002b60, e=0x7f7268033aa0, left=0x0, right=0x0, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:480
10 0x00007f729b8423e4 in exp_bin (be=0x7f7268002b60, e=0x7f7268033b30, left=0x0, right=0x0, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:450
.....
## Comment 26370
Date: 2018-04-14 22:14:30 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [ff8bda80d298](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ff8bda80d298) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ff8bda80d298](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ff8bda80d298)
Changeset description:
Added test for Bug #6579
## Comment 26386
Date: 2018-04-20 10:07:46 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [629384f37bb0](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=629384f37bb0) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=629384f37bb0](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=629384f37bb0)
Changeset description:
fixes for bug #6579, ie split aggr/function into a lower
project earlier.
## Comment 26998
Date: 2019-05-06 11:22:26 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [49fdc1899528](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=49fdc1899528) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=49fdc1899528](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=49fdc1899528)
Changeset description:
Adding more sqlite tests failing queries:
Bug #6605 outputs only 3 columns out of 6 on default.
Bug #6566 crashes on both Apr2019 and default.
Bug #6579 crashes on default.
| Sqlitelogic test infinite loop while compiling SQL query | https://api.github.com/repos/MonetDB/MonetDB/issues/6579/comments | 0 | 2020-11-30T16:27:01Z | 2024-06-27T13:03:55Z | https://github.com/MonetDB/MonetDB/issues/6579 | 753,620,170 | 6,579 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-13 14:24:32 +0200
From: @swingbit
To: SQL devs <<bugs-sql>>
Version: 11.27.13 (Jul2017-SP4)
CC: @kutsurak, @njnes
Last updated: 2018-06-20 11:02:11 +0200
## Comment 26365
Date: 2018-04-13 14:24:32 +0200
From: @swingbit
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
This happens in both Jul2017 and Mar2018
Starting with these two tables, x and y:
START TRANSACTION;
CREATE TABLE "x" (
"a1" CHARACTER LARGE OBJECT,
"a2" CHARACTER LARGE OBJECT
);
COPY 5 RECORDS INTO "x" FROM stdin USING DELIMITERS '|','\n','"';
"fiets"|"damesfiets"
"fiets"|"herenfiets"
"auto"|"personenwag"
"auto"|"bedrijfsauto"
"auto"|"personenauto"
COMMIT;
START TRANSACTION;
CREATE TABLE "y" (
"paramname" CHARACTER LARGE OBJECT,
"value" CHARACTER LARGE OBJECT
);
COPY 1 RECORDS INTO "y" FROM stdin USING DELIMITERS '|','\n','"';
"1"|"something"
COMMIT;
** The following gives back 3 results:
start transaction;
insert into y values ('0','boot');
insert into y values ('0','auto');
select count(*)
from x,
(select value as a1 from y where paramname='0') as v
where x.a1 = v.a1;
rollback;
** However, the following (where the two insert statements have been merged into one statement), gives 0 results:
start transaction;
insert into y values ('0','boot'), ('0','auto');
select count(*)
from x,
(select value as a1 from y where paramname='0') as v
where x.a1 = v.a1;
rollback;
Notice that joining directly x and y behaves as expected. Only with the subselect into v, it fails.
Reproducible: Always
## Comment 26369
Date: 2018-04-14 22:07:33 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [fbc3ebb69aab](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=fbc3ebb69aab) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=fbc3ebb69aab](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=fbc3ebb69aab)
Changeset description:
Added test for bug #6578
## Comment 26371
Date: 2018-04-16 08:25:49 +0200
From: @kutsurak
As far as I can tell the problem appears in the subquery, but only if autocommit is off:
sql>start transaction;
auto commit mode: off
sql>insert into y values ('0','boot'), ('0','auto');
2 affected rows
sql>select * from y where paramname='0';
+-----------+-----------+
| paramname | value |
+===========+===========+
| 1 | something |
| 0 | boot |
+-----------+-----------+
2 tuples
sql>rollback;
auto commit mode: on
sql>insert into y values ('0','boot'), ('0','auto');
2 affected rows
sql>select * from y;
+-----------+-----------+
| paramname | value |
+===========+===========+
| 1 | something |
| 0 | boot |
| 0 | auto |
+-----------+-----------+
3 tuples
sql>select * from y where paramname='0';
+-----------+-------+
| paramname | value |
+===========+=======+
| 0 | boot |
| 0 | auto |
+-----------+-------+
2 tuples
I tested this in Mar2018.
## Comment 26373
Date: 2018-04-16 13:15:48 +0200
From: @njnes
fixed by keeping the head seqbase stable under the first (and later) appends
| One two-tuple insert gives different results than two single inserts | https://api.github.com/repos/MonetDB/MonetDB/issues/6578/comments | 0 | 2020-11-30T16:26:58Z | 2024-06-27T13:03:54Z | https://github.com/MonetDB/MonetDB/issues/6578 | 753,620,114 | 6,578 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-13 00:58:37 +0200
From: Anton Kravchenko <<kravchenko.anton86>>
To: SQL devs <<bugs-sql>>
Version: 11.29.3 (Mar2018)
CC: @njnes
Last updated: 2018-06-20 11:02:14 +0200
## Comment 26364
Date: 2018-04-13 00:58:37 +0200
From: Anton Kravchenko <<kravchenko.anton86>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
creating temp table constrains the original query to use a single CPU
Reproducible: Always
### Steps to Reproduce:
1. run time=4.6s, 4 CPUs are used
select count(t1.v1) from t1, t2 where t1.v1=t2.v1;
2. run time=15s, 1 CPU is used
create temp table t3 as
(select count(t1.v1) from t1, t2 where t1.v1=t2.v1)
on commit preserve rows;
3. run time=15s, 1 CPU is used
create temp table t3 as
(select t1.v1 from t1, t2 where t1.v1=t2.v1)
on commit preserve rows;
Tables t1, t2 were created from:
create or replace function test1() RETURNS TABLE (v0 INT, v1 INT)
LANGUAGE PYTHON {
import numpy as np
nrows=100000000
i_var=np.random.randint(1,10000000, (100000000,))
return [[i for i in range(nrows)],i_var]
};
create table t1 as (select * from test1()) with data;
create table t2 as (select distinct v1 from t1) with data;
### Actual Results:
run time=15s, 1 CPU is used
create temp table t3 as
(select t1.v1 from t1, t2 where t1.v1=t2.v1)
on commit preserve rows;
### Expected Results:
run time=4.6s, 4 CPUs are used
create temp table t3 as
(select t1.v1 from t1, t2 where t1.v1=t2.v1)
on commit preserve rows;
MonetDB 5 server v11.29.3 "Mar2018" (64-bit, 128-bit integers)
Copyright (c) 1993 - July 2008 CWI
Copyright (c) August 2008 - 2018 MonetDB B.V., all rights reserved
Visit https://www.monetdb.org/ for further information
Found 31.3GiB available memory, 4 available cpu cores
Libraries:
libpcre: 8.32 2012-11-30 (compiled with 8.32)
openssl: OpenSSL 1.0.2k 26 Jan 2017 (compiled with OpenSSL 1.0.2l 25 May 2017)
libxml2: 2.9.1 (compiled with 2.9.1)
Compiled by: akravchenko@cent7-1 (x86_64-pc-linux-gnu)
Compilation: gcc -std=gnu99 -O3 -fomit-frame-pointer -pipe -D_FORTIFY_SOURCE=2
Linking : /usr/bin/ld -m elf_x86_64 -Wl,-Bsymbolic-functions
## Comment 26414
Date: 2018-04-29 11:39:19 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [dede692aec19](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=dede692aec19) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=dede692aec19](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=dede692aec19)
Changeset description:
Added tests for bugs 6577 and 6594
## Comment 26436
Date: 2018-05-11 10:18:53 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [f3fcfcad3fbc](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=f3fcfcad3fbc) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=f3fcfcad3fbc](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=f3fcfcad3fbc)
Changeset description:
fixes for bug #6577 and 6600
## Comment 26440
Date: 2018-05-11 13:15:26 +0200
From: @njnes
fixed, ie also update queries are now using mitosis
| creating temp table kills performance of the original query | https://api.github.com/repos/MonetDB/MonetDB/issues/6577/comments | 0 | 2020-11-30T16:26:54Z | 2024-06-27T13:03:53Z | https://github.com/MonetDB/MonetDB/issues/6577 | 753,620,053 | 6,577 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-12 10:12:08 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: 11.29.3 (Mar2018)
CC: @njnes
Last updated: 2019-09-02 16:05:27 +0200
## Comment 26361
Date: 2018-04-12 10:12:08 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
Running several simple SELECT queries with math expressions involving negatives numbers causes a MAL exception to be thrown.
Reproducible: Always
### Steps to Reproduce:
At the end of each line after the comment the expected result is shown.
SELECT -2 * 50; -- -100
SELECT ALL - 0 * 36; -- 0
SELECT ALL - 33 + - 77 * 0 AS col1; -- -33
SELECT + 12 + - 0 * + 70; -- 12
SELECT - ( ( 0 ) ) * 78; -- 0
SELECT ALL - 4 * + 0 * - - 48 col1; -- 0
SELECT - 0 * - 15 AS col0; -- 0
SELECT - 48 * + + 0 + + 94 AS col1; -- 94
SELECT ALL + - 32 * - 32 + - - 76 - - - 43 * COUNT ( * ) + - + 84 * 0 * + + 39 + + 44 * + 76; -- 4401
SELECT 38 + - 63 * 0 * - 13 AS col2; -- 38
SELECT 64 + - 0 * 40 AS col1; -- 64
SELECT ALL - 35 * - 0 AS col2; -- 0
SELECT + - 36 * - + 33 * - + 49 * + CASE 50 WHEN - 21 + 0 + - COUNT ( * ) THEN NULL
WHEN + 99 * - 43 + - - 35 THEN - 75 ELSE NULL END * - - 51 * + 0 AS col2; -- NULL
SELECT - 66 * 0 * - 87 AS col2; -- 0
SELECT - 13 * - ( + 0 ) + 39 * - COUNT ( * ) * - COUNT ( * ) col1; -- 39
SELECT COUNT ( - + 70 ) - - 99 * - 0 AS col2; -- 1
SELECT - ( - - 15 ) + - 53 * - 0; -- -15
SELECT - + 22 * + COUNT ( * ) + - - 58 - + 83 * + MAX ( - 94 * 0 ); -- 36
SELECT DISTINCT - 16 + ( + 37 ) - - 56 * - - 11 * - 0 + + MAX ( + 29 ) AS col2; -- 50
SELECT - 54 * + 0; -- 0
SELECT ALL - 30 + + 85, - ( - + 43 ) - 71 + - 0 * - - 78 + - 78 + - - 47 * - 93 + 32 * + SUM ( 90 ) AS col0; -- 55, -1597
SELECT ALL + COUNT ( * ) - - + 0 * - 32; -- 1
### Actual Results:
MAL exception:
TypeException:user.main[6]:'calc.-' undefined in: X_162759:sht := calc.-(X_162758:bte);
### Expected Results:
Check the steps above.
## Comment 26362
Date: 2018-04-12 10:12:53 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [780897b729df](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=780897b729df) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=780897b729df](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=780897b729df)
Changeset description:
Added tests for bugs 6574 and 6576
## Comment 26366
Date: 2018-04-14 16:02:58 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [efe5b9e9ef51](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=efe5b9e9ef51) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=efe5b9e9ef51](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=efe5b9e9ef51)
Changeset description:
fixes for bug #6576
## Comment 26991
Date: 2019-05-02 14:31:58 +0200
From: @PedroTadim
Updated test with failing query.
## Comment 26995
Date: 2019-05-03 10:37:52 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [2ecd89950f47](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=2ecd89950f47) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=2ecd89950f47](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=2ecd89950f47)
Changeset description:
Adding more sqlite tests failing queries:
Bug #6560 gives right results in Apr2019 branch, but wrong on default.
Bug #6561 we should analyze first, but I think MonetDB outputs wrong results.
Bug #6576 more overflows in calculations.
## Comment 27001
Date: 2019-05-06 11:25:56 +0200
From: @PedroTadim
The overflows were fixed on changeset [e8c60cd2b678](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e8c60cd2b678)
| Sqlitelogictest aritmetic expressions with negative numbers handling | https://api.github.com/repos/MonetDB/MonetDB/issues/6576/comments | 0 | 2020-11-30T16:26:50Z | 2024-06-27T13:03:52Z | https://github.com/MonetDB/MonetDB/issues/6576 | 753,620,013 | 6,576 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-11 17:57:27 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @kutsurak, @njnes
Last updated: 2018-08-31 13:23:13 +0200
## Comment 26359
Date: 2018-04-11 17:57:27 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
On default branch, run the query bellow for an assertion error. My guess is that the MAL generation is not correct, thus generating a function call with a bulk implementation.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab1(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab1 VALUES(22,6,8), (28,57,45), (82,44,71);
3. SELECT ALL - 52 * + COALESCE ( + col2, + 47 / + 25 ) FROM tab1 AS cor0 GROUP BY cor0.col2, cor0.col2;
### Actual Results:
Assertion error in mserver5: mal_scenario.c:552: runScenarioBody: Assertion `c->curprg->def->errors == NULL' failed.
### Expected Results:
A single row with results -2340, -3692, -416
Log:
WARNING To speedup calc./ a bulk operator implementation is needed
X_290:bat[:bte] := mal.multiplex("calc":str, "/":str, 47:bte, 25:bte, X_288:bat[:oid]);
mserver5: mal_scenario.c:552: runScenarioBody: Assertion `c->curprg->def->errors == NULL' failed.
Backtrace:
0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:51
1 0x00007ffff4781c41 in __GI_abort () at abort.c:79
2 0x00007ffff4778f7a in __assert_fail_base (fmt=0x7ffff48c9260 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", assertion=assertion@entry=0x7ffff7b644e8 "c->curprg->def->errors == NULL",
file=file@entry=0x7ffff7b644d5 "mal_scenario.c", line=line@entry=552, function=function@entry=0x7ffff7b64560 <__PRETTY_FUNCTION__.11456> "runScenarioBody") at assert.c:92
3 0x00007ffff4778ff2 in __GI___assert_fail (assertion=0x7ffff7b644e8 "c->curprg->def->errors == NULL", file=0x7ffff7b644d5 "mal_scenario.c", line=552,
function=0x7ffff7b64560 <__PRETTY_FUNCTION__.11456> "runScenarioBody") at assert.c:101
4 0x00007ffff7a0dcc2 in runScenarioBody (c=0x7fffe973c368, once=0) at mal_scenario.c:552
5 0x00007ffff7a0ddad in runScenario (c=0x7fffe973c368, once=0) at mal_scenario.c:569
6 0x00007ffff7a0fcdf in MSserveClient (dummy=0x7fffe973c368) at mal_session.c:519
7 0x00007ffff7a0f7d4 in MSscheduleClient (command=0x7fffb0000b30 "0", challenge=0x7fffc72e1d9b "wd3NDp6N", fin=0x181ab40, fout=0x7fffc0002d90, protocol=PROTOCOL_9, blocksize=8190, compute_column_widths=0)
at mal_session.c:397
8 0x00007ffff7ab150e in doChallenge (data=0x181ab50) at mal_mapi.c:279
9 0x00007ffff7697240 in thread_starter (arg=0x9c8710) at gdk_system.c:476
10 0x00007ffff4b0950b in start_thread (arg=0x7fffc72e2700) at pthread_create.c:465
11 0x00007ffff484116f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95
## Comment 26360
Date: 2018-04-11 17:58:39 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [9f73e7e8a96b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9f73e7e8a96b) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=9f73e7e8a96b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=9f73e7e8a96b)
Changeset description:
Added test for bug #6575
## Comment 26367
Date: 2018-04-14 18:56:06 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [0e4c696bd1fe](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=0e4c696bd1fe) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=0e4c696bd1fe](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=0e4c696bd1fe)
Changeset description:
fixes for bug #6575
| Sqlitelogictest crash on groupby query with coalesce call | https://api.github.com/repos/MonetDB/MonetDB/issues/6575/comments | 0 | 2020-11-30T16:26:48Z | 2024-06-27T13:03:51Z | https://github.com/MonetDB/MonetDB/issues/6575 | 753,619,975 | 6,575 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-11 00:53:41 +0200
From: Anton Kravchenko <<kravchenko.anton86>>
To: SQL devs <<bugs-sql>>
Version: 11.29.3 (Mar2018)
CC: @njnes
Last updated: 2018-06-20 11:02:08 +0200
## Comment 26355
Date: 2018-04-11 00:53:41 +0200
From: Anton Kravchenko <<kravchenko.anton86>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
crash when creating a table from query with ordered index
Reproducible: Always
### Steps to Reproduce:
1)
create table t1(v0 int, v1 char(1));
insert into t1 values (1,'a'),(2,'b'),(3,'c');
create ordered index index_t1_v1 on t1(v1);
2)
create table t2(v1 char(1));
insert into t2 values ('a');
3)
create temp table t3 as
(select t1.v0 from t1
where
exists(select * from t2 where t2.v1=t1.v1))
with data on commit preserve rows;
### Actual Results:
2018-04-10 15:38:49 ERR db1[31908]: could not find t1.%index_t1_v1
2018-04-10 15:38:49 ERR db1[31908]: t1.v0
2018-04-10 15:38:49 ERR db1[31908]: t1.v1
2018-04-10 15:38:49 ERR db1[31908]: t1.%TID%
2018-04-10 15:38:49 ERR db1[31908]: t1.v0
2018-04-10 15:38:49 ERR db1[31908]: t1.v1
2018-04-10 15:38:49 ERR db1[31908]: t1.%TID%
2018-04-10 15:38:49 ERR db1[31908]: mserver5: /home/akravchenko/MonetDB-Mar2018/MonetDB-11.29.3/sql/backends/monet5/rel_bin.c:576: exp_bin: Assertion `s' failed.
2018-04-10 15:38:52 MSG merovingian[26313]: database 'db1' (31908) has crashed (dumped core)
### Expected Results:
It works as expected when no table is created:
select t1.v0 from t1
where
exists(select * from t2 where t2.v1=t1.v1);
MonetDB 5 server v11.29.3 "Mar2018" (64-bit, 128-bit integers)
Copyright (c) 1993 - July 2008 CWI
Copyright (c) August 2008 - 2018 MonetDB B.V., all rights reserved
Visit https://www.monetdb.org/ for further information
Found 31.3GiB available memory, 4 available cpu cores
Libraries:
libpcre: 8.32 2012-11-30 (compiled with 8.32)
openssl: OpenSSL 1.0.2k 26 Jan 2017 (compiled with OpenSSL 1.0.2l 25 May 2017)
libxml2: 2.9.1 (compiled with 2.9.1)
Compiled by: akravchenko@cent7-1 (x86_64-pc-linux-gnu)
Compilation: gcc -std=gnu99 -g
Linking : /usr/bin/ld -m elf_x86_64 -Wl,-Bsymbolic-functions
## Comment 26358
Date: 2018-04-11 16:20:15 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [885e050fdec6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=885e050fdec6) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=885e050fdec6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=885e050fdec6)
Changeset description:
fixes for bugs 6572, 6573 and 6574
also fix select -2*50 cast problem, fix type of function in rewriter.
6572/6574 fixed in dead code optimizer
6573 fixed problem in 'IN' handling with constants
## Comment 26363
Date: 2018-04-12 10:12:55 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [780897b729df](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=780897b729df) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=780897b729df](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=780897b729df)
Changeset description:
Added tests for bugs 6574 and 6576
| server crashed could not find ordered index while creating a table | https://api.github.com/repos/MonetDB/MonetDB/issues/6574/comments | 0 | 2020-11-30T16:26:44Z | 2024-06-27T13:03:50Z | https://github.com/MonetDB/MonetDB/issues/6574 | 753,619,921 | 6,574 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-09 10:45:04 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-06-20 11:02:16 +0200
## Comment 26352
Date: 2018-04-09 10:45:04 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
Run the query bellow for an assertion error.
Reproducible: Always
### Steps to Reproduce:
1. SELECT - ( 5 ) AS col1, + CASE WHEN NOT - 43 NOT IN ( + + 33 * 65, - 31 ) THEN NULL WHEN ( + ( - 41 ) ) IN ( 73 * CASE + + 33 WHEN - 25 THEN - CAST ( NULL AS INTEGER ) ELSE NULL END ) THEN 64 WHEN NOT ( + CAST ( + CASE + 52 WHEN COUNT ( * ) THEN NULL WHEN 51 THEN - 68 ELSE NULL END AS INTEGER ) ) = 50 THEN NULL END * 11 + NULLIF ( ( - 3 ) + 51, 32 ) AS col1;
### Actual Results:
Assertion error in mserver5: rel_bin.c:576: exp_bin: Assertion `s' failed.
### Expected Results:
A single result set row with values -5, NULL
Backtrace:
0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:51
1 0x00007ffff4782c41 in __GI_abort () at abort.c:79
2 0x00007ffff4779f7a in __assert_fail_base (fmt=0x7ffff48ca260 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", assertion=assertion@entry=0x7fffe84fe738 "s", file=file@entry=0x7fffe84fe6c8 "rel_bin.c",
line=line@entry=576, function=function@entry=0x7fffe84fef08 <__PRETTY_FUNCTION__.18603> "exp_bin") at assert.c:92
3 0x00007ffff4779ff2 in __GI___assert_fail (assertion=0x7fffe84fe738 "s", file=0x7fffe84fe6c8 "rel_bin.c", line=576, function=0x7fffe84fef08 <__PRETTY_FUNCTION__.18603> "exp_bin") at assert.c:101
4 0x00007fffe8384d0e in exp_bin (be=0x7fffb0002b60, e=0x7fffb00f0c30, left=0x7fffb010c470, right=0x7fffb010c660, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:576
5 0x00007fffe83841a2 in exp_bin (be=0x7fffb0002b60, e=0x7fffb00fb480, left=0x7fffb010c470, right=0x7fffb010c660, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:450
6 0x00007fffe838e003 in rel2bin_project (be=0x7fffb0002b60, rel=0x7fffb00f47f0, refs=0x7fffb010be40, topn=0x0) at rel_bin.c:2481
7 0x00007fffe8399bed in subrel_bin (be=0x7fffb0002b60, rel=0x7fffb00f47f0, refs=0x7fffb010be40) at rel_bin.c:5068
8 0x00007fffe838df30 in rel2bin_project (be=0x7fffb0002b60, rel=0x7fffb00fb030, refs=0x7fffb010be40, topn=0x0) at rel_bin.c:2469
9 0x00007fffe8399bed in subrel_bin (be=0x7fffb0002b60, rel=0x7fffb00fb030, refs=0x7fffb010be40) at rel_bin.c:5068
10 0x00007fffe8399ed5 in _subrel_bin (be=0x7fffb0002b60, rel=0x7fffb00fb030, refs=0x7fffb010be40) at rel_bin.c:5130
11 0x00007fffe8399fed in output_rel_bin (be=0x7fffb0002b60, rel=0x7fffb00fb030) at rel_bin.c:5153
12 0x00007fffe83b6384 in sql_relation2stmt (be=0x7fffb0002b60, r=0x7fffb00fb030) at sql_gencode.c:542
13 0x00007fffe83b6577 in backend_dumpstmt (be=0x7fffb0002b60, mb=0x7fffb0125ac0, r=0x7fffb00fb030, top=1, add_end=1,
query=0x7fffb010fa80 "select - ( 5 ) as col1, + case when not - 43 not in ( + + 33 * 65, - 31 ) then null when ( + ( - 41 ) ) in ( 73 * case + + 33 when - 25 then - cast ( null as integer ) else null end ) then 64 when not"...) at sql_gencode.c:582
14 0x00007fffe83b6def in backend_dumpproc (be=0x7fffb0002b60, c=0x7fffe973d368, cq=0x7fffb01158a0, r=0x7fffb00fb030) at sql_gencode.c:712
15 0x00007fffe8378978 in SQLparser (c=0x7fffe973d368) at sql_scenario.c:1233
16 0x00007ffff7a0f378 in runPhase (c=0x7fffe973d368, phase=1) at mal_scenario.c:510
17 0x00007ffff7a0f481 in runScenarioBody (c=0x7fffe973d368, once=0) at mal_scenario.c:532
18 0x00007ffff7a0f73a in runScenario (c=0x7fffe973d368, once=0) at mal_scenario.c:569
19 0x00007ffff7a11666 in MSserveClient (dummy=0x7fffe973d368) at mal_session.c:519
20 0x00007ffff7a11161 in MSscheduleClient (command=0x7fffb0000b30 "\360\030", challenge=0x7fffc72e1d9b "8HgaRnl3at5", fin=0x17ff660, fout=0x7fffc0002d90, protocol=PROTOCOL_9, blocksize=8190,
compute_column_widths=0) at mal_session.c:397
21 0x00007ffff7ab22c8 in doChallenge (data=0x17ff670) at mal_mapi.c:279
22 0x00007ffff769936e in thread_starter (arg=0x9c4bc0) at gdk_system.c:476
23 0x00007ffff4b0a50b in start_thread (arg=0x7fffc72e2700) at pthread_create.c:465
24 0x00007ffff484216f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95
## Comment 26354
Date: 2018-04-09 10:46:30 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [799859bfffd7](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=799859bfffd7) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=799859bfffd7](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=799859bfffd7)
Changeset description:
Added tests for bugs 6572 and 6573
## Comment 26357
Date: 2018-04-11 16:20:14 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [885e050fdec6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=885e050fdec6) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=885e050fdec6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=885e050fdec6)
Changeset description:
fixes for bugs 6572, 6573 and 6574
also fix select -2*50 cast problem, fix type of function in rewriter.
6572/6574 fixed in dead code optimizer
6573 fixed problem in 'IN' handling with constants
| Sqlitelogictest crash on complex select query with case statement | https://api.github.com/repos/MonetDB/MonetDB/issues/6573/comments | 0 | 2020-11-30T16:26:41Z | 2024-06-27T13:03:49Z | https://github.com/MonetDB/MonetDB/issues/6573 | 753,619,894 | 6,573 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-07 00:08:50 +0200
From: Anton Kravchenko <<kravchenko.anton86>>
To: SQL devs <<bugs-sql>>
Version: 11.29.3 (Mar2018)
CC: @njnes
Last updated: 2018-06-20 11:02:13 +0200
## Comment 26348
Date: 2018-04-07 00:08:50 +0200
From: Anton Kravchenko <<kravchenko.anton86>>
Here is a reproducible example.
create table t1(v0 int, v1 char(1));
insert into t1 values (1,'a'),(2,'b'),(3,'c');
create ordered index index_t1_v1 on t1(v1);
create table t2(v1 char(1));
insert into t2 values ('a');
create temp table t3 as
(select t1.v0 from t1,t2 where trim(t2.v1)=t1.v1)
on commit preserve rows;
1) --enable-assert=no
mclient gives "Error in optimizer garbageCollector"
merver5 gives "could not find t1.%index_t1_v1
t1.v0
t1.v1
t1.%TID%
t1.v0
t1.v1
t1.%TID%"
2) --enable-assert=yes
merver5 crashed "could not find t1.%index_t1_v1
t1.v0
t1.v1
t1.%TID%
t1.v0
t1.v1
t1.%TID%
mserver5: /home/akravchenko/MonetDB-Mar2018/MonetDB-11.29.3/sql/backends/monet5/rel_bin.c:576: exp_bin: Assertion `s' failed.
Aborted (core dumped)
I am using the latest MonetDB version:
MonetDB 5 server v11.29.3 "Mar2018" (64-bit, 128-bit integers)
Copyright (c) 1993 - July 2008 CWI
Copyright (c) August 2008 - 2018 MonetDB B.V., all rights reserved
Visit https://www.monetdb.org/ for further information
Found 31.3GiB available memory, 4 available cpu cores
Libraries:
libpcre: 8.32 2012-11-30 (compiled with 8.32)
openssl: OpenSSL 1.0.2k 26 Jan 2017 (compiled with )
libxml2: 2.9.1 (compiled with 2.9.1)
Compiled by: mockbuild@ (x86_64-redhat-linux-gnu)
Compilation: gcc -std=gnu99 -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches -m64 -mtune=generic
Linking : /usr/bin/ld -m elf_x86_64 -Wl,-z,relro -Wl,-Bsymbolic-functions
## Comment 26353
Date: 2018-04-09 10:46:29 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [799859bfffd7](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=799859bfffd7) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=799859bfffd7](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=799859bfffd7)
Changeset description:
Added tests for bugs 6572 and 6573
## Comment 26356
Date: 2018-04-11 16:20:11 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [885e050fdec6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=885e050fdec6) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=885e050fdec6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=885e050fdec6)
Changeset description:
fixes for bugs 6572, 6573 and 6574
also fix select -2*50 cast problem, fix type of function in rewriter.
6572/6574 fixed in dead code optimizer
6573 fixed problem in 'IN' handling with constants
| ordered index Error in optimizer garbageCollector | https://api.github.com/repos/MonetDB/MonetDB/issues/6572/comments | 0 | 2020-11-30T16:26:37Z | 2024-06-27T13:03:48Z | https://github.com/MonetDB/MonetDB/issues/6572 | 753,619,846 | 6,572 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-04 18:21:22 +0200
From: Manuel <<manuel>>
To: clients devs <<bugs-clients>>
Version: 11.25.21 (Dec2016-SP4)
CC: martin.van.dinther, @PedroTadim
Last updated: 2019-09-26 17:18:08 +0200
## Comment 26344
Date: 2018-04-04 18:21:22 +0200
From: Manuel <<manuel>>
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
I have a long sequence of queries, that when executed by a specific automation software causes monetdb jdbc driver to hang indefinitely.
I have access to the sequence of queries, but I have been unable to reproduce the issue on a more controllable environment: I still can reproduce the issue 9 times out of ten.
When this happens the thread submitting the query is stuck here:
at java.net.SocketInputStream.socketRead0(Native Method)
at java.net.SocketInputStream.socketRead(SocketInputStream.java:116)
at java.net.SocketInputStream.read(SocketInputStream.java:171)
at java.net.SocketInputStream.read(SocketInputStream.java:141)
at java.io.BufferedInputStream.fill(BufferedInputStream.java:246)
at java.io.BufferedInputStream.read1(BufferedInputStream.java:286)
at java.io.BufferedInputStream.read(BufferedInputStream.java:345)
at nl.cwi.monetdb.mcl.net.MapiSocket$BlockInputStream._read(Unknown Source)
at nl.cwi.monetdb.mcl.net.MapiSocket$BlockInputStream.readBlock(Unknown Source)
at nl.cwi.monetdb.mcl.net.MapiSocket$BlockInputStream.read(Unknown Source)
at sun.nio.cs.StreamDecoder.readBytes(StreamDecoder.java:284)
at sun.nio.cs.StreamDecoder.implRead(StreamDecoder.java:326)
at sun.nio.cs.StreamDecoder.read(StreamDecoder.java:178)
at java.io.InputStreamReader.read(InputStreamReader.java:184)
at java.io.BufferedReader.fill(BufferedReader.java:161)
at java.io.BufferedReader.readLine(BufferedReader.java:324)
at java.io.BufferedReader.readLine(BufferedReader.java:389)
at nl.cwi.monetdb.mcl.io.BufferedMCLReader.readLine(Unknown Source)
at nl.cwi.monetdb.mcl.io.BufferedMCLReader.waitForPrompt(Unknown Source)
at nl.cwi.monetdb.jdbc.MonetConnection$ResponseList.executeQuery(Unknown Source)
at nl.cwi.monetdb.jdbc.MonetConnection$ResponseList.processQuery(Unknown Source)
at nl.cwi.monetdb.jdbc.MonetStatement.internalExecute(Unknown Source)
at nl.cwi.monetdb.jdbc.MonetStatement.execute(Unknown Source)
at nl.cwi.monetdb.jdbc.MonetStatement.executeQuery(Unknown Source)
While this is happening, it is possible to submit other queries just fine:
If I execute the query
select * from "sys"."queue"
while the tread is stuck, the query submitted by the hanging thread is not in the result set: It seems that the thread is waiting for a query that has already finished, or it will never be executed. Not 100% certain it is a jdbc issue.
I am aware of https://www.monetdb.org/bugzilla/show_bug.cgi?id=3973, but that seems more a workaround for the jdbc connection to get stuck, whereas I will be more interested into avoiding it in the first place.
The DBMS is running on the same host the client is running.
I can submit the entire sequence of queries, as well as the datasets, but I was not able to reproduce the issue that way.
Is this a known issue? Is there anything I can try to gather more information about what is happening ?
Reproducible: Sometimes
### Actual Results:
JDBC connection hanging
## Comment 26346
Date: 2018-04-05 18:23:33 +0200
From: Martin van Dinther <<martin.van.dinther>>
Which JDBC driver version do you use? The latest version is monetdb-jdbc-2.27.jar
Is the "sequence of queries" sent concurrently (multiple java execution threads running in parallel) or using one thread (running queries sequentially) by the specific automation software?
If multiple threads, try running it with one thread to determine if it is related to this.
Also if multiple threads are used, can you try to make a separate connection for each thread and verify if that resolves the problem (as each connection will use a separate Socket). This might help in identifying the root cause.
It will be difficult for us to analyse (and fix) a problem if we have no reproduction script/program.
Can you try to create a standalone java program which reproduces the problem?
FYI: the MonetDB JDBC driver sources can be viewed from https://dev.monetdb.org/hg/monetdb-java/file/tip/src/main/java/nl/cwi/monetdb
The mcl subdir contains the java code dealing with network IO.
Aside: The Version field lists Dec2016-SP4. Is that correct? If so, you should consider upgrading MonetDB to a newer version (Mar2018) as much has been changed and improved since Dec2016-SP4.
## Comment 26347
Date: 2018-04-05 18:55:49 +0200
From: Manuel <<manuel>>
Hi Martin,
I am using the latest version of the monetdb driver: monetdb-jdbc-2.27.jar.
The problem occurs when running on a single thread, or by submitting multiple queries at the same time. I also tried opening a new connection every time a new query is submitted.
On the monetdb side, I tried setting the parameters gdk_nr_threads=1 and sql_optimizer=sequential_pipe, but it did not help.
Yes, I am currently using Dec2016-SP4. I am aware the new version is available: the issue being that upgrading monetdb might require a non trivial amount of work on our side, and I by checking at the releases notes I haven't spotted anything useful for our use cases. Still that might be worth a try if we are not able to understand what is going on.
I understand it is hard for you to analyse the problem. I was wondering if there is a way for me to take a "thread dump" of monetdb to see what is happening, or enable some logging that would help us understand what happened to the query being submitted.
I am currently to find a way to reproduce the issue easily, and on any machine.
## Comment 27168
Date: 2019-07-25 16:08:38 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset 637899bda602, made by Martin van Dinther <martin.van.dinther@monetdbsolutions.com> in the monetdb-java repo, refers to this bug.
For complete details, see https://dev.monetdb.org/hg/monetdb-java?cmd=changeset;node=637899bda602
Changeset description:
Removed SendThread class and its usage from MonetConnection.
It appeared the usage of a SendThread object (per connection) resulted sometimes to a non-responding/hanging JDBC connection, see bugs 6571 and 6693.
Tests with large query scripts and large results returned show that the SendThread class is not needed in the JDBC driver to function properly.
Also submitter of bug #6693 verified that the removal of SendThread resolved the hanging JDBC-driver problem without any side effects.
## Comment 27326
Date: 2019-09-26 17:18:08 +0200
From: Martin van Dinther <<martin.van.dinther>>
This is fixed in the new: monetdb-jdbc-2.29.jre7.jar release.
It is downloadable from https://www.monetdb.org/downloads/Java/
| Jdbc connection hangs | https://api.github.com/repos/MonetDB/MonetDB/issues/6571/comments | 0 | 2020-11-30T16:26:35Z | 2024-06-27T13:03:47Z | https://github.com/MonetDB/MonetDB/issues/6571 | 753,619,813 | 6,571 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-04 17:13:20 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-06-20 11:02:15 +0200
## Comment 26340
Date: 2018-04-04 17:13:20 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
The query bellow generates a MAL exception.
Reproducible: Always
### Steps to Reproduce:
1. SELECT ALL 37 * + COALESCE ( - 65, 81 ) * - 36 / + - 36 AS col1;
### Actual Results:
Semantic errors TypeException:user.update[21]:'calc.*' undefined in: X_558:sht := calc.*(X_555:bte, X_557:int);
### Expected Results:
-2405
## Comment 26343
Date: 2018-04-04 17:16:21 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [44290b405a9f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=44290b405a9f) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=44290b405a9f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=44290b405a9f)
Changeset description:
Added tests for bugs 6568, 6569 and 6570
## Comment 26351
Date: 2018-04-08 14:22:40 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [ad3e7db083e1](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ad3e7db083e1) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ad3e7db083e1](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ad3e7db083e1)
Changeset description:
fixes for bug #6570, ie properly lookup the sql_mul function (and
return type) after rewriting math expressions
| Sqlitelogictest select coalesce undefined calc | https://api.github.com/repos/MonetDB/MonetDB/issues/6570/comments | 0 | 2020-11-30T16:26:32Z | 2024-06-27T13:03:46Z | https://github.com/MonetDB/MonetDB/issues/6570 | 753,619,779 | 6,570 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-04 16:48:13 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-06-20 11:02:19 +0200
## Comment 26339
Date: 2018-04-04 16:48:13 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
The query select query bellow returns NULL instead of 78
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab1(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab1 VALUES(51,14,96), (85,5,59), (91,47,68);
3. SELECT ALL + MIN ( ALL + ( + col0 ) ) + - 9 * - ( + COUNT ( * ) ) AS col0 FROM tab1 AS cor0 WHERE NOT ( - col0 + col0 ) BETWEEN ( - 39 ) AND ( - 92 );
4. DROP TABLE tab1;
### Actual Results:
NULL
### Expected Results:
78
## Comment 26342
Date: 2018-04-04 17:16:20 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [44290b405a9f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=44290b405a9f) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=44290b405a9f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=44290b405a9f)
Changeset description:
Added tests for bugs 6568, 6569 and 6570
## Comment 26349
Date: 2018-04-08 11:08:26 +0200
From: @njnes
the problem (if any) is in the range select handling (with the low value >= high value). If the sql is written correctly (ie use symmetric between or between -92 and - 39) the output is correct.
## Comment 26372
Date: 2018-04-16 12:23:22 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [021999eef451](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=021999eef451) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=021999eef451](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=021999eef451)
Changeset description:
An anti-select for an empty range should return all non-nil values.
This fixes bug #6569.
| Sqlitelogictest select query with not between cause with wrong results | https://api.github.com/repos/MonetDB/MonetDB/issues/6569/comments | 0 | 2020-11-30T16:26:29Z | 2024-06-27T13:03:45Z | https://github.com/MonetDB/MonetDB/issues/6569 | 753,619,743 | 6,569 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-04 16:27:57 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes, stely1936
Last updated: 2020-10-06 13:12:24 +0200
## Comment 26338
Date: 2018-04-04 16:27:57 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
Just run the query bellow for an assertion error in rel_bin.c
Reproducible: Always
### Steps to Reproduce:
1. SELECT - 76 AS col2, CASE WHEN CAST ( - MAX ( DISTINCT - 75 ) AS INTEGER ) IS NOT NULL THEN + 4 * - + 48 * + 51 * - - 54 * - 77 * 46 WHEN NOT ( NULL ) IN ( + - ( - - 49 ) * - 1, - 41, - 14 - - CASE - - 1 WHEN - - 82 THEN - - 23 END / + 46 - - 87 * + 52 ) THEN 56 ELSE NULL END AS col2;
### Actual Results:
mserver5: rel_bin.c:576: exp_bin: Assertion `s' failed.
### Expected Results:
A single column with values -76 and 1872896256
Backtrace:
0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:51
1 0x00007ffff4781c41 in __GI_abort () at abort.c:79
2 0x00007ffff4778f7a in __assert_fail_base (fmt=0x7ffff48c9260 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", assertion=assertion@entry=0x7fffe84fd5f8 "s", file=file@entry=0x7fffe84fd588 "rel_bin.c",
line=line@entry=576, function=function@entry=0x7fffe84fddc8 <__PRETTY_FUNCTION__.18599> "exp_bin") at assert.c:92
3 0x00007ffff4778ff2 in __GI___assert_fail (assertion=0x7fffe84fd5f8 "s", file=0x7fffe84fd588 "rel_bin.c", line=576, function=0x7fffe84fddc8 <__PRETTY_FUNCTION__.18599> "exp_bin") at assert.c:101
4 0x00007fffe8383cee in exp_bin (be=0x7fffb0002b60, e=0x7fffb0223430, left=0x7fffb04d63a0, right=0x7fffb04d6510, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:576
5 0x00007fffe838cfe3 in rel2bin_project (be=0x7fffb0002b60, rel=0x7fffb022b880, refs=0x7fffb04d62c0, topn=0x0) at rel_bin.c:2481
6 0x00007fffe8398bcd in subrel_bin (be=0x7fffb0002b60, rel=0x7fffb022b880, refs=0x7fffb04d62c0) at rel_bin.c:5068
7 0x00007fffe838cf10 in rel2bin_project (be=0x7fffb0002b60, rel=0x7fffb022d430, refs=0x7fffb04d62c0, topn=0x0) at rel_bin.c:2469
8 0x00007fffe8398bcd in subrel_bin (be=0x7fffb0002b60, rel=0x7fffb022d430, refs=0x7fffb04d62c0) at rel_bin.c:5068
9 0x00007fffe8398eb5 in _subrel_bin (be=0x7fffb0002b60, rel=0x7fffb022d430, refs=0x7fffb04d62c0) at rel_bin.c:5130
10 0x00007fffe8398fcd in output_rel_bin (be=0x7fffb0002b60, rel=0x7fffb022d430) at rel_bin.c:5153
11 0x00007fffe83b5364 in sql_relation2stmt (be=0x7fffb0002b60, r=0x7fffb022d430) at sql_gencode.c:542
12 0x00007fffe83b5557 in backend_dumpstmt (be=0x7fffb0002b60, mb=0x7fffb00b2340, r=0x7fffb022d430, top=1, add_end=1,
query=0x7fffb00dc890 "select - 76 as col2, case when cast ( - max ( distinct - 75 ) as integer ) is not null then + 4 * - + 48 * + 51 * - - 54 * - 77 * 46 when not ( null ) in ( + - ( - - 49 ) * - 1, - 41, - 14 - - case - "...) at sql_gencode.c:582
13 0x00007fffe83b5dcf in backend_dumpproc (be=0x7fffb0002b60, c=0x7fffe973c368, cq=0x7fffb009a150, r=0x7fffb022d430) at sql_gencode.c:712
14 0x00007fffe8377958 in SQLparser (c=0x7fffe973c368) at sql_scenario.c:1233
15 0x00007ffff7a0f392 in runPhase (c=0x7fffe973c368, phase=1) at mal_scenario.c:510
16 0x00007ffff7a0f49b in runScenarioBody (c=0x7fffe973c368, once=0) at mal_scenario.c:532
17 0x00007ffff7a0f754 in runScenario (c=0x7fffe973c368, once=0) at mal_scenario.c:569
18 0x00007ffff7a11680 in MSserveClient (dummy=0x7fffe973c368) at mal_session.c:519
19 0x00007ffff7a1117b in MSscheduleClient (command=0x7fffb0000b30 "0", challenge=0x7fffc72e1d9b "pb9keHDd", fin=0x1805700, fout=0x7fffc0002d90, protocol=PROTOCOL_9, blocksize=8190, compute_column_widths=0)
at mal_session.c:397
20 0x00007ffff7ab22e2 in doChallenge (data=0x1805710) at mal_mapi.c:279
21 0x00007ffff7699357 in thread_starter (arg=0x9b5780) at gdk_system.c:476
22 0x00007ffff4b0950b in start_thread (arg=0x7fffc72e2700) at pthread_create.c:465
23 0x00007ffff484116f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95
## Comment 26341
Date: 2018-04-04 17:16:18 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [44290b405a9f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=44290b405a9f) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=44290b405a9f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=44290b405a9f)
Changeset description:
Added tests for bugs 6568, 6569 and 6570
## Comment 26350
Date: 2018-04-08 11:09:45 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [a6f3af2adabf](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=a6f3af2adabf) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=a6f3af2adabf](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=a6f3af2adabf)
Changeset description:
fix bug #6568
| Sqlitelogictest crash on complex case query | https://api.github.com/repos/MonetDB/MonetDB/issues/6568/comments | 0 | 2020-11-30T16:26:26Z | 2024-06-27T13:03:44Z | https://github.com/MonetDB/MonetDB/issues/6568 | 753,619,696 | 6,568 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-04 13:56:17 +0200
From: @mlkersten
To: SQL devs <<bugs-sql>>
Version: -- development
Last updated: 2018-04-04 14:01:55 +0200
## Comment 26337
Date: 2018-04-04 13:56:17 +0200
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.13; rv:59.0) Gecko/20100101 Firefox/59.0
Build Identifier:
I expect also (conditional) create/drop schemas.
mk@toke::~> mclient -d demo
Welcome to mclient, the MonetDB/SQL interactive terminal (unreleased)
Database: MonetDB v11.30.0 (unreleased), 'mapi:monetdb://toke.da.cwi.nl:50000/demo'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>create schema sqlscalpel;
operation successful
sql>set schema sqlscalpel;
auto commit mode: on
sql>create table queue(i int);
operation successful
sql>drop table queue;
operation successful
sql>drop schema sqlscalpel;
DROP SCHEMA: cannot drop current schema
Reproducible: Always
| Drop schema not allowed | https://api.github.com/repos/MonetDB/MonetDB/issues/6567/comments | 0 | 2020-11-30T16:26:23Z | 2024-06-28T06:48:14Z | https://github.com/MonetDB/MonetDB/issues/6567 | 753,619,660 | 6,567 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-03 16:22:40 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2019-05-06 19:58:55 +0200
## Comment 26333
Date: 2018-04-03 16:22:40 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
Despite activating FULL_IMPLEMENTATION macro in gdk_calc.c in changeset [ff833040fa32](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ff833040fa32) for bug #6513, the SQL compiler still generates unavailable MAL operations.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab0(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab0 VALUES(97,1,99), (15,81,47), (87,21,10);
3. SELECT - col0 - - - col0 * + CAST ( NULL AS INTEGER ) FROM tab0;
4. SELECT ALL - col1 * - 16 * + CAST ( NULL AS INTEGER ) FROM tab0 cor0;
### Actual Results:
In the first SELECT query, the following MAL exception is generated:
Semantic errors TypeException:user.sql.init[12]:'calc.-' undefined in: X_32176:hge := calc.-(X_32175:int);
In the second SELECT query:
Semantic errors TypeException:user.sql.init[14]:'calc.-' undefined in: X_32228:lng := calc.-(X_32227:int);
### Expected Results:
For both SELECT queries a single column result set with 3 NULL values.
## Comment 26335
Date: 2018-04-03 16:36:57 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [549d90c8aba5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=549d90c8aba5) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=549d90c8aba5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=549d90c8aba5)
Changeset description:
Added tests for bugs 6565 and 6566
## Comment 27000
Date: 2019-05-06 11:22:28 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [49fdc1899528](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=49fdc1899528) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=49fdc1899528](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=49fdc1899528)
Changeset description:
Adding more sqlite tests failing queries:
Bug #6605 outputs only 3 columns out of 6 on default.
Bug #6566 crashes on both Apr2019 and default.
Bug #6579 crashes on default.
## Comment 27002
Date: 2019-05-06 19:58:55 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [dd2ccbff24f0](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=dd2ccbff24f0) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=dd2ccbff24f0](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=dd2ccbff24f0)
Changeset description:
fixes for bug #6566, ie make sure we keep atleast one expression in a groupby
| Sqlitelogictest unavailable calc.- MAL operations | https://api.github.com/repos/MonetDB/MonetDB/issues/6566/comments | 0 | 2020-11-30T16:26:18Z | 2024-06-27T13:03:41Z | https://github.com/MonetDB/MonetDB/issues/6566 | 753,619,613 | 6,566 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-04-03 16:02:52 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-06-20 11:02:09 +0200
## Comment 26332
Date: 2018-04-03 16:02:52 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
Run the query bellow for a SIGSEGV on rel_project2groupby call on rel_select.c.
Reproducible: Always
### Steps to Reproduce:
1. SELECT + NULLIF ( + 67, + + NULLIF ( + + 46, 66 - CASE WHEN 51 IN ( + 91 ) THEN + SUM ( CAST ( NULL AS INTEGER ) ) + - 92 ELSE - ( - 47 ) END ) ) / - CAST ( NULL AS INTEGER ) + + NULLIF ( - COALESCE ( CAST ( - 35 AS INTEGER ), + 6, COUNT ( ALL 28 ) ), 75 + 14 );
### Actual Results:
SIGSEGV on rel_project2groupby call on rel_select.c
### Expected Results:
A single NULL value
Backtrace:
0 0x00007fffe840ab95 in rel_project2groupby (sql=0x7fffb000a0d0, g=0x7fffb01e3c30) at rel_select.c:187
1 0x00007fffe841beb2 in _rel_aggr (sql=0x7fffb000a0d0, rel=0x7fffc72e1780, distinct=0, s=0x1946150, aname=0x7fffb01e1ab0 "sum", args=0x7fffb01e1c80, f=2) at rel_select.c:3607
2 0x00007fffe841d4d0 in rel_aggr (sql=0x7fffb000a0d0, rel=0x7fffc72e1780, se=0x7fffb01e1ca0, f=2) at rel_select.c:3812
3 0x00007fffe84211bd in rel_value_exp2 (sql=0x7fffb000a0d0, rel=0x7fffc72e1780, se=0x7fffb01e1ca0, f=2, ek=..., is_last=0x7fffc72e06cc) at rel_select.c:4663
4 0x00007fffe8421e66 in rel_value_exp (sql=0x7fffb000a0d0, rel=0x7fffc72e1780, se=0x7fffb01e1ca0, f=2, ek=...) at rel_select.c:4845
5 0x00007fffe841b369 in rel_binop (sql=0x7fffb000a0d0, rel=0x7fffc72e1780, se=0x7fffb01e1e80, f=2, ek=...) at rel_select.c:3450
6 0x00007fffe8421166 in rel_value_exp2 (sql=0x7fffb000a0d0, rel=0x7fffc72e1780, se=0x7fffb01e1e80, f=2, ek=..., is_last=0x7fffc72e088c) at rel_select.c:4659
7 0x00007fffe8421e66 in rel_value_exp (sql=0x7fffb000a0d0, rel=0x7fffc72e1780, se=0x7fffb01e1e80, f=2, ek=...) at rel_select.c:4845
8 0x00007fffe841d9ee in rel_case (sql=0x7fffb000a0d0, rel=0x7fffc72e1780, token=46, opt_cond=0x0, when_search_list=0x7fffb01e1f20, opt_else=0x7fffb01e2000, f=2) at rel_select.c:3863
9 0x00007fffe841e3fe in rel_case_exp (sql=0x7fffb000a0d0, rel=0x7fffc72e1780, se=0x7fffb01e2080, f=2) at rel_select.c:3983
10 0x00007fffe8421d63 in rel_value_exp2 (sql=0x7fffb000a0d0, rel=0x7fffc72e1780, se=0x7fffb01e2080, f=2, ek=..., is_last=0x7fffc72e0b7c) at rel_select.c:4819
11 0x00007fffe8421e66 in rel_value_exp (sql=0x7fffb000a0d0, rel=0x7fffc72e1780, se=0x7fffb01e2080, f=2, ek=...) at rel_select.c:4845
12 0x00007fffe841b3ac in rel_binop (sql=0x7fffb000a0d0, rel=0x7fffc72e1780, se=0x7fffb01e21a0, f=2, ek=...) at rel_select.c:3451
13 0x00007fffe8421166 in rel_value_exp2 (sql=0x7fffb000a0d0, rel=0x7fffc72e1780, se=0x7fffb01e21a0, f=2, ek=..., is_last=0x7fffc72e0d3c) at rel_select.c:4659
...
## Comment 26334
Date: 2018-04-03 16:36:56 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [549d90c8aba5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=549d90c8aba5) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=549d90c8aba5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=549d90c8aba5)
Changeset description:
Added tests for bugs 6565 and 6566
## Comment 26336
Date: 2018-04-04 13:44:59 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [52858676c04b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=52858676c04b) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=52858676c04b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=52858676c04b)
Changeset description:
fxes for bugs 6565 and 6565
6565 make sure we handle just created projections in the project2groupby
6566 when rewriting a function with a 'NULL' argument into NULL, cast it into the right type
| Sqlitelogictest crash on complex select query with coalesce call | https://api.github.com/repos/MonetDB/MonetDB/issues/6565/comments | 0 | 2020-11-30T16:26:16Z | 2024-06-27T13:03:40Z | https://github.com/MonetDB/MonetDB/issues/6565 | 753,619,582 | 6,565 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-03-29 17:45:14 +0200
From: @kutsurak
To: SQL devs <<bugs-sql>>
Version: 11.27.13 (Jul2017-SP4)
Last updated: 2018-08-31 13:23:08 +0200
## Comment 26599
Date: 2018-08-10 14:52:18 +0200
From: @kutsurak
This issue is about the authentication used when an execution plan needs data from a remote table. We did not actually need to change the protocol at all, but we needed to extend the way a remote table is defined. For more information take a look at:
https://www.monetdb.org/remote-tables-auth
## Comment 26600
Date: 2018-08-10 14:54:12 +0200
From: @kutsurak
The correct URL is: https://www.monetdb.org/blog/remote-tables-auth
| Changes to the Remote Table definition | https://api.github.com/repos/MonetDB/MonetDB/issues/6564/comments | 0 | 2020-11-30T16:26:14Z | 2024-06-27T13:03:39Z | https://github.com/MonetDB/MonetDB/issues/6564 | 753,619,563 | 6,564 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-03-26 15:27:38 +0200
From: @kutsurak
To: Merovingian devs <<bugs-merovingian>>
Version: 11.27.13 (Jul2017-SP4)
CC: alexo15
Last updated: 2018-10-25 20:51:05 +0200
## Comment 26293
Date: 2018-03-26 15:27:38 +0200
From: @kutsurak
User-Agent: Mozilla/5.0 (X11; Fedora; Linux x86_64; rv:59.0) Gecko/20100101 Firefox/59.0
Build Identifier:
Right now monetdbd accepts connections either only from localhost (listenaddr=localhost), or from everywhere (listenaddr=0.0.0.0).
The user should be able to specify one IP address plus a subnet mask in CIDR notation for greater flexibility.
Reproducible: Always
## Comment 26657
Date: 2018-10-25 20:51:05 +0200
From: Alex Ossipov <<alexo15>>
Using ipv4 notation does not allow remote connections on ipv6 only networks.
Any attempts to start monetdbd with listenaddr set to ipv6 fail with next error:
monetdbd: cannot find host <ipv6 address>
When starting with default localhost value I see errors in the merovingian.log:
ERR discovery[741733]: error while sending broadcast message: Network is unreachable
| monetdbd should provide more fine grained control over hosts allowed to connect | https://api.github.com/repos/MonetDB/MonetDB/issues/6563/comments | 0 | 2020-11-30T16:26:12Z | 2024-06-28T13:39:44Z | https://github.com/MonetDB/MonetDB/issues/6563 | 753,619,531 | 6,563 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-03-25 20:00:35 +0200
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-06-20 11:02:10 +0200
## Comment 26291
Date: 2018-03-25 20:00:35 +0200
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
Run the query below for an assertion error in rel_bin.c
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab1(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab1 VALUES(22,6,8), (28,57,45), (82,44,71);
3. SELECT - + 48 * COUNT ( * ) FROM tab1 AS cor0 GROUP BY col2, col2, col0 HAVING NOT NULL NOT IN ( - + col2, COUNT ( * ), col2 / + 30 + + COUNT ( * ), - 19 );
### Actual Results:
Assertion error in rel_bin.c
### Expected Results:
An empty result set
No backtrace needed :)
## Comment 26292
Date: 2018-03-25 20:01:10 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [7e482660d6ee](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=7e482660d6ee) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=7e482660d6ee](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=7e482660d6ee)
Changeset description:
Added test for Bug #6562
## Comment 26294
Date: 2018-03-28 11:23:27 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [be9ea021c85d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=be9ea021c85d) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=be9ea021c85d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=be9ea021c85d)
Changeset description:
fixed bug #6562, ie only add expressions in the selection phase
| Sqlitelogictest crash on group by query with not in operator | https://api.github.com/repos/MonetDB/MonetDB/issues/6562/comments | 0 | 2020-11-30T16:26:09Z | 2024-06-27T13:03:37Z | https://github.com/MonetDB/MonetDB/issues/6562 | 753,619,481 | 6,562 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-03-24 20:40:35 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2019-05-03 10:37:51 +0200
## Comment 26287
Date: 2018-03-24 20:40:35 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
Just run the query bellow for an assertion error in rel_bin.c
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab0(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab0 VALUES(83,0,38), (26,0,79), (43,81,24);
3. SELECT - col0 * - + 8 + + + col0 + + + col0 + + + 42 * col0 FROM tab0 GROUP BY col0, col0, col1 HAVING NOT 18 + - col0 IN ( + CAST ( + col0 AS INTEGER ) );
### Actual Results:
An assertion error in rel_bin.c:746: exp_bin: Assertion `0' failed.
### Expected Results:
A single column with values 1352, 2236, 4316
The backtrace is already known from previous bugs :)
## Comment 26288
Date: 2018-03-24 20:42:41 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [26af4405a2a5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=26af4405a2a5) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=26af4405a2a5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=26af4405a2a5)
Changeset description:
Added test for Bug #6561
## Comment 26289
Date: 2018-03-25 18:14:46 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [78237ac1720c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=78237ac1720c) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=78237ac1720c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=78237ac1720c)
Changeset description:
fixed bug #6561, ie crash in groupby having in/cast (use rename expression properly)
no longer support (persistent) statistics on temporary tables
## Comment 26321
Date: 2018-03-29 15:39:22 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
## Comment 26994
Date: 2019-05-03 10:37:51 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [2ecd89950f47](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=2ecd89950f47) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=2ecd89950f47](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=2ecd89950f47)
Changeset description:
Adding more sqlite tests failing queries:
Bug #6560 gives right results in Apr2019 branch, but wrong on default.
Bug #6561 we should analyze first, but I think MonetDB outputs wrong results.
Bug #6576 more overflows in calculations.
| Sqlitelogictest crash on group by query with having not in clause | https://api.github.com/repos/MonetDB/MonetDB/issues/6561/comments | 0 | 2020-11-30T16:26:05Z | 2024-06-27T13:03:36Z | https://github.com/MonetDB/MonetDB/issues/6561 | 753,619,436 | 6,561 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-03-23 13:50:42 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2019-05-03 10:37:50 +0200
## Comment 26278
Date: 2018-03-23 13:50:42 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36
Build Identifier:
Just run the query bellow, it gives an assertion error in rel_bin.c.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab1(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab1 VALUES(22,6,8), (28,57,45), (82,44,71);
3. SELECT + - col2 FROM tab1 AS cor0 GROUP BY col0, col2 HAVING ( col2 / + 15 + + 88 ) IN ( AVG ( col2 ) );
### Actual Results:
mserver5: rel_bin.c:576: exp_bin: Assertion `s' failed.
### Expected Results:
An empty result set.
Backtrace:
0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:51
1 0x00007ffff4781c41 in __GI_abort () at abort.c:79
2 0x00007ffff4778f7a in __assert_fail_base (fmt=0x7ffff48c9260 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", assertion=assertion@entry=0x7fffe84ff378 "s", file=file@entry=0x7fffe84ff308 "rel_bin.c",
line=line@entry=576, function=function@entry=0x7fffe84ffb48 <__PRETTY_FUNCTION__.18599> "exp_bin") at assert.c:92
3 0x00007ffff4778ff2 in __GI___assert_fail (assertion=0x7fffe84ff378 "s", file=0x7fffe84ff308 "rel_bin.c", line=576, function=0x7fffe84ffb48 <__PRETTY_FUNCTION__.18599> "exp_bin") at assert.c:101
4 0x00007fffe8385d6e in exp_bin (be=0x7fffb0002b60, e=0x7fffb00f1260, left=0x7fffb00f5780, right=0x7fffb00f67e0, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:576
5 0x00007fffe838f063 in rel2bin_project (be=0x7fffb0002b60, rel=0x7fffb00f1370, refs=0x7fffb00f48e0, topn=0x0) at rel_bin.c:2481
6 0x00007fffe839ac4d in subrel_bin (be=0x7fffb0002b60, rel=0x7fffb00f1370, refs=0x7fffb00f48e0) at rel_bin.c:5068
7 0x00007fffe838c783 in rel2bin_semijoin (be=0x7fffb0002b60, rel=0x7fffb00f18f0, refs=0x7fffb00f48e0) at rel_bin.c:1939
8 0x00007fffe839ab93 in subrel_bin (be=0x7fffb0002b60, rel=0x7fffb00f18f0, refs=0x7fffb00f48e0) at rel_bin.c:5052
9 0x00007fffe838ef90 in rel2bin_project (be=0x7fffb0002b60, rel=0x7fffb00ef560, refs=0x7fffb00f48e0, topn=0x0) at rel_bin.c:2469
10 0x00007fffe839ac4d in subrel_bin (be=0x7fffb0002b60, rel=0x7fffb00ef560, refs=0x7fffb00f48e0) at rel_bin.c:5068
11 0x00007fffe838ef90 in rel2bin_project (be=0x7fffb0002b60, rel=0x7fffb00f2320, refs=0x7fffb00f48e0, topn=0x0) at rel_bin.c:2469
12 0x00007fffe839ac4d in subrel_bin (be=0x7fffb0002b60, rel=0x7fffb00f2320, refs=0x7fffb00f48e0) at rel_bin.c:5068
13 0x00007fffe839af35 in _subrel_bin (be=0x7fffb0002b60, rel=0x7fffb00f2320, refs=0x7fffb00f48e0) at rel_bin.c:5130
14 0x00007fffe839b04d in output_rel_bin (be=0x7fffb0002b60, rel=0x7fffb00f2320) at rel_bin.c:5153
15 0x00007fffe83b72c4 in sql_relation2stmt (be=0x7fffb0002b60, r=0x7fffb00f2320) at sql_gencode.c:542
16 0x00007fffe83b74b7 in backend_dumpstmt (be=0x7fffb0002b60, mb=0x7fffb0125a30, r=0x7fffb00f2320, top=1, add_end=1,
query=0x7fffb0110990 "select + - col2 from tab1 as cor0 group by col0, col2 having ( col2 / + 15 + + 88 ) in ( avg ( col2 ) );") at sql_gencode.c:582
17 0x00007fffe83b7d2f in backend_dumpproc (be=0x7fffb0002b60, c=0x7fffe973e368, cq=0x7fffb0116670, r=0x7fffb00f2320) at sql_gencode.c:712
18 0x00007fffe83799d8 in SQLparser (c=0x7fffe973e368) at sql_scenario.c:1233
19 0x00007ffff7a0da85 in runPhase (c=0x7fffe973e368, phase=1) at mal_scenario.c:510
20 0x00007ffff7a0db8e in runScenarioBody (c=0x7fffe973e368, once=0) at mal_scenario.c:532
21 0x00007ffff7a0de47 in runScenario (c=0x7fffe973e368, once=0) at mal_scenario.c:569
22 0x00007ffff7a0fd82 in MSserveClient (dummy=0x7fffe973e368) at mal_session.c:514
23 0x00007ffff7a0f8d1 in MSscheduleClient (command=0x7fffb0000b30 "0", challenge=0x7fffc72dbd9b "xfouFWzfo", fin=0x17ff770, fout=0x7fffc0002d90, protocol=PROTOCOL_9, blocksize=8190, compute_column_widths=0)
at mal_session.c:397
24 0x00007ffff7ab04d5 in doChallenge (data=0x17ff780) at mal_mapi.c:279
25 0x00007ffff7699479 in thread_starter (arg=0x9c4c10) at gdk_system.c:476
26 0x00007ffff4b0950b in start_thread (arg=0x7fffc72dc700) at pthread_create.c:465
27 0x00007ffff484116f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95
## Comment 26279
Date: 2018-03-23 13:55:15 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [288504db76dc](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=288504db76dc) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=288504db76dc](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=288504db76dc)
Changeset description:
Added test for bug #6560
## Comment 26285
Date: 2018-03-24 13:49:59 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [60cf89593bf0](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=60cf89593bf0) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=60cf89593bf0](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=60cf89593bf0)
Changeset description:
fixed bug #6560, ie handle aliases better in groupby
## Comment 26309
Date: 2018-03-29 15:39:13 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
## Comment 26993
Date: 2019-05-03 10:37:50 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [2ecd89950f47](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=2ecd89950f47) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=2ecd89950f47](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=2ecd89950f47)
Changeset description:
Adding more sqlite tests failing queries:
Bug #6560 gives right results in Apr2019 branch, but wrong on default.
Bug #6561 we should analyze first, but I think MonetDB outputs wrong results.
Bug #6576 more overflows in calculations.
| Sqlitelogictest crash on group by query with having in | https://api.github.com/repos/MonetDB/MonetDB/issues/6560/comments | 0 | 2020-11-30T16:26:01Z | 2024-06-27T13:03:35Z | https://github.com/MonetDB/MonetDB/issues/6560 | 753,619,387 | 6,560 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-03-22 19:02:19 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.27.13 (Jul2017-SP4)
CC: @njnes
Last updated: 2018-03-29 15:39:15 +0200
## Comment 26275
Date: 2018-03-22 19:02:19 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Fedora; Linux x86_64; rv:59.0) Gecko/20100101 Firefox/59.0
Build Identifier:
after creation of a temporary table and analysing its columns data (such that sys.statistics is populated) and next dropping the temporary table explicitly, the associated rows in sys.statistics are not automatically dropped also.
Reproducible: Always
### Steps to Reproduce:
Run SQL:
DELETE FROM sys.statistics WHERE column_id NOT IN (SELECT id FROM sys.columns);
SELECT COUNT(*) as COUNT FROM sys.statistics WHERE column_id NOT IN (SELECT id FROM sys.columns);
-- should be 0
CREATE TABLE tbl_normal (id int, name varchar(99));
SELECT * FROM tbl_normal;
ANALYZE sys.tbl_normal;
SELECT COUNT(*) as COUNT FROM sys.statistics WHERE column_id NOT IN (SELECT id FROM sys.columns);
-- should be o
SELECT type, width, count, "unique", nils FROM sys.statistics WHERE column_id NOT IN (SELECT id FROM sys.columns);
CREATE TEMP TABLE tbl_temp as SELECT * FROM sys.statistics ON COMMIT PRESERVE ROWS;
SELECT * FROM tbl_temp;
SELECT * FROM tmp.tbl_temp;
SELECT name, type, "system", commit_action, access FROM tmp._tables;
-- shows 1 row
SELECT name, type, type_digits, type_scale, "null", number FROM tmp._columns;
-- shows 12 rows
ANALYZE tmp.tbl_temp;
SELECT COUNT(*) as COUNT FROM sys.statistics;
SELECT COUNT(*) as COUNT FROM sys.statistics WHERE column_id NOT IN (SELECT id FROM sys.columns);
-- SELECT * FROM sys.statistics;
SELECT type, width, count, "unique", nils FROM sys.statistics WHERE column_id NOT IN (SELECT id FROM sys.columns);
DROP TABLE tbl_normal;
SELECT COUNT(*) as COUNT FROM sys.statistics;
SELECT COUNT(*) as COUNT FROM sys.statistics WHERE column_id NOT IN (SELECT id FROM sys.columns);
SELECT type, width, count, "unique", nils FROM sys.statistics WHERE column_id NOT IN (SELECT id FROM sys.columns);
DROP TABLE tmp.tbl_temp;
DROP TABLE IF EXISTS tmp.tbl_temp;
SELECT * FROM tmp._tables;
SELECT * FROM tmp._columns;
SELECT COUNT(*) as COUNT FROM sys.statistics;
SELECT COUNT(*) as COUNT FROM sys.statistics WHERE column_id NOT IN (SELECT id FROM sys.columns);
-- should have been 0
SELECT type, width, count, "unique", nils FROM sys.statistics WHERE column_id NOT IN (SELECT id FROM sys.columns);
-- should have been no rows
## Comment 26286
Date: 2018-03-24 15:26:41 +0100
From: @njnes
we should no keep statistics on temporary tables at all.
## Comment 26290
Date: 2018-03-25 18:15:16 +0200
From: @njnes
disallowed statistics on temporary tables
## Comment 26314
Date: 2018-03-29 15:39:15 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| rows in sys.statistics are not removed when a temporary table is dropped. | https://api.github.com/repos/MonetDB/MonetDB/issues/6559/comments | 0 | 2020-11-30T16:25:58Z | 2024-06-27T13:03:34Z | https://github.com/MonetDB/MonetDB/issues/6559 | 753,619,332 | 6,559 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-03-20 18:52:50 +0100
From: Guillaume de GENTILE <<gentile_g>>
To: clients devs <<bugs-clients>>
Version: 11.27.9 (Jul2017-SP2)
CC: @PedroTadim
Last updated: 2018-03-23 13:42:17 +0100
## Comment 26273
Date: 2018-03-20 18:52:50 +0100
From: Guillaume de GENTILE <<gentile_g>>
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/64.0.3282.186 Safari/537.36
Build Identifier:
Hi Pedro,
first of all, I am really impressed by the MonetDB database and I would like to thanks all the team for your work.
Originally I wanted to play with large data (50 million rows with 50 columns) but then I realised that this amount is already quite large for most of the database (non-column store DB) as I was facing some major performance issues.
This is when I tested successfully MonerDB which did perfectly what I wanted :)
Today I saw on your web site that you have a Java embedded version with MonetDBLite-Java which exatly what I was looking for (I don't want to have a separate launcher for the DB server as it increases complexity)
I have just tested the new api, but unfortunately I was not able to use it with the apache DBCP connection pool (works fine if not defined inside the pool).
I get below exception:
Caused by: java.sql.SQLException: isValid() returned false
at org.apache.commons.dbcp2.PoolableConnection.validate(PoolableConnection.java:284)
at org.apache.commons.dbcp2.PoolableConnectionFactory.validateConnection(PoolableConnectionFactory.java:357)
at org.apache.commons.dbcp2.BasicDataSource.validateConnectionFactory(BasicDataSource.java:2307)
at org.apache.commons.dbcp2.BasicDataSource.createPoolableConnectionFactory(BasicDataSource.java:2290)
... 10 more
Do you think that this is something you could fix ?
Thanks and regards,
Guillaume de GENTILE
Reproducible: Always
### Steps to Reproduce:
1. Configure the MonetDBLite-Java JDBC driver with apache DBCP connection pool
2. initialize the pool
3.
### Actual Results:
The pool fails to initialize:
Caused by: java.sql.SQLException: isValid() returned false
at org.apache.commons.dbcp2.PoolableConnection.validate(PoolableConnection.java:284)
at org.apache.commons.dbcp2.PoolableConnectionFactory.validateConnection(PoolableConnectionFactory.java:357)
at org.apache.commons.dbcp2.BasicDataSource.validateConnectionFactory(BasicDataSource.java:2307)
at org.apache.commons.dbcp2.BasicDataSource.createPoolableConnectionFactory(BasicDataSource.java:2290)
... 10 more
### Expected Results:
Pool should initialize correctly
Versions:
commons-dbcp2-2.1.1
monetdb-jdbc-new-2.33
monetdb-java-lite-2.34
## Comment 26277
Date: 2018-03-23 13:42:17 +0100
From: @PedroTadim
Fixed this bug in changeset 211:3d6834ac2039 in monetdb-java repository, with a new release
| Cannot use MonetDBLite-Java with Apache DBCP connection pool | https://api.github.com/repos/MonetDB/MonetDB/issues/6558/comments | 0 | 2020-11-30T16:25:55Z | 2024-06-27T13:03:33Z | https://github.com/MonetDB/MonetDB/issues/6558 | 753,619,280 | 6,558 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-03-15 10:33:50 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2019-08-20 12:07:27 +0200
## Comment 26271
Date: 2018-03-15 10:33:50 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.162 Safari/537.36
Build Identifier:
The query bellow triggers an assertion error in rel_bin.c
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab0(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab0 VALUES(83,0,38), (26,0,79), (43,81,24);
3. SELECT DISTINCT col2 FROM tab0 GROUP BY col2, col1 HAVING NOT NULL NOT IN ( AVG ( ALL + col1 ) );
### Actual Results:
Assertion error on rel_bin.c:576: Assertion `s' failed.
### Expected Results:
An empty result set.
Backtrace:
0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:51
1 0x00007ffff477ac41 in __GI_abort () at abort.c:79
2 0x00007ffff4771f7a in __assert_fail_base (fmt=0x7ffff48c2260 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", assertion=assertion@entry=0x7fffe84f65b8 "s", file=file@entry=0x7fffe84f6548 "rel_bin.c",
line=line@entry=576, function=function@entry=0x7fffe84f6d88 <__PRETTY_FUNCTION__.18599> "exp_bin") at assert.c:92
3 0x00007ffff4771ff2 in __GI___assert_fail (assertion=0x7fffe84f65b8 "s", file=0x7fffe84f6548 "rel_bin.c", line=576, function=0x7fffe84f6d88 <__PRETTY_FUNCTION__.18599> "exp_bin") at assert.c:101
4 0x00007fffe837d01e in exp_bin (be=0x7fffa8002b60, e=0x7fffa84cdce0, left=0x7fffa84cfaa0, right=0x7fffa84cfed0, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:576
5 0x00007fffe8387729 in rel2bin_groupby (be=0x7fffa8002b60, rel=0x7fffa84cd4d0, refs=0x7fffa84cf460) at rel_bin.c:2757
6 0x00007fffe8391f41 in subrel_bin (be=0x7fffa8002b60, rel=0x7fffa84cd4d0, refs=0x7fffa84cf460) at rel_bin.c:5064
7 0x00007fffe8386ca3 in rel2bin_select (be=0x7fffa8002b60, rel=0x7fffa84ce720, refs=0x7fffa84cf460) at rel_bin.c:2605
8 0x00007fffe8391f13 in subrel_bin (be=0x7fffa8002b60, rel=0x7fffa84ce720, refs=0x7fffa84cf460) at rel_bin.c:5060
9 0x00007fffe8386240 in rel2bin_project (be=0x7fffa8002b60, rel=0x7fffa84cec70, refs=0x7fffa84cf460, topn=0x0) at rel_bin.c:2469
10 0x00007fffe8391ee5 in subrel_bin (be=0x7fffa8002b60, rel=0x7fffa84cec70, refs=0x7fffa84cf460) at rel_bin.c:5056
11 0x00007fffe83871ce in rel2bin_groupby (be=0x7fffa8002b60, rel=0x7fffa84cd7d0, refs=0x7fffa84cf460) at rel_bin.c:2686
12 0x00007fffe8391f41 in subrel_bin (be=0x7fffa8002b60, rel=0x7fffa84cd7d0, refs=0x7fffa84cf460) at rel_bin.c:5064
13 0x00007fffe83921cd in _subrel_bin (be=0x7fffa8002b60, rel=0x7fffa84cd7d0, refs=0x7fffa84cf460) at rel_bin.c:5118
14 0x00007fffe83922e5 in output_rel_bin (be=0x7fffa8002b60, rel=0x7fffa84cd7d0) at rel_bin.c:5141
15 0x00007fffe83ae55c in sql_relation2stmt (be=0x7fffa8002b60, r=0x7fffa84cd7d0) at sql_gencode.c:542
16 0x00007fffe83ae74f in backend_dumpstmt (be=0x7fffa8002b60, mb=0x7fffa8094960, r=0x7fffa84cd7d0, top=1, add_end=1,
query=0x7fffa8087a40 "select distinct col2 from tab0 group by col2, col1 having not null not in ( avg ( all + col1 ) );") at sql_gencode.c:582
17 0x00007fffe83aefc7 in backend_dumpproc (be=0x7fffa8002b60, c=0x7fffe9737368, cq=0x7fffa8068e50, r=0x7fffa84cd7d0) at sql_gencode.c:712
18 0x00007fffe8370c88 in SQLparser (c=0x7fffe9737368) at sql_scenario.c:1233
19 0x00007ffff7a0cff5 in runPhase (c=0x7fffe9737368, phase=1) at mal_scenario.c:510
20 0x00007ffff7a0d0fe in runScenarioBody (c=0x7fffe9737368, once=0) at mal_scenario.c:532
21 0x00007ffff7a0d3b7 in runScenario (c=0x7fffe9737368, once=0) at mal_scenario.c:569
22 0x00007ffff7a0f2f2 in MSserveClient (dummy=0x7fffe9737368) at mal_session.c:514
23 0x00007ffff7a0ee41 in MSscheduleClient (command=0x7fffa8000b30 "0", challenge=0x7fffbf2dbd9b "dJ15zqmWYVE", fin=0x17f0040, fout=0x7fffb8002d90, protocol=PROTOCOL_9, blocksize=8190, compute_column_widths=0)
at mal_session.c:397
24 0x00007ffff7aafa45 in doChallenge (data=0x17f0050) at mal_mapi.c:279
25 0x00007ffff7693f11 in thread_starter (arg=0x9b5600) at gdk_system.c:476
26 0x00007ffff4b0250b in start_thread (arg=0x7fffbf2dc700) at pthread_create.c:465
27 0x00007ffff483a16f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95
## Comment 26272
Date: 2018-03-15 10:37:51 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [cfe2a4fe1c9d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=cfe2a4fe1c9d) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=cfe2a4fe1c9d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=cfe2a4fe1c9d)
Changeset description:
Added test for bug #6557
## Comment 26276
Date: 2018-03-23 13:09:00 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [326148a495d3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=326148a495d3) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=326148a495d3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=326148a495d3)
Changeset description:
fixed bug #6557, ie properly name aggregate expressions
## Comment 26320
Date: 2018-03-29 15:39:21 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
## Comment 26663
Date: 2018-11-06 15:01:13 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [6d4152a6de27](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=6d4152a6de27) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=6d4152a6de27](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=6d4152a6de27)
Changeset description:
Added test and fix for bug #6557.
## Comment 26992
Date: 2019-05-02 14:44:52 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [1b22d2eee345](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=1b22d2eee345) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=1b22d2eee345](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=1b22d2eee345)
Changeset description:
Updated bug #6557 with similar query that should give an error.
The error is thrown in Apr2019 branch, but it's crashing on default, maybe because of subquery branch changes.
## Comment 27198
Date: 2019-08-02 15:54:44 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [f7e5625612ef](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=f7e5625612ef) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=f7e5625612ef](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=f7e5625612ef)
Changeset description:
Updated test for bug #6557 with failing query on default branch.
## Comment 27248
Date: 2019-08-20 12:07:27 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [ad472c9f29d3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ad472c9f29d3) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ad472c9f29d3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ad472c9f29d3)
Changeset description:
Updated bug #6557 test with query with wrong result.
| Sqlitelogictest crash on aggregation query with not in | https://api.github.com/repos/MonetDB/MonetDB/issues/6557/comments | 0 | 2020-11-30T16:25:52Z | 2024-06-27T13:03:31Z | https://github.com/MonetDB/MonetDB/issues/6557 | 753,619,239 | 6,557 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-03-14 13:44:18 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-08-31 13:23:10 +0200
## Comment 26266
Date: 2018-03-14 13:44:18 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.146 Safari/537.36
Build Identifier:
MonetDB gets a division by zero while processing the query bellow. From my finds it comes from the COALESCE function call. The error happens by just typing:
SELECT COALESCE ( + 55, + 71 * - 23 + 20 + 74 / - 1 / + 75 );
Reproducible: Always
### Steps to Reproduce:
1. SELECT COALESCE ( + 55, + 71 * - 23 + 20 + 74 / - COUNT ( ALL - 0 ) / + 75, - 85 / - 71 ) - + - 62 + - - CAST ( + 88 AS INTEGER ) * + + 83 AS col0;
### Actual Results:
Division by zero error.
### Expected Results:
A result set with the single value 7421.
## Comment 26280
Date: 2018-03-23 20:21:06 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [1e3453a67fe1](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=1e3453a67fe1) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=1e3453a67fe1](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=1e3453a67fe1)
Changeset description:
fixe bug #6556, ie push conditional code into the batcalc operators
(scalar case is handled by mal barriers)
## Comment 26283
Date: 2018-03-24 10:29:08 +0100
From: @njnes
fixed in default
| Sqlitelogictest division by zero on COALESCE call | https://api.github.com/repos/MonetDB/MonetDB/issues/6556/comments | 0 | 2020-11-30T16:25:48Z | 2024-06-27T13:03:30Z | https://github.com/MonetDB/MonetDB/issues/6556 | 753,619,185 | 6,556 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-03-14 13:22:47 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-03-29 15:39:06 +0200
## Comment 26264
Date: 2018-03-14 13:22:47 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.146 Safari/537.36
Build Identifier:
MonetDB generates redundant columns in query stated bellow.
Reproducible: Always
### Steps to Reproduce:
1. SELECT DISTINCT CASE WHEN + CASE + - COUNT ( * ) WHEN - 10 THEN NULL WHEN + - 23 - 53 THEN ( COUNT ( DISTINCT + ( + 21 ) ) ) * 72 + COALESCE ( + + 45, 9 ) WHEN - 40 + - 7 + - 1 THEN NULL END NOT IN ( + 14 + + 18 ) THEN NULL ELSE ( - - 46 ) END;
### Actual Results:
This result set:
+------+------+------+------+------+
| L3 | L4 | L5 | L6 | L10 |
+======+======+======+======+======+
| 1 | 1 | 1 | 1 | 46 |
+------+------+------+------+------+
### Expected Results:
This result set:
+------+
| L10 |
+======+
| 46 |
+------+
## Comment 26268
Date: 2018-03-14 13:55:03 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [55091706b910](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=55091706b910) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=55091706b910](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=55091706b910)
Changeset description:
Added tests for bugs 6554, 6555 and 65556
## Comment 26270
Date: 2018-03-14 16:28:18 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [3f518a29f2a9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=3f518a29f2a9) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=3f518a29f2a9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=3f518a29f2a9)
Changeset description:
fixes for bugs 6554 and 6555
## Comment 26299
Date: 2018-03-29 15:39:06 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| Sqlitelogictest wrong result set generated from complex case statement | https://api.github.com/repos/MonetDB/MonetDB/issues/6555/comments | 0 | 2020-11-30T16:25:45Z | 2024-06-27T13:03:29Z | https://github.com/MonetDB/MonetDB/issues/6555 | 753,619,139 | 6,555 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-03-14 13:13:54 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-03-29 15:39:25 +0200
## Comment 26263
Date: 2018-03-14 13:13:54 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.146 Safari/537.36
Build Identifier:
Mserver gets a stack overflow in the rel_optimizer while optimizing the query bellow.
Reproducible: Always
### Steps to Reproduce:
1. SELECT 50, CASE WHEN NOT - COUNT ( * ) IN ( - - 41 * + 92 * + + ( + + 10 ) + - MAX ( DISTINCT + 74 ), + 14, + 11 * + COUNT ( * ) * + ( - + CASE + 77 WHEN - 52 THEN + + 43 + + 71 + - AVG ( ALL + + 93 ) + 12 / - - COUNT ( * ) ELSE NULL END ), COUNT ( - 91 ) + + + 71 * + + CASE - + 53 WHEN - + 94 + - 39 - + - 10 + 13 + SUM ( DISTINCT - + 24 ) THEN NULL WHEN - AVG ( ALL 53 ) THEN 76 ELSE - 17 END ) THEN - AVG ( ALL - ( + 63 ) ) END * + 62 AS col0;
### Actual Results:
Stack overflow in the rel_optimizer.
### Expected Results:
A single result set with values 50, NULL
Backtrace:
0 0x00007fffe8468f10 in rel_reset_subquery (rel=0x7fffb0102a30) at rel_optimizer.c:9281
1 0x00007fffe8468f15 in rel_reset_subquery (rel=0x7fffb01025a0) at rel_optimizer.c:9281
2 0x00007fffe8468f15 in rel_reset_subquery (rel=0x7fffb0102a30) at rel_optimizer.c:9281
...
## Comment 26267
Date: 2018-03-14 13:55:01 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [55091706b910](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=55091706b910) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=55091706b910](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=55091706b910)
Changeset description:
Added tests for bugs 6554, 6555 and 65556
## Comment 26269
Date: 2018-03-14 16:28:16 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [3f518a29f2a9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=3f518a29f2a9) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=3f518a29f2a9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=3f518a29f2a9)
Changeset description:
fixes for bugs 6554 and 6555
## Comment 26325
Date: 2018-03-29 15:39:25 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| Sqlitelogictest crash on complex case statement | https://api.github.com/repos/MonetDB/MonetDB/issues/6554/comments | 0 | 2020-11-30T16:25:42Z | 2024-06-27T13:03:28Z | https://github.com/MonetDB/MonetDB/issues/6554 | 753,619,105 | 6,554 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-03-12 10:30:27 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-03-29 15:39:20 +0200
## Comment 26252
Date: 2018-03-12 10:30:27 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.146 Safari/537.36
Build Identifier:
Infinite recursion happens in mserver in rel_reset_subquery call, leading to a stack overflow while optimizing the query bellow.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab1(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab1 VALUES(22,6,8), (28,57,45), (82,44,71);
3. SELECT DISTINCT + cor0.col1 / cor0.col1 col2 FROM tab1 AS cor0 GROUP BY col1 HAVING NOT NULL IN ( - col1 * cor0.col1 );
### Actual Results:
Infinite recursion in rel_reset_subquery.
### Expected Results:
An empty result set.
Backtrace:
0 0x00007fffe843ad85 in rel_reset_subquery (rel=0x7fffb4254950) at rel_optimizer.c:9281
1 0x00007fffe843ad8a in rel_reset_subquery (rel=0x7fffb4254950) at rel_optimizer.c:9281
2 0x00007fffe843ad8a in rel_reset_subquery (rel=0x7fffb4254950) at rel_optimizer.c:9281
.....
## Comment 26254
Date: 2018-03-12 10:35:16 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [5100acc0528d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=5100acc0528d) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=5100acc0528d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=5100acc0528d)
Changeset description:
Added tests for bugs 6552 and 6553
## Comment 26255
Date: 2018-03-13 10:13:40 +0100
From: @sjoerdmullender
The problem is that rel==rel->l so when rel_reset_subquery calls itself recursively on rel->l, it just repeats infinitely.
The assignment where rel->l is assigned the value rel is in rel_logical_value_exp where it says
if (outer)
outer->l = z;
because at this point, outer == *rel, left == *rel, and hence z == *rel (this is a different rel, *rel here is the same as the rel above).
## Comment 26257
Date: 2018-03-14 10:03:52 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [8c9f8c70fb25](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=8c9f8c70fb25) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=8c9f8c70fb25](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=8c9f8c70fb25)
Changeset description:
fixed bugs 6552 and 6553, by more defensive code
## Comment 26258
Date: 2018-03-14 10:04:05 +0100
From: @njnes
fixed crash
## Comment 26319
Date: 2018-03-29 15:39:20 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| Sqlitelogictest crash on aggregation with having statement | https://api.github.com/repos/MonetDB/MonetDB/issues/6553/comments | 0 | 2020-11-30T16:25:39Z | 2024-06-27T13:03:27Z | https://github.com/MonetDB/MonetDB/issues/6553 | 753,619,060 | 6,553 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-03-12 10:11:24 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-03-29 15:39:15 +0200
## Comment 26251
Date: 2018-03-12 10:11:24 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.146 Safari/537.36
Build Identifier:
Just run the query bellow.
Reproducible: Always
### Steps to Reproduce:
1. SELECT CASE WHEN + - 44 <> 44 * - - 17 + 30 AND NOT NULL IS NOT NULL THEN COUNT ( * ) WHEN NULL NOT IN ( COUNT ( DISTINCT + 86 ) * - COUNT ( * ) ) THEN + 53 ELSE ( 66 ) * - CAST ( NULL AS INTEGER ) + 10 END;
### Actual Results:
SIGSEV in rel_select.c:2170 for dereferencing a NULL pointer.
### Expected Results:
1
Backtrace:
0 0x00007fffe83dfc90 in rel_logical_value_exp (sql=0x7fffb400a080, rel=0x7fffcb388780, sc=0x7fffb41e30e0, f=2) at rel_select.c:2170
1 0x00007fffe83e9c23 in rel_case (sql=0x7fffb400a080, rel=0x7fffcb388780, token=46, opt_cond=0x0, when_search_list=0x7fffb41e2aa0, opt_else=0x7fffb41e3820, f=2) at rel_select.c:3891
2 0x00007fffe83ea312 in rel_case_exp (sql=0x7fffb400a080, rel=0x7fffcb388780, se=0x7fffb41e38a0, f=2) at rel_select.c:3974
3 0x00007fffe83edc6f in rel_value_exp2 (sql=0x7fffb400a080, rel=0x7fffcb388780, se=0x7fffb41e38a0, f=2, ek=..., is_last=0x7fffcb38869c) at rel_select.c:4810
4 0x00007fffe83edd72 in rel_value_exp (sql=0x7fffb400a080, rel=0x7fffcb388780, se=0x7fffb41e38a0, f=2, ek=...) at rel_select.c:4836
5 0x00007fffe83ede71 in column_exp (sql=0x7fffb400a080, rel=0x7fffcb388780, column_e=0x7fffb41e3920, f=2) at rel_select.c:4849
6 0x00007fffe83ee0d8 in rel_column_exp (sql=0x7fffb400a080, rel=0x7fffcb388780, column_e=0x7fffb41e3920, f=2) at rel_select.c:4893
7 0x00007fffe83ee254 in rel_simple_select (sql=0x7fffb400a080, rel=0x7fffb41e3b10, where=0x0, selection=0x7fffb41e3940, distinct=0) at rel_select.c:4922
8 0x00007fffe83efe37 in rel_query (sql=0x7fffb400a080, rel=0x0, sq=0x7fffb41e3a20, toplevel=1, ek=..., apply=8) at rel_select.c:5337
9 0x00007fffe83f182b in rel_subquery (sql=0x7fffb400a080, rel=0x0, sq=0x7fffb41e3a20, ek=..., apply=8) at rel_select.c:5716
10 0x00007fffe83f1952 in rel_selects (sql=0x7fffb400a080, s=0x7fffb41e3a20) at rel_select.c:5737
11 0x00007fffe83d5f2e in rel_semantic (sql=0x7fffb400a080, s=0x7fffb41e3a20) at rel_semantic.c:231
12 0x00007fffe831ed0a in sql_symbol2relation (c=0x7fffb400a080, sym=0x7fffb41e3a20) at sql.c:118
13 0x00007fffe833ff99 in SQLparser (c=0x7fffe9709368) at sql_scenario.c:1171
14 0x00007ffff7a0c09f in runPhase (c=0x7fffe9709368, phase=1) at mal_scenario.c:510
15 0x00007ffff7a0c1a8 in runScenarioBody (c=0x7fffe9709368, once=0) at mal_scenario.c:532
16 0x00007ffff7a0c461 in runScenario (c=0x7fffe9709368, once=0) at mal_scenario.c:569
17 0x00007ffff7a0e39c in MSserveClient (dummy=0x7fffe9709368) at mal_session.c:514
18 0x00007ffff7a0deeb in MSscheduleClient (command=0x7fffb4000b30 "\260<", challenge=0x7fffcb388d9b "9IUXRjfyi", fin=0x17df890, fout=0x7fffc4002d90, protocol=PROTOCOL_9, blocksize=8190,
compute_column_widths=0) at mal_session.c:397
19 0x00007ffff7aaeaef in doChallenge (data=0x17df8a0) at mal_mapi.c:279
20 0x00007ffff7692f06 in thread_starter (arg=0x7fffc4004e80) at gdk_system.c:476
21 0x00007ffff4af661b in start_thread (arg=0x7fffcb389700) at pthread_create.c:465
22 0x00007ffff4823c2f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95
## Comment 26253
Date: 2018-03-12 10:35:15 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [5100acc0528d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=5100acc0528d) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=5100acc0528d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=5100acc0528d)
Changeset description:
Added tests for bugs 6552 and 6553
## Comment 26256
Date: 2018-03-14 10:03:51 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [8c9f8c70fb25](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=8c9f8c70fb25) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=8c9f8c70fb25](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=8c9f8c70fb25)
Changeset description:
fixed bugs 6552 and 6553, by more defensive code
## Comment 26259
Date: 2018-03-14 10:05:01 +0100
From: @njnes
fixed crash
## Comment 26313
Date: 2018-03-29 15:39:15 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| Sqlitelogictest crash on complex case statement | https://api.github.com/repos/MonetDB/MonetDB/issues/6552/comments | 0 | 2020-11-30T16:25:36Z | 2024-06-27T13:03:26Z | https://github.com/MonetDB/MonetDB/issues/6552 | 753,618,999 | 6,552 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-03-05 11:03:11 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-03-29 15:38:59 +0200
## Comment 26242
Date: 2018-03-05 11:03:11 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/64.0.3282.186 Safari/537.36
Build Identifier:
In the SQL query bellow, by providing the + sign before the 77, changes the semantics of the query returning 0 instead of NULL.
Reproducible: Always
### Steps to Reproduce:
1. SELECT 0 * + 77 * CAST ( NULL AS INTEGER ) AS col2;
### Expected Results:
NULL
## Comment 26244
Date: 2018-03-05 11:05:12 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [7582f0753563](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=7582f0753563) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=7582f0753563](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=7582f0753563)
Changeset description:
Added tests for bugs 6550 and 6551
## Comment 26247
Date: 2018-03-10 15:09:43 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [d6b55762ae95](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d6b55762ae95) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=d6b55762ae95](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=d6b55762ae95)
Changeset description:
fixes for bug #6551, ie make sure we detect NULL better, when rewriting simple math expressions.
## Comment 26248
Date: 2018-03-10 17:05:49 +0100
From: @njnes
fixed by handling null;s better in optimizer
## Comment 26296
Date: 2018-03-29 15:38:59 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| Sqlitelogictest wrong NULL value cast | https://api.github.com/repos/MonetDB/MonetDB/issues/6551/comments | 0 | 2020-11-30T16:25:33Z | 2024-06-27T13:03:25Z | https://github.com/MonetDB/MonetDB/issues/6551 | 753,618,960 | 6,551 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-03-05 10:39:42 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-03-29 15:39:14 +0200
## Comment 26241
Date: 2018-03-05 10:39:42 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/64.0.3282.186 Safari/537.36
Build Identifier:
Mserver gets an assertion error while compiling the query bellow.
Reproducible: Always
### Steps to Reproduce:
1. SELECT ALL CASE WHEN NOT NULL NOT IN ( - COALESCE ( + - CAST ( + 45 AS INTEGER ), - 66, - 9 * COALESCE ( 80, NULLIF ( - 12, + + ( + 26 ) + - - MAX ( 18 ) + 66 + 98 ) + + + AVG ( ALL 35 ), - - 55 - - - 67, + 42 ) ), + COUNT ( * ) * - + 23, + 66 * + 93 + - ( - 61 ) + + 10 * + 92 * + - 40 * - 47 / 29, COUNT ( * ) * 30 / 28 + 73 ) THEN COUNT ( * ) / - - 86 ELSE NULL END - - 66;
### Actual Results:
Assertion error on rel_bin.c
### Expected Results:
A single result set with NULL
Assertion error:
mserver5: rel_bin.c:576: exp_bin: Assertion `s' failed.
Backtrace:
0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:51
1 0x00007ffff4744381 in __GI_abort () at abort.c:79
2 0x00007ffff473a8fa in __assert_fail_base (fmt=0x7ffff48b5c28 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", assertion=assertion@entry=0x7fffe84c77f8 "s", file=file@entry=0x7fffe84c7788 "rel_bin.c",
line=line@entry=576, function=function@entry=0x7fffe84c7fc8 <__PRETTY_FUNCTION__.18611> "exp_bin") at assert.c:92
3 0x00007ffff473a972 in __GI___assert_fail (assertion=0x7fffe84c77f8 "s", file=0x7fffe84c7788 "rel_bin.c", line=576, function=0x7fffe84c7fc8 <__PRETTY_FUNCTION__.18611> "exp_bin") at assert.c:101
4 0x00007fffe834b531 in exp_bin (be=0x7fffb4002b60, e=0x7fffb420d030, left=0x7fffb4260330, right=0x7fffb4260460, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:576
5 0x00007fffe834abe8 in exp_bin (be=0x7fffb4002b60, e=0x7fffb42499d0, left=0x7fffb4260330, right=0x7fffb4260460, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:479
6 0x00007fffe834a9c5 in exp_bin (be=0x7fffb4002b60, e=0x7fffb4249a50, left=0x7fffb4260330, right=0x7fffb4260460, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:450
7 0x00007fffe8354826 in rel2bin_project (be=0x7fffb4002b60, rel=0x7fffb4201d20, refs=0x7fffb4259ab0, topn=0x0) at rel_bin.c:2481
8 0x00007fffe83603f8 in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb4201d20, refs=0x7fffb4259ab0) at rel_bin.c:5056
9 0x00007fffe8354753 in rel2bin_project (be=0x7fffb4002b60, rel=0x7fffb4248d20, refs=0x7fffb4259ab0, topn=0x0) at rel_bin.c:2469
10 0x00007fffe83603f8 in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb4248d20, refs=0x7fffb4259ab0) at rel_bin.c:5056
11 0x00007fffe83606e0 in _subrel_bin (be=0x7fffb4002b60, rel=0x7fffb4248d20, refs=0x7fffb4259ab0) at rel_bin.c:5118
12 0x00007fffe83607f8 in output_rel_bin (be=0x7fffb4002b60, rel=0x7fffb4248d20) at rel_bin.c:5141
13 0x00007fffe837d189 in sql_relation2stmt (be=0x7fffb4002b60, r=0x7fffb4248d20) at sql_gencode.c:532
14 0x00007fffe837d37c in backend_dumpstmt (be=0x7fffb4002b60, mb=0x7fffb4210bf0, r=0x7fffb4248d20, top=1, add_end=1,
query=0x7fffb420f490 "select all case when not null not in ( - coalesce ( + - cast ( + 45 as integer ), - 66, - 9 * coalesce ( 80, nullif ( - 12, + + ( + 26 ) + - - max ( 18 ) + 66 + 98 ) + + + avg ( all 35 ), - - 55 - - -"...) at sql_gencode.c:572
15 0x00007fffe837dbf4 in backend_dumpproc (be=0x7fffb4002b60, c=0x7fffe9708368, cq=0x7fffb42153d0, r=0x7fffb4248d20) at sql_gencode.c:702
16 0x00007fffe833efb6 in SQLparser (c=0x7fffe9708368) at sql_scenario.c:1233
17 0x00007ffff7a0b07f in runPhase (c=0x7fffe9708368, phase=1) at mal_scenario.c:510
18 0x00007ffff7a0b188 in runScenarioBody (c=0x7fffe9708368, once=0) at mal_scenario.c:532
19 0x00007ffff7a0b441 in runScenario (c=0x7fffe9708368, once=0) at mal_scenario.c:569
20 0x00007ffff7a0d37c in MSserveClient (dummy=0x7fffe9708368) at mal_session.c:514
21 0x00007ffff7a0cecb in MSscheduleClient (command=0x7fffb4000b30 "0+", challenge=0x7fffcae5bd9b "aTVCQTWqy", fin=0x7fffc4000ca0, fout=0x7fffc4002d90, protocol=PROTOCOL_9, blocksize=8190,
compute_column_widths=0) at mal_session.c:397
22 0x00007ffff7aadc14 in doChallenge (data=0x7fffc4000cb0) at mal_mapi.c:290
23 0x00007ffff7691f06 in thread_starter (arg=0x7fffc4000ce0) at gdk_system.c:476
24 0x00007ffff4af561b in start_thread (arg=0x7fffcae5c700) at pthread_create.c:465
25 0x00007ffff482298f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95
## Comment 26243
Date: 2018-03-05 11:05:10 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [7582f0753563](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=7582f0753563) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=7582f0753563](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=7582f0753563)
Changeset description:
Added tests for bugs 6550 and 6551
## Comment 26249
Date: 2018-03-11 18:54:40 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [81dd28eef303](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=81dd28eef303) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=81dd28eef303](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=81dd28eef303)
Changeset description:
fixed bug #6550
## Comment 26250
Date: 2018-03-11 21:32:41 +0100
From: @njnes
fixed complicated combination of in, case and aggregation.
## Comment 26311
Date: 2018-03-29 15:39:14 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| Sqlitelogictest crash on complex CASE statement | https://api.github.com/repos/MonetDB/MonetDB/issues/6550/comments | 0 | 2020-11-30T16:25:29Z | 2024-06-27T13:03:24Z | https://github.com/MonetDB/MonetDB/issues/6550 | 753,618,911 | 6,550 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-03-01 10:11:52 +0100
From: Zhouyang <<jiazhouyang09>>
To: clients devs <<bugs-clients>>
Version: 11.27.13 (Jul2017-SP4)
Last updated: 2018-03-29 15:39:08 +0200
## Comment 26234
Date: 2018-03-01 10:11:52 +0100
From: Zhouyang <<jiazhouyang09>>
Created attachment 600
Fix reported problems
Hi,
I analyzed the monetdb source code and found some potential API bugs.
The function listen may fail, when it happens, there would be unexpected results, which are hard to debug without proper logs.
This patch (against the latest master branch) adds log messages when listen fails.
Best,
Zhouyang
> Attached file: [listen_monetdb-trunk.patch](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_6549_listen_monetdb-trunk.patch_600) (text/plain, 1550 bytes)
> Description: Fix reported problems
## Comment 26235
Date: 2018-03-01 15:09:38 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [a7ac54dc0fa4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=a7ac54dc0fa4) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=a7ac54dc0fa4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=a7ac54dc0fa4)
Changeset description:
Fixes for bug #6549, check for system calls failures
## Comment 26245
Date: 2018-03-08 16:00:08 +0100
From: @sjoerdmullender
Patch was applied.
## Comment 26303
Date: 2018-03-29 15:39:08 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| Add log messages when listen fails | https://api.github.com/repos/MonetDB/MonetDB/issues/6549/comments | 0 | 2020-11-30T16:25:26Z | 2024-06-27T13:03:23Z | https://github.com/MonetDB/MonetDB/issues/6549 | 753,618,871 | 6,549 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-27 11:50:10 +0100
From: Nick <<nickchernov77>>
To: SQL devs <<bugs-sql>>
Version: 11.27.13 (Jul2017-SP4)
CC: @njnes
Last updated: 2018-03-29 15:39:16 +0200
## Comment 26231
Date: 2018-02-27 11:50:10 +0100
From: Nick <<nickchernov77>>
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:58.0) Gecko/20100101 Firefox/58.0
Build Identifier:
I have two databases (v11.27.13 both) on two Debian8 machines. One with
corrupted table (in my case it got corrupted due to bug in Jul2017-SP1) and
another with remote table to the corrupted table. When I run the SELECT
query on the remote table I get an error. And the session opened by this
query to the first DB did not close. I see it by querying 'sys.sessions'
table. And every time I run this SELECT I get additional record in
'sys.sessions'. This happens only if en error occurs, on successful SELECT
the session immediately disappears. How can I force this sessions to close
or is it a bug?
Reproducible: Always
### Steps to Reproduce:
1. Have two instances on MonetDB with 'storage' databases in each.
2. In the first instance create regular table like this:
create table storage.ttt(col1 varchar(10));
3. Fill it with some data.
4. Corrupt this table somehow (Don't know how to reproduce this step. I got it corrupted accidentally due to bug in Jul2017-SP1)
5. In the second instance create remote table to the table created in step 2:
create remote table storage.ttt(col1 varchar(10))ON 'mapi:monetdb://<hostname>:50000/storage';
7. Check opened sessions on the first instance by running the query under 'monetdb' user:
select * from sys.sessions;
1 tuple (3.907ms)
8. Run query on remote table:
select * from storage.ttt;
and get an error about corrupted BAT or something.
9. Check opened sessions again and the additional inactive session will be listed which will never close.
### Actual Results:
Increasing number of inactive sessions from the second instance.
### Expected Results:
Close each session after query on the remote table has failed.
mserver5 --version
MonetDB 5 server v11.27.13 "Jul2017-SP4" (64-bit, 128-bit integers)
Copyright (c) 1993 - July 2008 CWI
Copyright (c) August 2008 - 2018 MonetDB B.V., all rights reserved
Visit https://www.monetdb.org/ for further information
Found 11.8GiB available memory, 4 available cpu cores
Libraries:
libpcre: 8.35 2014-04-04 (compiled with 8.35)
openssl: OpenSSL 1.0.1t 3 May 2016 (compiled with )
libxml2: 2.9.1 (compiled with 2.9.1)
Compiled by: root@dev.monetdb.org (x86_64-pc-linux-gnu)
Compilation: gcc -O3 -fomit-frame-pointer -pipe -g -D_FORTIFY_SOURCE=2
Linking : /usr/bin/ld -m elf_x86_64
## Comment 26232
Date: 2018-02-27 14:26:05 +0100
From: Nick <<nickchernov77>>
In my case the error I got on step 8 is:
Cannot subtract delete column
BATdiff: inputs not compatible.
## Comment 26260
Date: 2018-03-14 11:46:27 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [a20ec2286704](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=a20ec2286704) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=a20ec2286704](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=a20ec2286704)
Changeset description:
fixes for bug #6548 (ie add exception block to catch exceptions and disconnect from the
remote mserver)
also fixed problem in newExitStmt, don't push a newAssignment statement
as that is allready pushed.
## Comment 26261
Date: 2018-03-14 11:47:50 +0100
From: @njnes
solved by an exception block
## Comment 26315
Date: 2018-03-29 15:39:16 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| Select from remote table leaves session open | https://api.github.com/repos/MonetDB/MonetDB/issues/6548/comments | 0 | 2020-11-30T16:25:22Z | 2024-06-27T13:03:22Z | https://github.com/MonetDB/MonetDB/issues/6548 | 753,618,822 | 6,548 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-20 11:55:50 +0100
From: Manuel <<manuel>>
To: SQL devs <<bugs-sql>>
Version: 11.27.11 (Jul2017-SP3)
Last updated: 2018-03-29 15:39:26 +0200
## Comment 26225
Date: 2018-02-20 11:55:50 +0100
From: Manuel <<manuel>>
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/64.0.3282.167 Safari/537.36
Build Identifier:
The following query
select max("A") from "unitTestDontDelete" where "A" ilike ('%' || '' || '%')
on the table
CREATE TABLE "sys"."unitTestDontDelete" (
"A" VARCHAR(255),
"B" BIGINT,
"C" DOUBLE,
"D" TIMESTAMP
);
INSERT INTO "sys"."unitTestDontDelete" VALUES (NULL, NULL, NULL, NULL);
INSERT INTO "sys"."unitTestDontDelete" VALUES ('Cat1', 0, 0.5, '2013-06-10 11:10:10.000000');
INSERT INTO "sys"."unitTestDontDelete" VALUES ('Cat2', 1, 1.5, '2013-06-11 12:11:11.000000');
INSERT INTO "sys"."unitTestDontDelete" VALUES ('Cat1', 2, 2.5, '2013-06-12 13:12:12.000000');
INSERT INTO "sys"."unitTestDontDelete" VALUES ('Cat2', 3, 3.5, '2013-06-13 14:13:13.000000');
INSERT INTO "sys"."unitTestDontDelete" VALUES ('Cat1', 4, 4.5, '2013-06-14 15:14:14.000000');
INSERT INTO "sys"."unitTestDontDelete" VALUES ('Cat2', 5, 5.5, '2013-06-15 16:15:15.000000');
INSERT INTO "sys"."unitTestDontDelete" VALUES ('Cat1', 6, 6.5, '2013-06-16 17:16:16.000000');
INSERT INTO "sys"."unitTestDontDelete" VALUES ('Cat2', 7, 7.5, '2013-06-17 18:17:17.000000');
INSERT INTO "sys"."unitTestDontDelete" VALUES ('Cat1', 8, 8.5, '2013-06-18 19:18:18.000000');
yields different results depending on the OS:
On Windows it returns null,
On Linux/Mac it returns 'Cat2'
Reproducible: Always
### Steps to Reproduce:
1. Import the table unitTestDontDelete on different OSs
2. Execute the query select max("A") from "unitTestDontDelete" where "A" ilike ('%' || '' || '%')
### Actual Results:
OS dependent
## Comment 26226
Date: 2018-02-20 13:43:02 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [b70be09e345c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=b70be09e345c) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=b70be09e345c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=b70be09e345c)
Changeset description:
Fixed implementation of strcasestr for when the library doesn't provide it.
The old implementation had several flaws: if needle is "" it would
return NULL instead of haystack; if haystack is "aaab" and needle is
"aab" it would return NULL instead of a pointer to the second 'a'.
This fixes bug #6547.
## Comment 26227
Date: 2018-02-20 13:43:06 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [77f42f1749f8](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=77f42f1749f8) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=77f42f1749f8](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=77f42f1749f8)
Changeset description:
Test for bug #6547.
## Comment 26328
Date: 2018-03-29 15:39:26 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| OS-dependent behaviour for ilike | https://api.github.com/repos/MonetDB/MonetDB/issues/6547/comments | 0 | 2020-11-30T16:25:19Z | 2024-06-27T13:03:21Z | https://github.com/MonetDB/MonetDB/issues/6547 | 753,618,772 | 6,547 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-19 10:27:47 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-03-29 15:39:10 +0200
## Comment 26219
Date: 2018-02-19 10:27:47 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/64.0.3282.167 Safari/537.36
Build Identifier:
Mserver5 crashes while compiling a SELECT query with IN operator.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab0(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab0 VALUES (97,1,99),(15,81,47),(87,21,10);
3. SELECT * FROM tab0 cor0 WHERE ( col1 ) IN ( + 57, + col1 / - col2 );
### Actual Results:
Assertion error in rel_bin.c
### Expected Results:
Empty result set.
The assertion error:
mserver5: rel_bin.c:576: exp_bin: Assertion `s' failed.
Backtrace:
0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:51
1 0x00007ffff4748381 in __GI_abort () at abort.c:79
2 0x00007ffff473e8fa in __assert_fail_base (fmt=0x7ffff48b9c28 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", assertion=assertion@entry=0x7fffe84cbd58 "s", file=file@entry=0x7fffe84cbce8 "rel_bin.c",
line=line@entry=576, function=function@entry=0x7fffe84cc518 <__PRETTY_FUNCTION__.18615> "exp_bin") at assert.c:92
3 0x00007ffff473e972 in __GI___assert_fail (assertion=0x7fffe84cbd58 "s", file=0x7fffe84cbce8 "rel_bin.c", line=576, function=0x7fffe84cc518 <__PRETTY_FUNCTION__.18615> "exp_bin") at assert.c:101
4 0x00007fffe8350b22 in exp_bin (be=0x7fffb4002b60, e=0x7fffb4147130, left=0x7fffb414d5a0, right=0x7fffb414db80, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:576
5 0x00007fffe8359e17 in rel2bin_project (be=0x7fffb4002b60, rel=0x7fffb41471c0, refs=0x7fffb414b840, topn=0x0) at rel_bin.c:2481
6 0x00007fffe836596a in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb41471c0, refs=0x7fffb414b840) at rel_bin.c:5053
7 0x00007fffe8359d44 in rel2bin_project (be=0x7fffb4002b60, rel=0x7fffb4149b70, refs=0x7fffb414b840, topn=0x0) at rel_bin.c:2469
8 0x00007fffe836596a in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb4149b70, refs=0x7fffb414b840) at rel_bin.c:5053
9 0x00007fffe8358506 in rel2bin_union (be=0x7fffb4002b60, rel=0x7fffb4148e10, refs=0x7fffb414b840) at rel_bin.c:2160
10 0x00007fffe83658de in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb4148e10, refs=0x7fffb414b840) at rel_bin.c:5041
11 0x00007fffe8357537 in rel2bin_semijoin (be=0x7fffb4002b60, rel=0x7fffb4146160, refs=0x7fffb414b840) at rel_bin.c:1939
12 0x00007fffe83658b0 in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb4146160, refs=0x7fffb414b840) at rel_bin.c:5037
13 0x00007fffe8359d44 in rel2bin_project (be=0x7fffb4002b60, rel=0x7fffb4146200, refs=0x7fffb414b840, topn=0x0) at rel_bin.c:2469
14 0x00007fffe836596a in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb4146200, refs=0x7fffb414b840) at rel_bin.c:5053
15 0x00007fffe8365c52 in _subrel_bin (be=0x7fffb4002b60, rel=0x7fffb4146200, refs=0x7fffb414b840) at rel_bin.c:5115
16 0x00007fffe8365d6a in output_rel_bin (be=0x7fffb4002b60, rel=0x7fffb4146200) at rel_bin.c:5138
17 0x00007fffe8382471 in sql_relation2stmt (be=0x7fffb4002b60, r=0x7fffb4146200) at sql_gencode.c:532
18 0x00007fffe8382664 in backend_dumpstmt (be=0x7fffb4002b60, mb=0x7fffb40e8a80, r=0x7fffb4146200, top=1, add_end=1, query=0x7fffb41398a0 "select * from tab0 cor0 where ( col1 ) in ( + 57, + col1 / - col2 );")
at sql_gencode.c:572
19 0x00007fffe8382edc in backend_dumpproc (be=0x7fffb4002b60, c=0x7fffe970d368, cq=0x7fffb411d5d0, r=0x7fffb4146200) at sql_gencode.c:702
20 0x00007fffe8344f71 in SQLparser (c=0x7fffe970d368) at sql_scenario.c:1233
21 0x00007ffff7a0d039 in runPhase (c=0x7fffe970d368, phase=1) at mal_scenario.c:510
22 0x00007ffff7a0d142 in runScenarioBody (c=0x7fffe970d368, once=0) at mal_scenario.c:532
23 0x00007ffff7a0d3fb in runScenario (c=0x7fffe970d368, once=0) at mal_scenario.c:569
24 0x00007ffff7a0f336 in MSserveClient (dummy=0x7fffe970d368) at mal_session.c:514
25 0x00007ffff7a0ee85 in MSscheduleClient (command=0x7fffb4000b30 "0", challenge=0x7fffcb392d9b "9VJwBVfN", fin=0x17f9200, fout=0x7fffc4002d90, protocol=PROTOCOL_9, blocksize=8190, compute_column_widths=0)
at mal_session.c:397
26 0x00007ffff7aaf07e in doChallenge (data=0x17f9210) at mal_mapi.c:288
27 0x00007ffff7694a72 in thread_starter (arg=0x7fffc4004e80) at gdk_system.c:476
28 0x00007ffff4af961b in start_thread (arg=0x7fffcb393700) at pthread_create.c:465
29 0x00007ffff482698f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95
## Comment 26223
Date: 2018-02-19 10:48:03 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [e156ab0c6065](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e156ab0c6065) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=e156ab0c6065](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=e156ab0c6065)
Changeset description:
Added tests for bugs 6545 and 6546
## Comment 26239
Date: 2018-03-01 19:35:56 +0100
From: @njnes
resolved after fixes for bug #6529
## Comment 26306
Date: 2018-03-29 15:39:10 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| Sqlitelogictest crash in IN query with division | https://api.github.com/repos/MonetDB/MonetDB/issues/6546/comments | 0 | 2020-11-30T16:25:16Z | 2024-06-27T13:03:20Z | https://github.com/MonetDB/MonetDB/issues/6546 | 753,618,717 | 6,546 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-19 10:15:31 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-03-29 15:39:13 +0200
## Comment 26218
Date: 2018-02-19 10:15:31 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/64.0.3282.167 Safari/537.36
Build Identifier:
The query bellow crashes in MonetDB with an assertion error in rel_bin.c
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab0(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab0 VALUES (97,1,99), (15,81,47), (87,21,10);
3. SELECT DISTINCT * FROM tab0 AS cor0 WHERE - 60 + ( + - col0 ) IN ( + - CAST ( 35 AS INTEGER ) * + ( - 3 ) * - 91 * + + 87, - col0 * + col2 * 18 );
### Actual Results:
Assertion error in rel_bin.c
### Expected Results:
Empty result set
The assertion error:
mserver5: rel_bin.c:576: exp_bin: Assertion `s' failed
Backtrace:
0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:51
1 0x00007ffff4748381 in __GI_abort () at abort.c:79
2 0x00007ffff473e8fa in __assert_fail_base (fmt=0x7ffff48b9c28 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", assertion=assertion@entry=0x7fffe84cbd58 "s", file=file@entry=0x7fffe84cbce8 "rel_bin.c",
line=line@entry=576, function=function@entry=0x7fffe84cc518 <__PRETTY_FUNCTION__.18615> "exp_bin") at assert.c:92
3 0x00007ffff473e972 in __GI___assert_fail (assertion=0x7fffe84cbd58 "s", file=0x7fffe84cbce8 "rel_bin.c", line=576, function=0x7fffe84cc518 <__PRETTY_FUNCTION__.18615> "exp_bin") at assert.c:101
4 0x00007fffe8350b22 in exp_bin (be=0x7fffb4002b60, e=0x7fffb41f6b50, left=0x7fffb41fed20, right=0x7fffb41ff450, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:576
5 0x00007fffe8359e17 in rel2bin_project (be=0x7fffb4002b60, rel=0x7fffb41f6be0, refs=0x7fffb41fbfe0, topn=0x0) at rel_bin.c:2481
6 0x00007fffe836596a in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb41f6be0, refs=0x7fffb41fbfe0) at rel_bin.c:5053
7 0x00007fffe8359d44 in rel2bin_project (be=0x7fffb4002b60, rel=0x7fffb41fa040, refs=0x7fffb41fbfe0, topn=0x0) at rel_bin.c:2469
8 0x00007fffe836596a in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb41fa040, refs=0x7fffb41fbfe0) at rel_bin.c:5053
9 0x00007fffe8358506 in rel2bin_union (be=0x7fffb4002b60, rel=0x7fffb41f9030, refs=0x7fffb41fbfe0) at rel_bin.c:2160
10 0x00007fffe83658de in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb41f9030, refs=0x7fffb41fbfe0) at rel_bin.c:5041
11 0x00007fffe8357537 in rel2bin_semijoin (be=0x7fffb4002b60, rel=0x7fffb41f51c0, refs=0x7fffb41fbfe0) at rel_bin.c:1939
12 0x00007fffe83658b0 in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb41f51c0, refs=0x7fffb41fbfe0) at rel_bin.c:5037
13 0x00007fffe8359d44 in rel2bin_project (be=0x7fffb4002b60, rel=0x7fffb41f7ac0, refs=0x7fffb41fbfe0, topn=0x0) at rel_bin.c:2469
14 0x00007fffe836596a in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb41f7ac0, refs=0x7fffb41fbfe0) at rel_bin.c:5053
15 0x00007fffe835acd2 in rel2bin_groupby (be=0x7fffb4002b60, rel=0x7fffb41f5260, refs=0x7fffb41fbfe0) at rel_bin.c:2686
16 0x00007fffe83659c6 in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb41f5260, refs=0x7fffb41fbfe0) at rel_bin.c:5061
17 0x00007fffe8365c52 in _subrel_bin (be=0x7fffb4002b60, rel=0x7fffb41f5260, refs=0x7fffb41fbfe0) at rel_bin.c:5115
18 0x00007fffe8365d6a in output_rel_bin (be=0x7fffb4002b60, rel=0x7fffb41f5260) at rel_bin.c:5138
19 0x00007fffe8382471 in sql_relation2stmt (be=0x7fffb4002b60, r=0x7fffb41f5260) at sql_gencode.c:532
20 0x00007fffe8382664 in backend_dumpstmt (be=0x7fffb4002b60, mb=0x7fffb41ea7b0, r=0x7fffb41f5260, top=1, add_end=0,
query=0x7fffb423b3c0 "select distinct * from tab0 as cor0 where - 60 + ( + - col0 ) in ( + - cast ( 35 as integer ) * + ( - 3 ) * - 91 * + + 87, - col0 * + col2 * 18 );") at sql_gencode.c:572
21 0x00007fffe8344bd8 in SQLparser (c=0x7fffe970d368) at sql_scenario.c:1195
22 0x00007ffff7a0d039 in runPhase (c=0x7fffe970d368, phase=1) at mal_scenario.c:510
23 0x00007ffff7a0d142 in runScenarioBody (c=0x7fffe970d368, once=0) at mal_scenario.c:532
24 0x00007ffff7a0d3fb in runScenario (c=0x7fffe970d368, once=0) at mal_scenario.c:569
25 0x00007ffff7a0f336 in MSserveClient (dummy=0x7fffe970d368) at mal_session.c:514
26 0x00007ffff7a0ee85 in MSscheduleClient (command=0x7fffb4000b30 "0+", challenge=0x7fffcb392d9b "41Q48ySCa", fin=0x7fffc4000ca0, fout=0x7fffc4002d90, protocol=PROTOCOL_9, blocksize=8190,
compute_column_widths=0) at mal_session.c:397
27 0x00007ffff7aaf07e in doChallenge (data=0x7fffc4000cb0) at mal_mapi.c:288
28 0x00007ffff7694a72 in thread_starter (arg=0x7fffc4000ce0) at gdk_system.c:476
29 0x00007ffff4af961b in start_thread (arg=0x7fffcb393700) at pthread_create.c:465
30 0x00007ffff482698f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95
## Comment 26220
Date: 2018-02-19 10:43:48 +0100
From: @PedroTadim
This crash only happens when I compile MonetDB without hugeint support.
## Comment 26222
Date: 2018-02-19 10:48:02 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [e156ab0c6065](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e156ab0c6065) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=e156ab0c6065](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=e156ab0c6065)
Changeset description:
Added tests for bugs 6545 and 6546
## Comment 26238
Date: 2018-03-01 19:34:57 +0100
From: @njnes
resolved after fixes for bug #6529
## Comment 26310
Date: 2018-03-29 15:39:13 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| Sqlitelogictest crash in IN query | https://api.github.com/repos/MonetDB/MonetDB/issues/6545/comments | 0 | 2020-11-30T16:25:11Z | 2024-06-27T13:03:19Z | https://github.com/MonetDB/MonetDB/issues/6545 | 753,618,656 | 6,545 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-15 18:26:35 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.27.13 (Jul2017-SP4)
Last updated: 2018-03-08 16:08:09 +0100
## Comment 26203
Date: 2018-02-15 18:26:35 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Fedora; Linux x86_64; rv:57.0) Gecko/20100101 Firefox/57.0
Build Identifier:
Table bam.files has 0 rows.
When running command:
TRUNCATE TABLE bam.files;
it returns with an error:
42000: !TRUNCATE: FOREIGN KEY sq.sq_fkey_file_id depends on files
As the table has 0 rows, the truncate table command does not have to do anything and therefore should succeed without an error, even if it is referenced via an fkey.
So the internal code should check the BATcount() value first, before doing fk checks.
Reproducible: Always
### Steps to Reproduce:
-- this requires the SAMTools are available, loaded and the bam schema and tables are created during initialisation of a new database
SELECT COUNT(*) AS count_rows FROM bam.files;
TRUNCATE TABLE bam.files;
SELECT COUNT(*) AS count_rows FROM bam.files;
### Actual Results:
no 42000: !TRUNCATE: FOREIGN KEY sq.sq_fkey_file_id depends on files
error message.
### Expected Results:
no fk error message when table has 0 rows
## Comment 26206
Date: 2018-02-15 19:00:11 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [5ac9ac5e4878](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=5ac9ac5e4878) made by Martin van Dinther <martin.van.dinther@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=5ac9ac5e4878](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=5ac9ac5e4878)
Changeset description:
Add test for bug #6544
## Comment 26207
Date: 2018-02-16 13:22:56 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [d5dd594db6c5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d5dd594db6c5) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=d5dd594db6c5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=d5dd594db6c5)
Changeset description:
Fixes for bugs 6543 and 6544
## Comment 26246
Date: 2018-03-08 16:08:09 +0100
From: @sjoerdmullender
Seems to be working.
Resolve as fixed since this is a new feature.
| Mar2018: truncating a table of 0 rows should be successful even if it has a fk references to it | https://api.github.com/repos/MonetDB/MonetDB/issues/6544/comments | 0 | 2020-11-30T16:25:09Z | 2024-06-27T13:03:18Z | https://github.com/MonetDB/MonetDB/issues/6544 | 753,618,625 | 6,544 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-15 17:17:23 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.27.13 (Jul2017-SP4)
CC: @njnes
Last updated: 2018-03-29 15:39:27 +0200
## Comment 26202
Date: 2018-02-15 17:17:23 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Fedora; Linux x86_64; rv:57.0) Gecko/20100101 Firefox/57.0
Build Identifier:
This issue applies to Mar2018 release, which supports truncate table statements.
It appears that truncating system tables (such as sys._tables, tmp._tables, sys.schemas) is possible and really removes the data in those tables.
This corrupts the MonetDB server internal system and eventually leads to a segmentation fault.
Truncating data from system tables (all tables listed by: SELECT s.name||'.'||t.name as qname FROM sys.tables t join sys.schemas s ON t.schema_id = s.id WHERE t."system" = true AND t.query IS NULL AND s.name IN ('sys', 'tmp') AND t.name NOT LIKE 'netcdf_%' AND t.name <> 'statistics') should be disallowed.
It should return error: "Truncate not allowed on system table %s.%s".
Reproducible: Always
### Steps to Reproduce:
SELECT (COUNT(*) > 0) AS has_rows FROM tmp.triggers;
TRUNCATE TABLE tmp.triggers;
SELECT (COUNT(*) > 0) AS has_rows FROM tmp.triggers;
SELECT (COUNT(*) > 0) AS has_rows FROM tmp.objects;
TRUNCATE TABLE tmp.objects;
SELECT (COUNT(*) > 0) AS has_rows FROM tmp.objects;
SELECT (COUNT(*) > 0) AS has_rows FROM tmp.keys;
TRUNCATE TABLE tmp.keys;
SELECT (COUNT(*) > 0) AS has_rows FROM tmp.keys;
SELECT (COUNT(*) > 0) AS has_rows FROM tmp.idxs;
TRUNCATE TABLE tmp.idxs;
SELECT (COUNT(*) > 0) AS has_rows FROM tmp.idxs;
SELECT (COUNT(*) > 0) AS has_rows FROM tmp._tables;
TRUNCATE TABLE tmp._tables;
SELECT (COUNT(*) > 0) AS has_rows FROM tmp._tables;
SELECT (COUNT(*) > 0) AS has_rows FROM tmp._columns;
TRUNCATE TABLE tmp._columns;
SELECT (COUNT(*) > 0) AS has_rows FROM tmp._columns;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.user_role;
TRUNCATE TABLE sys.user_role;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.user_role;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.types;
TRUNCATE TABLE sys.types;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.types;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.triggers;
TRUNCATE TABLE sys.triggers;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.triggers;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.table_types;
TRUNCATE TABLE sys.table_types;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.table_types;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.systemfunctions;
TRUNCATE TABLE sys.systemfunctions;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.systemfunctions;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.storagemodelinput;
TRUNCATE TABLE sys.storagemodelinput;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.storagemodelinput;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.statistics;
TRUNCATE TABLE sys.statistics;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.statistics;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.spatial_ref_sys;
TRUNCATE TABLE sys.spatial_ref_sys;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.spatial_ref_sys;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.sequences;
TRUNCATE TABLE sys.sequences;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.sequences;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.schemas;
TRUNCATE TABLE sys.schemas;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.schemas;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.privileges;
TRUNCATE TABLE sys.privileges;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.privileges;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.privilege_codes;
TRUNCATE TABLE sys.privilege_codes;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.privilege_codes;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.objects;
TRUNCATE TABLE sys.objects;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.objects;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.keywords;
TRUNCATE TABLE sys.keywords;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.keywords;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.keys;
TRUNCATE TABLE sys.keys;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.keys;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.key_types;
TRUNCATE TABLE sys.key_types;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.key_types;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.index_types;
TRUNCATE TABLE sys.index_types;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.index_types;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.idxs;
TRUNCATE TABLE sys.idxs;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.idxs;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.functions;
TRUNCATE TABLE sys.functions;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.functions;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.function_types;
TRUNCATE TABLE sys.function_types;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.function_types;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.function_languages;
TRUNCATE TABLE sys.function_languages;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.function_languages;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.dependency_types;
TRUNCATE TABLE sys.dependency_types;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.dependency_types;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.dependencies;
TRUNCATE TABLE sys.dependencies;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.dependencies;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.db_user_info;
TRUNCATE TABLE sys.db_user_info;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.db_user_info;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.comments;
TRUNCATE TABLE sys.comments;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.comments;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.auths;
TRUNCATE TABLE sys.auths;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.auths;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.args;
TRUNCATE TABLE sys.args;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.args;
SELECT (COUNT(*) > 0) AS has_rows FROM sys._tables;
TRUNCATE TABLE sys._tables;
SELECT (COUNT(*) > 0) AS has_rows FROM sys._tables;
SELECT (COUNT(*) > 0) AS has_rows FROM sys._columns;
TRUNCATE TABLE sys._columns;
SELECT (COUNT(*) > 0) AS has_rows FROM sys._columns;
-- the next truncates should be allowed, as they are extension tables and not system tables
SELECT (COUNT(*) > 0) AS has_rows FROM sys.netcdf_vars;
TRUNCATE TABLE sys.netcdf_vars;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.netcdf_vars;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.netcdf_vardim;
TRUNCATE TABLE sys.netcdf_vardim;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.netcdf_vardim;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.netcdf_files;
TRUNCATE TABLE sys.netcdf_files;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.netcdf_files;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.netcdf_dims;
TRUNCATE TABLE sys.netcdf_dims;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.netcdf_dims;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.netcdf_attrs;
TRUNCATE TABLE sys.netcdf_attrs;
SELECT (COUNT(*) > 0) AS has_rows FROM sys.netcdf_attrs;
-- the next truncates on bam tables should be allowed, as they are extension tables and not system tables
SELECT (COUNT(*) > 0) AS has_rows FROM bam.sq;
TRUNCATE TABLE bam.sq;
SELECT (COUNT(*) > 0) AS has_rows FROM bam.sq;
SELECT (COUNT(*) > 0) AS has_rows FROM bam.rg;
TRUNCATE TABLE bam.rg;
SELECT (COUNT(*) > 0) AS has_rows FROM bam.rg;
SELECT (COUNT(*) > 0) AS has_rows FROM bam.pg;
TRUNCATE TABLE bam.pg;
SELECT (COUNT(*) > 0) AS has_rows FROM bam.pg;
SELECT (COUNT(*) > 0) AS has_rows FROM bam.files;
TRUNCATE TABLE bam.files;
SELECT (COUNT(*) > 0) AS has_rows FROM bam.files;
SELECT (COUNT(*) > 0) AS has_rows FROM bam.export;
TRUNCATE TABLE bam.export;
SELECT (COUNT(*) > 0) AS has_rows FROM bam.export;
### Actual Results:
Truncate data of system tables is executed without error, except for the tables with static content (which are created as read only).
### Expected Results:
error for all truncate statements, except for truncate table sys.statistics;
Note: it should also be possible to truncate sys.netcdf_% tables (an optional extension) and tables in non system schemas such as bam (an optional extension)
## Comment 26204
Date: 2018-02-15 18:31:09 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [a590d416b444](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=a590d416b444) made by Martin van Dinther <martin.van.dinther@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=a590d416b444](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=a590d416b444)
Changeset description:
Adding tests for bug #6543
## Comment 26208
Date: 2018-02-16 13:22:57 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [d5dd594db6c5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d5dd594db6c5) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=d5dd594db6c5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=d5dd594db6c5)
Changeset description:
Fixes for bugs 6543 and 6544
## Comment 26229
Date: 2018-02-22 17:56:28 +0100
From: @sjoerdmullender
I don't agree with the assessment here.
If you are the "superuser" (monetdb) you should be allowed to shoot yourself in your foot. And in fact, you can using DELETE FROM table. So, the check whether TRUNCATE is allowed should be based solely on whether you have been GRANTed the privilege. By default, a "normal" user (i.e. not monetdb) is not allowed to DELETE FROM or TRUNCATE a system table, but they are allowed to DELETE FROM or TRUNCATE a table in the tmp schema.
It is artificial to disallow TRUNCATE on a system table when you do allow DELETE FROM. Therefore, I will undo this particular change.
## Comment 26230
Date: 2018-02-22 17:57:34 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [3067eb415093](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=3067eb415093) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=3067eb415093](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=3067eb415093)
Changeset description:
Allow the superuser to shoot him/herself in the foot.
See bug #6543, comment 3.
## Comment 26274
Date: 2018-03-22 18:04:43 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [3942bf20e695](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=3942bf20e695) made by Martin van Dinther <martin.van.dinther@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=3942bf20e695](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=3942bf20e695)
Changeset description:
Split the truncation of the system core tables in schema tmp from the script which truncates the system core tables in schema sys.
The behavior in schema sys (returns an error for testuser) is different for schema tmp (does NOT return an error for testuser).
The truncate_tmp_tables.Bug-6543.stable.* files contain the desired output.
## Comment 26284
Date: 2018-03-24 11:57:14 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [599b73a8b007](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=599b73a8b007) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=599b73a8b007](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=599b73a8b007)
Changeset description:
more fixes for bug #6543, ie disallow direct updates on system temporaries
## Comment 26329
Date: 2018-03-29 15:39:27 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| Mar2018: truncate on SQL system tables should NOT be allowed | https://api.github.com/repos/MonetDB/MonetDB/issues/6543/comments | 0 | 2020-11-30T16:25:05Z | 2024-06-27T13:03:17Z | https://github.com/MonetDB/MonetDB/issues/6543 | 753,618,579 | 6,543 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-15 16:27:47 +0100
From: Martin van Dinther <<martin.van.dinther>>
To: SQL devs <<bugs-sql>>
Version: 11.27.13 (Jul2017-SP4)
CC: @njnes
Last updated: 2018-03-29 15:39:00 +0200
## Comment 26201
Date: 2018-02-15 16:27:47 +0100
From: Martin van Dinther <<martin.van.dinther>>
User-Agent: Mozilla/5.0 (X11; Fedora; Linux x86_64; rv:57.0) Gecko/20100101 Firefox/57.0
Build Identifier:
SQL query: select count(*) from sys.commented_function_signatures;
produces assertion failure:
mserver5: ../../../../dev/sql/backends/monet5/rel_bin.c:575: exp_bin: Assertion `s' failed.
Occurs on Mar2018 release, but potentially also on Jul2017 release.
Note the view sys.commented_function_signatures is introduced in Mar2018 release, see https://dev.monetdb.org/hg/MonetDB/file/Mar2018/sql/scripts/97_comments.sql
Reproducible: Always
### Steps to Reproduce:
1. The view sys.commented_function_signatures is introduced in March2018 release
2. so compile March2018 release with assertions enabled
3. run query: select count(*) from sys.commented_function_signatures;
### Actual Results:
bash-4.4$ ./start_INSTALL_mserver5.sh
builtin opt gdk_dbpath = /export/scratch1/dinther/INSTALL/var/monetdb5/dbfarm/demo
builtin opt gdk_debug = 0
builtin opt gdk_vmtrim = no
builtin opt monet_prompt = >
builtin opt monet_daemon = no
builtin opt mapi_port = 50000
builtin opt mapi_open = false
builtin opt mapi_autosense = false
builtin opt sql_optimizer = default_pipe
builtin opt sql_debug = 0
cmdline opt embedded_r = true
cmdline opt embedded_py = true
cmdline opt mapi_port = 41000
cmdline opt gdk_debug = 10
MonetDB 5 server v11.29.0
This is an unreleased version
Serving database 'demo', using 4 threads
Compiled for x86_64-unknown-linux-gnu/64bit with 128bit integers
Found 31.307 GiB available main-memory.
Copyright (c) 1993 - July 2008 CWI.
Copyright (c) August 2008 - 2018 MonetDB B.V., all rights reserved
Visit https://www.monetdb.org/ for further information
Listening for connection requests on mapi:monetdb://127.0.0.1:41000/
MonetDB/GIS module loaded
MonetDB/SQL module loaded
MonetDB/Python2 module loaded
MonetDB/R module loaded
>could not find L37.L37
f.id
L2.fid
mserver5: ../../../../dev/sql/backends/monet5/rel_bin.c:575: exp_bin: Assertion `s' failed.
./start_INSTALL_mserver5.sh: line 7: 3531 Aborted (core dumped) /export/scratch1/dinther/INSTALL/bin/mserver5 -d10 --set embedded_r=true --set embedded_py=true --set mapi_port=41000
bash-4.4$
### Expected Results:
no assertion failure.
a normal count(*) result returned.
## Comment 26205
Date: 2018-02-15 18:45:31 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [d3b3b1e73871](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d3b3b1e73871) made by Martin van Dinther <martin.van.dinther@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=d3b3b1e73871](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=d3b3b1e73871)
Changeset description:
Add test for bug #6542
## Comment 26297
Date: 2018-03-29 15:39:00 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| assertion failure when querying: select count(*) from sys.commented_function_signatures; | https://api.github.com/repos/MonetDB/MonetDB/issues/6542/comments | 0 | 2020-11-30T16:25:02Z | 2024-06-27T13:03:16Z | https://github.com/MonetDB/MonetDB/issues/6542 | 753,618,536 | 6,542 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-13 15:49:49 +0100
From: @drstmane
To: clients devs <<bugs-clients>>
Version: 11.27.13 (Jul2017-SP4)
Last updated: 2018-03-29 15:39:12 +0200
## Comment 26190
Date: 2018-02-13 15:49:49 +0100
From: @drstmane
$ echo -e ' select 0; \n select avg(value) from sys.generate_series(0,1000000000); \n select 0; '
select 0;
select avg(value) from sys.generate_series(0,1000000000);
select 0;
$ echo -e ' select 0; \n select avg(value) from sys.generate_series(0,1000000000); \n select 0; ' | mclient
+------+
| L2 |
+======+
| 0 |
+------+
1 tuple
clk: 5.00 sec
+--------------------------+
| L6 |
+==========================+
| 499999999.5 |
+--------------------------+
1 tuple
clk: 5.00 sec
+------+
| L2 |
+======+
| 0 |
+------+
1 tuple
clk: 5.00 sec
## Comment 26191
Date: 2018-02-13 16:05:41 +0100
From: @drstmane
The problem is the very problem that was the reason with mclient reported (per query) timing only in "interactve" (or "echo") mode, i.e., only when mclient sends individual queries (or at least lines) to the server, and thus indeed measures the wall-clock time for individual queries, rather than sending the entire input / file, and this measuring the wall-clock time for the entire batch.
Without "--interative" or "--echo", mclient now reports per query the time it took to execute the entire batch of queries sent to the server.
To ensure that the mclient indeed report the wall-clock time for what it actually measures, mclient should
(1) either send individual queries to the server (much like "--interactive", which enabled interpretation of '\'-commands, and "--echo", which echos each query sent to the server) when the timermode is other than none,
(2) or report wall-clock time (only) for what is indeed measure, i.e., at the end of an entire batch of queries sent to the server in "batch-mode"; and per query only if mclient indeed sends individual queries to the server, e.g., when --interactive or --echo are used.
## Comment 26192
Date: 2018-02-14 09:56:31 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [9e804389aee9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9e804389aee9) made by Stefan Manegold <Stefan.Manegold@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=9e804389aee9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=9e804389aee9)
Changeset description:
mclient: report only the wall-clock times we do indeed measure:
when mclient sends individual queries (lines) to the server
in interactive mode, report wall-clock per query;
otherwise, report (only) total wall-clock for the entire
batch sent, executed, and measured.
This should fix bug #6541.
## Comment 26195
Date: 2018-02-14 10:15:08 +0100
From: @drstmane
Fixed by changeset [9e804389aee9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9e804389aee9) .
## Comment 26307
Date: 2018-03-29 15:39:12 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| [Mar2018]: mclient reports incorrect wall-clock time | https://api.github.com/repos/MonetDB/MonetDB/issues/6541/comments | 0 | 2020-11-30T16:24:58Z | 2024-06-27T13:03:15Z | https://github.com/MonetDB/MonetDB/issues/6541 | 753,618,493 | 6,541 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-12 12:33:45 +0100
From: Gaetan Butez <<gaetan.butez>>
To: clients devs <<bugs-clients>>
Version: 11.27.11 (Jul2017-SP3)
CC: renalkoclok
Last updated: 2020-06-10 16:49:31 +0200
## Comment 26189
Date: 2018-02-12 12:33:45 +0100
From: Gaetan Butez <<gaetan.butez>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36
Build Identifier:
Microsoft Access is returning the error : "Reserved error (-7731); there is no message for this error." when trying to link a table with the External Table Wizard after selecting the table that I wanted to link.
More over, Query Pass-Through works fine.
Software environment :
- Microsoft Access 2013 (32 bits)
- MonetDB Server (32 bits)
- MonetDB ODBC (32 bits + 64 bits both installed)
- Windows 7 (64 bits)
Reproducible: Always
### Steps to Reproduce:
1. Open a Microsoft Access Database
2. Open "Get External Data - ODBC Database" in External Data ribbon
3. Select "Link to the data source by creating a linked table."
4. Select MonetDB in Machine Data Source (default MonetDB DSN created by ODBC driver installer)
5. Select the table that you want to import in the "Link Tables Window" and click "OK" button. (checking "Save password" or not does not have an impact on the results)
6. MS Access Reserved Error (-7731) shows and doesn't link the table in the database
### Actual Results:
External table is not linked in the MS Access database
### Expected Results:
MS Access should have linked the table as a Linked External Table.
When selecting "Import the source data into a new table in the current database" during step 3., MS Access shows the same error and in the "Save failures" created table, the column "Failure Reason" shows : "Could not open the object". Maybe the ODBC driver opens the table as "Read only" and should be as "Read/Write" or without restrictions of some sort.
Thank you very much in advance for you help as I'm trying to use MonetDB as my OLAP DB Engine and MS Access as my UI.
## Comment 26209
Date: 2018-02-16 14:29:19 +0100
From: @sjoerdmullender
I don't have access installed, so I can't (easily) debug.
Can you please provide log information? You can get log information by starting your ODBC using application with an extra environment variable whose value should be the name of a file into which the log is to be written (i.e. it must be writable by the application). The name of the environment variable that you need to set for this is ODBCDEBUG.
Do check the log file for sensitive information.
## Comment 26212
Date: 2018-02-17 19:03:04 +0100
From: Gaetan Butez <<gaetan.butez>>
Created attachment 597
SQL LOG
> Attached file: [SQL.LOG](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_6540_SQL.LOG_597) (text/plain, 110023 bytes)
> Description: SQL LOG
## Comment 26213
Date: 2018-02-17 19:03:48 +0100
From: Gaetan Butez <<gaetan.butez>>
Created attachment 598
SQL LOG
> Attached file: [SQL.LOG](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_6540_SQL.LOG_598) (text/plain, 110023 bytes)
> Description: SQL LOG
## Comment 26214
Date: 2018-02-17 19:04:22 +0100
From: Gaetan Butez <<gaetan.butez>>
Created attachment 599
SQL LOG
> Attached file: [SQL.LOG](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_6540_SQL.LOG_599) (text/plain, 110023 bytes)
> Description: SQL LOG
## Comment 26215
Date: 2018-02-17 19:06:35 +0100
From: Gaetan Butez <<gaetan.butez>>
Hi Sjoerd,
Thanks for your help on this issue.
You'll find attached the log for the ODBC driver.
This log was created on my home computer with the following environnment and having the same error shown in MS Access:
- Win 10 Pro (64 bits)
- MS Access 2016 (64 bits)
- MonetDB Server (64 bits - latest)
- MonetDB ODBC Driver (64 bits - latest)
Best regards,
Gaëtan
## Comment 26224
Date: 2018-02-19 11:00:18 +0100
From: @sjoerdmullender
Were you by any chance trying to link to a table called MSysConf?
If so, the problem is in my opinion firmly on the side of the application. In SQL, table and column names are case insensitive, unless the names are quoted. In MonetDB, you need to quote the names with double quotes (as the ODBC driver would tell you if you (or your application) were to ask with SQLGetInfo with parameter SQL_IDENTIFIER_QUOTE_CHAR).
It looks like the name of the table is MSysConf, but since the name is not quoted in the query that the application uses (SELECT Config, nValue FROM MSysConf), the server translates the name to msysconf, which it then can't find.
The same problem would occur with the column names: they too need to be quoted.
## Comment 26228
Date: 2018-02-20 13:54:42 +0100
From: Gaetan Butez <<gaetan.butez>>
Hi Sjoerd,
The table I was trying to link is : "testtable" in the schema "sys".
Actually, I found out that the MSysConf is a parameter table that a DBA create in a database that might be linked in MS Access. MSysConf specifies to MS Access if the user has the rights to save the password locally on his machine or not. This functionality is described here by MS : https://support.microsoft.com/en-us/help/209502/how-to-store-sql-database-login-ids-and-passwords-locally.
After implementing the MsysConf table with the parameters defined in the MS article, the MsysConf error doesn't show in ODBC log. However, linking table in Ms Access still doesn't work.
Could it come from the "EXIT SQLGetStmtAttrW with return code -1 (SQL_ERROR) [...] DIAG [HYC00] [MonetDB][ODBC Driver 11.27.11]Optional feature not implemented (0)" error ?
Thanks again for your help.
Regards,
Gaëtan
## Comment 26518
Date: 2018-06-27 16:54:12 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [70f093ad98de](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=70f093ad98de) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=70f093ad98de](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=70f093ad98de)
Changeset description:
Always return 0 for SQL_ATTR_QUERY_TIMEOUT (no timeout).
This is one of the errors reported in the log in bug #6540.
## Comment 26519
Date: 2018-06-27 16:54:16 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [74e643111c76](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=74e643111c76) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=74e643111c76](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=74e643111c76)
Changeset description:
Allow setting of SQL_ATTR_MAX_LENGTH and SQL_ATTR_MAX_ROW.
(but only to 0 and 2147483647 (2**31 - 1).)
This is one of the errors reported in the log in bug #6540.
## Comment 26520
Date: 2018-06-28 09:41:56 +0200
From: @sjoerdmullender
I looked at the latest log and saw errors corresponding to the table MSysConf which you say you fixed, and to attributes SQL_ATTR_QUERY_TIMEOUT (getting the value) and SQL_ATTR_MAX_LENGTH (setting the value). I implemented those attributes (the timeout is 0, meaning no timeout, and max length can only be set to 0 or 2147483647, meaning we always get the complete data).
Can you try the nightly build to see whether that made a difference? You can find it at http://monetdb.cwi.nl/testweb/web/66773:74e643111c76/ .
## Comment 26535
Date: 2018-07-08 16:07:10 +0200
From: Gaetan Butez <<gaetan.butez>>
Created attachment 604
SQL Log Mysql Working
Works even if Msysconf is not present in the database
> Attached file: [SQL - MySql 080718.LOG](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_6540_SQL_-_MySql_080718.LOG_604) (text/plain, 60779 bytes)
> Description: SQL Log Mysql Working
## Comment 26536
Date: 2018-07-08 16:07:54 +0200
From: Gaetan Butez <<gaetan.butez>>
Created attachment 605
Sql Log MonetDB with odbc nightbuild
> Attached file: [SQL - MonetDB 080718.LOG](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_6540_SQL_-_MonetDB_080718.LOG_605) (text/plain, 117742 bytes)
> Description: Sql Log MonetDB with odbc nightbuild
## Comment 26537
Date: 2018-07-08 16:08:25 +0200
From: Gaetan Butez <<gaetan.butez>>
Hi Sjoerd
I tried with the new odbc drivers from the nightbuild but it s still not working. You ll find attached two drivers tracing, one from MySQL odbc that is working when I want to link a table to access and one from MonetDB odbc.
As you can see, MSysconf table is requested by MySql odbc without being supplied and the drivers works.
Do I need also to install the nightbuild of the monetdb server or is it just comming from the odbc drivers ?
Thanks again in advance for your help.
Best regards,
Gaëtan Butez
## Comment 26568
Date: 2018-07-24 10:54:30 +0200
From: @sjoerdmullender
I see a bunch of differences between the MySQL and MonetDB logs, which is to be expected. For instance, the result of the call to SQLTablesW is quite different, but that should be ok.
What I find strange is that the call to SQLColumnsW is so different. In both cases, the column information of the test table (sys.test in MonetDB, table1 in MySQL) is requested. For MySQL, the call returns information about two columns, id and data. For MonetDB there is only a single column, id. And stranger still is that after this call and fetching the result, something that seemingly succeeds, the MySQL log continues, but the MonetDB log just stops with a disconnect.
I don't see why that would be. I see no errors in that area.
Were the two test tables the same (both with two columns)? What was the schema of the tables?
## Comment 27763
Date: 2020-06-03 09:09:14 +0200
From: @sjoerdmullender
The content of attachment 664 has been deleted for the following reason:
spam
## Comment 27817
Date: 2020-06-10 16:49:31 +0200
From: Gaetan Butez <<gaetan.butez>>
Found a better solution by using R and MonetDBLite instead of MS Access
| Can't link MonetDB table on MS Access 2013 | https://api.github.com/repos/MonetDB/MonetDB/issues/6540/comments | 0 | 2020-11-30T16:24:54Z | 2024-06-28T06:49:05Z | https://github.com/MonetDB/MonetDB/issues/6540 | 753,618,439 | 6,540 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-09 17:10:26 +0100
From: Manuel <<manuel>>
To: SQL devs <<bugs-sql>>
Version: 11.27.11 (Jul2017-SP3)
CC: @njnes
Last updated: 2018-03-02 12:52:52 +0100
## Comment 26184
Date: 2018-02-09 17:10:26 +0100
From: Manuel <<manuel>>
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/64.0.3282.140 Safari/537.36
Build Identifier:
I have a query involving the "NOT EXISTS" clause, as well as various joins and unions.
The complete query can be found in the file "full_query.sql" attached.
For sake of simplicity I created two views extracted from the query called "t9c" and "t9d".
The table used in the query can be imported via the file "bp.sql" also attached.
These are the content of the two views:
select * from "t9d" order by "c8v__y_g_0", "c8w__y_g_1" ;
c8v__y_g_0 c8w__y_g_1
---------------------------------
Collateralised Liq Cat (sum)
Collateralised Record count
Senior Unsecured Liq Cat (sum)
Senior Unsecured Record count
Subordinated Liq Cat (sum)
Subordinated Record count
Supranational Liq Cat (sum)
Supranational Record count
select * from "t9c" order by "c8l__y_g_0" , "c8m__y_g_1" ;
c8l__y_g_0 c8m__y_g_1 c8n_axis_1_sort_0_1
------------------------------------------------------
Collateralised Liq Cat (sum) 3
Collateralised Record count 3
Senior Unsecured Liq Cat (sum) 7
Senior Unsecured Record count 7
Subordinated Liq Cat (sum) 3
Subordinated Record count 3
Supranational Liq Cat (sum) 2
Supranational Record count 2
Once the views has been created, I expect the query
SELECT "t9d"."c8v__y_g_0" AS "c99__y_g_0",
"t9d"."c8w__y_g_1" AS "c9a__y_g_1",
CAST(NULL AS bigint) AS "c9b_axis_1_sort_0_1"
FROM "t9d"
WHERE
NOT EXISTS
(
SELECT
1 AS "one"
FROM
"t9c"
WHERE (
( "t9d"."c8v__y_g_0" = "t9c"."c8l__y_g_0"
OR
("t9d"."c8v__y_g_0" IS NULL AND "t9c"."c8l__y_g_0" IS NULL)
)
AND
( "t9d"."c8w__y_g_1" = "t9c"."c8m__y_g_1"
OR
("t9d"."c8w__y_g_1" IS NULL AND "t9c"."c8m__y_g_1" IS NULL)
)
)
)
to return no result (each pair (*g_0, *g_1) is contained in both tables).
However I get the records:
c99__y_g_0 c9a__y_g_1 c9b_axis_1_sort_0_1
------------------------------------------------------
Subordinated Liq Cat (sum) <null>
Supranational Liq Cat (sum) <null>
Senior Unsecured Liq Cat (sum) <null>
Collateralised Liq Cat (sum) <null>
Removing the condition on null values solves the issue and the result is empty as expected.
While investigating on this issue I found other queries returning the wrong result:
select * from "t9d", "t9c"
WHERE
(("t9d"."c8v__y_g_0" = "t9c"."c8l__y_g_0" OR ("t9d"."c8v__y_g_0" IS NULL AND "t9c"."c8l__y_g_0" IS NULL) ) AND ("t9d"."c8w__y_g_1" = "t9c"."c8m__y_g_1" OR ("t9c"."c8m__y_g_1" IS NULL AND "t9d"."c8w__y_g_1" IS NULL)))
select * from "t9d", "t9c"
WHERE
(("t9d"."c8v__y_g_0" = "t9c"."c8l__y_g_0" OR (1 = 2) ) AND ("t9d"."c8w__y_g_1" = "t9c"."c8m__y_g_1" OR ("t9c"."c8m__y_g_1" IS NULL AND "t9d"."c8w__y_g_1" IS NULL)))
select * from "t9d", "t9c"
WHERE
(not("t9d"."c8v__y_g_0" <> "t9c"."c8l__y_g_0" AND (1 = 1)) AND not("t9d"."c8w__y_g_1" <> "t9c"."c8m__y_g_1" and (1 = 1)))
Reproducible: Always
### Steps to Reproduce:
1. Import the table contained in "bp.sql" attached.
2. run the query contained in "full_query.sql", or create the views in create_view_t9c.sql create_view_t9d.sql and use the query described in the Details section
3.
### Actual Results:
c99__y_g_0 c9a__y_g_1 c9b_axis_1_sort_0_1
------------------------------------------------------
Subordinated Liq Cat (sum) <null>
Supranational Liq Cat (sum) <null>
Senior Unsecured Liq Cat (sum) <null>
Collateralised Liq Cat (sum) <null>
### Expected Results:
No records.
## Comment 26185
Date: 2018-02-09 17:11:05 +0100
From: Manuel <<manuel>>
Created attachment 593
bp.sql: the table used in the query
> Attached file: [bp.sql](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_6539_bp.sql_593) (text/plain, 244818 bytes)
> Description: bp.sql: the table used in the query
## Comment 26186
Date: 2018-02-09 17:12:32 +0100
From: Manuel <<manuel>>
Created attachment 594
full_query.sql: the full query exposing the issue.
> Attached file: [full_query.sql](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_6539_full_query.sql_594) (text/plain, 12659 bytes)
> Description: full_query.sql: the full query exposing the issue.
## Comment 26187
Date: 2018-02-09 17:13:33 +0100
From: Manuel <<manuel>>
Created attachment 595
create_view_t9c.sql: one of the view used in the query described in the details section
> Attached file: [create_view_t9c.sql](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_6539_create_view_t9c.sql_595) (text/plain, 8902 bytes)
> Description: create_view_t9c.sql: one of the view used in the query described in the details section
## Comment 26188
Date: 2018-02-09 17:13:50 +0100
From: Manuel <<manuel>>
Created attachment 596
create_view_t9d.sql: one of the view used in the query described in the details section
> Attached file: [create_view_t9d.sql](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_6539_create_view_t9d.sql_596) (text/plain, 2925 bytes)
> Description: create_view_t9d.sql: one of the view used in the query described in the details section
## Comment 26240
Date: 2018-03-02 12:52:52 +0100
From: @njnes
seems solved
| Query with not joins and condition returning the wrong result | https://api.github.com/repos/MonetDB/MonetDB/issues/6539/comments | 0 | 2020-11-30T16:24:51Z | 2024-06-27T13:03:12Z | https://github.com/MonetDB/MonetDB/issues/6539 | 753,618,409 | 6,539 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-08 16:14:29 +0100
From: @drstmane
To: GDK devs <<bugs-common>>
Version: -- development
Last updated: 2018-02-08 16:14:29 +0100
## Comment 26179
Date: 2018-02-08 16:14:29 +0100
From: @drstmane
Having a DBextra for transient storage separate from DBfarm for persistent storage is a great feature!
A very common and beneficial scenario is to have the transient DBextra on the fastest storage (e.g., NVMe of at least SSD), while the persistent DBfarm is on an "ordinary" disk RAID / JBOD / ...
However, the capacity of the fastest flash storage is usually considerably smaller than that of the disk storage.
In such cases, queries that process large amounts of data, and thus need to allocate large intermediate results, might fail given that transient intermediate result allocation is limited by the fast flash storage capacity.
One way to avoid query failures in such cases could be to have the allocation for transient intermediate results fall back to DBfarm in case allocation on DBextra fails. This could be either an automatics and transparent default, or an explicit configuration option of the server.
Thanks!
| transient BAT/COL/Heap-allocation fall-back to DBfarm in case DBextra fails | https://api.github.com/repos/MonetDB/MonetDB/issues/6538/comments | 0 | 2020-11-30T16:24:49Z | 2024-06-28T13:39:43Z | https://github.com/MonetDB/MonetDB/issues/6538 | 753,618,375 | 6,538 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-08 09:40:36 +0100
From: @drstmane
To: clients devs <<bugs-clients>>
Version: 11.27.11 (Jul2017-SP3)
Last updated: 2018-03-29 15:39:09 +0200
## Comment 26177
Date: 2018-02-08 09:40:36 +0100
From: @drstmane
as an alternative (or addition?) to feature request / bug #6536:
mclient's default timing mode should (again) be "none" rather than "clock"
much like feature request / bug #6536 this is mainly to avoid that the default timing output compromises/invalidates the format of mclient's output formatters csv, tab, xml.
## Comment 26193
Date: 2018-02-14 09:56:42 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [01645d98a71b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=01645d98a71b) made by Stefan Manegold <Stefan.Manegold@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=01645d98a71b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=01645d98a71b)
Changeset description:
mclient: make timermode "none" the default (again)
as per feature request / bug #6537
## Comment 26197
Date: 2018-02-14 10:17:30 +0100
From: @drstmane
Fixed / implemented by changeset [01645d98a71b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=01645d98a71b) .
## Comment 26305
Date: 2018-03-29 15:39:09 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| [Mar2018]: mclient's default timing mode should (again) be "none" rather than "clock" | https://api.github.com/repos/MonetDB/MonetDB/issues/6537/comments | 0 | 2020-11-30T16:24:46Z | 2024-06-27T13:03:10Z | https://github.com/MonetDB/MonetDB/issues/6537 | 753,618,347 | 6,537 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-08 07:30:45 +0100
From: @drstmane
To: clients devs <<bugs-clients>>
Version: 11.27.11 (Jul2017-SP3)
Last updated: 2018-03-29 15:39:07 +0200
## Comment 26171
Date: 2018-02-08 07:30:45 +0100
From: @drstmane
timing output of mclient -t / --timer= should go to stderr rather than stdout
## Comment 26174
Date: 2018-02-08 08:58:23 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [efcbf80ad9b0](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=efcbf80ad9b0) made by Stefan Manegold <Stefan.Manegold@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=efcbf80ad9b0](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=efcbf80ad9b0)
Changeset description:
mclient manpage: warn about default --timer=clock invalidating csv, tab, xml output formats
(as long as the timing output stil goes to stdout rather than stderr;
cf., feature request / bug #6536)
## Comment 26178
Date: 2018-02-08 09:41:37 +0100
From: @drstmane
for an alternative (or addition?), see also feature request / bug #6537.
much like feature request / bug #6537 this is mainly to avoid that the default timing output compromises/invalidates the format of mclient's output formatters csv, tab, xml.
## Comment 26194
Date: 2018-02-14 09:56:52 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [78546e443dd3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=78546e443dd3) made by Stefan Manegold <Stefan.Manegold@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=78546e443dd3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=78546e443dd3)
Changeset description:
mclient, timerHuman(): report all timing on stderr rather than stdout,
as per feature request / bug #6536 .
## Comment 26196
Date: 2018-02-14 10:16:40 +0100
From: @drstmane
Fixed / implemented by changeset [78546e443dd3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=78546e443dd3) .
## Comment 26301
Date: 2018-03-29 15:39:07 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| [Mar2018]: timing output of mclient -t / --timer= should go to stderr rather than stdout | https://api.github.com/repos/MonetDB/MonetDB/issues/6536/comments | 0 | 2020-11-30T16:24:43Z | 2024-06-27T13:03:09Z | https://github.com/MonetDB/MonetDB/issues/6536 | 753,618,310 | 6,536 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-08 07:20:39 +0100
From: @drstmane
To: clients devs <<bugs-clients>>
Version: 11.27.11 (Jul2017-SP3)
Last updated: 2018-03-29 15:39:06 +0200
## Comment 26169
Date: 2018-02-08 07:20:39 +0100
From: @drstmane
$ for i in clock minutes seconds milliseconds microseconds none ; do echo -e "\n\t$i\n" ; mclient -d$DB -s'select 0;' -t $i ; echo ; done
clock
+------+
| L2 |
+======+
| 0 |
+------+
1 tuple
clk: 0.579 ms
minutes
warning: invalid argument to -t: minutes
+------+
| L2 |
+======+
| 0 |
+------+
1 tuple
clk: 0.568 ms
seconds
warning: invalid argument to -t: seconds
+------+
| L2 |
+======+
| 0 |
+------+
1 tuple
clk: 0.576 ms
milliseconds
warning: invalid argument to -t: milliseconds
+------+
| L2 |
+======+
| 0 |
+------+
1 tuple
clk: 0.551 ms
microseconds
warning: invalid argument to -t: microseconds
+------+
| L2 |
+======+
| 0 |
+------+
1 tuple
clk: 0.567 ms
none
+------+
| L2 |
+======+
| 0 |
+------+
1 tuple
$ for i in clock minutes seconds milliseconds microseconds none ; do echo -e "\n\t$i\n" ; mclient -d$DB -s'select 0;' --timer=$i ; echo ; done
clock
+------+
| L2 |
+======+
| 0 |
+------+
1 tuple
clk: 0.568 ms
minutes
warning: invalid argument to -t: minutes
+------+
| L2 |
+======+
| 0 |
+------+
1 tuple
clk: 0.563 ms
seconds
warning: invalid argument to -t: seconds
+------+
| L2 |
+======+
| 0 |
+------+
1 tuple
clk: 0.709 ms
milliseconds
warning: invalid argument to -t: milliseconds
+------+
| L2 |
+======+
| 0 |
+------+
1 tuple
clk: 0.567 ms
microseconds
warning: invalid argument to -t: microseconds
+------+
| L2 |
+======+
| 0 |
+------+
1 tuple
clk: 0.570 ms
none
+------+
| L2 |
+======+
| 0 |
+------+
1 tuple
## Comment 26170
Date: 2018-02-08 07:22:56 +0100
From: @drstmane
$ mclient --help 2>&1 | grep timer
-t | --timer=format use time formatting {clock,minutes,seconds,milliseconds,microseconds,none}
## Comment 26172
Date: 2018-02-08 08:03:40 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [1c0a5c06a226](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=1c0a5c06a226) made by Stefan Manegold <Stefan.Manegold@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=1c0a5c06a226](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=1c0a5c06a226)
Changeset description:
fixing bug #6535: correct mclient help message for new --timer / -t / \t option
## Comment 26173
Date: 2018-02-08 08:04:48 +0100
From: @drstmane
fixed by correcting mclient's help message for new --timer / -t / \t option
## Comment 26300
Date: 2018-03-29 15:39:06 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| [Mar2018]: mclient -t / --timer does not work as documented | https://api.github.com/repos/MonetDB/MonetDB/issues/6535/comments | 0 | 2020-11-30T16:24:40Z | 2024-06-27T13:03:08Z | https://github.com/MonetDB/MonetDB/issues/6535 | 753,618,261 | 6,535 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-08 07:16:36 +0100
From: @drstmane
To: clients devs <<bugs-clients>>
Version: 11.27.11 (Jul2017-SP3)
Last updated: 2018-03-29 15:39:25 +0200
## Comment 26168
Date: 2018-02-08 07:16:36 +0100
From: @drstmane
$ for i in tab sql expanded x csv raw xml trash tab ; do echo -e "\n\t$i\n" ; mclient -d$DB -s'select 1,2;' -f $i ; echo ; done
tab
unsupported formatter
% .L2, .L4 table_name
% L2, L4 name
% tinyint, tinyint type
% 1, 1 length
[ 1, 2 ]
clk: 0.822 ms
sql
+------+------+
| L2 | L4 |
+======+======+
| 1 | 2 |
+------+------+
1 tuple
clk: 0.830 ms
expanded
-[ RECORD 1 ]-
L2 | 1
L4 | 2
clk: 0.811 ms
x
-[ RECORD 1 ]-
L2 | 1
L4 | 2
clk: 0.796 ms
csv
1,2
clk: 0.824 ms
raw
% .L2, .L4 table_name
% L2, L4 name
% tinyint, tinyint type
% 1, 1 length
[ 1, 2 ]
clk: 0.782 ms
xml
<?xml version='1.0' encoding='UTF-8'?>
<!DOCTYPE table [
<!ELEMENT table (row)*>
<!ELEMENT row (column)+>
<!ELEMENT column (PCDATA)>
<!ATTLIST table name CDATA IMPLIED>
<!ATTLIST column name CDATA IMPLIED
isnull (true|false) 'false'>]>
<table name=".L2">
<row><column name="L2">1</column><column name="L4">2</column></row>
</table>
clk: 0.979 ms
trash
clk: 0.932 ms
tab
unsupported formatter
% .L2, .L4 table_name
% L2, L4 name
% tinyint, tinyint type
% 1, 1 length
[ 1, 2 ]
clk: 0.836 ms
## Comment 26175
Date: 2018-02-08 09:27:33 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [bcc430e78d4c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=bcc430e78d4c) made by Stefan Manegold <Stefan.Manegold@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=bcc430e78d4c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=bcc430e78d4c)
Changeset description:
revert undocumented change of changeset [770e047b548c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=770e047b548c) "tab" -> "tsv" -> "tab"
changeset [770e047b548c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=770e047b548c) changed the name of mclient's tabular / tab-separated
output formatter from "tab" to "tsv" without either documenting this
change in the checkin message or updating any related documentation
(--help messages, manpage) accordingly.
To fix bug #6534, I herewith revert that change, i.e., mclient's
tabular / tab-seperated output formatter is now again called "tab"
as it used to be and as all documentation still claims.
(If you prefer "tsv" over "tab" after all, please feel free to change that
correctly and consistently in all code, documentation and test output.)
## Comment 26176
Date: 2018-02-08 09:29:04 +0100
From: @drstmane
fixed by reverting undocumented (and incomplete) change of changeset [770e047b548c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=770e047b548c)
## Comment 26326
Date: 2018-03-29 15:39:25 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| [Mar2018]: mclient -f tab / --format=tab complains about "unsupported formatter" | https://api.github.com/repos/MonetDB/MonetDB/issues/6534/comments | 0 | 2020-11-30T16:24:37Z | 2024-06-27T13:03:06Z | https://github.com/MonetDB/MonetDB/issues/6534 | 753,618,225 | 6,534 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-05 15:59:15 +0100
From: Richard Hughes <<richard.monetdb>>
To: SQL devs <<bugs-sql>>
Version: 11.27.11 (Jul2017-SP3)
Last updated: 2018-02-05 16:02:29 +0100
## Comment 26166
Date: 2018-02-05 15:59:15 +0100
From: Richard Hughes <<richard.monetdb>>
Created attachment 592
Implementation v1
The attached is a first attempt at a patch implementing support for SIMILAR TO expressions: "select count(*) from t where v similar to '%(b|d)%';"
My specific motivating use-case was queries of the form "select count(*) from big where v like ? or v like ? or v like ? or ...". Those are slower than I'd like because they perform N passes over the v column; the neatest way I could figure out to make MonetDB perform only one pass was to add regexes and thereby get alternation support.
Points worth mentioning (in no particular order):
1) I don't have a copy of the ANSI SQL standard, so I worked off the PostgreSQL documentation (https://www.postgresql.org/docs/10/static/functions-matching.htmlFUNCTIONS-SIMILARTO-REGEXP) excluding the substring() stuff.
2) That documentation doesn't talk about what exactly is supported in [] character classes. I have not filtered that part of the expression at all, so it's going to support the normal [^a-z] stuff and [:digit:] and the Perl extension [:^digit:]. It's highly unlikely that ANSI intended that last one to work, but I didn't see any harm in going beyond the standard there.
3) In fact, the whole similar2pcre() implementation is really scary. That needs to be reviewed very carefully to ensure I haven't managed to let any security problems through.
4) I went with the "similar to" syntax because it's the most standardish syntax I know of. I presume a function-based syntax would also work, however it would look slightly weird in the optimizers to put in a dependency on a specific extension. It may be worth observing that the always-reliable 'some random guy on Stack Overflow' says "SIMILAR TO is only supported in PostgreSQL because it ended up in early drafts of the SQL standard. They still haven't gotten rid of it. But there are plans to remove it and include regexp matches instead - or so I heard." (https://dba.stackexchange.com/questions/10694/pattern-matching-with-like-similar-to-or-regular-expressions-in-postgresql/1069610696).
5) There aren't any tests in this patch. I presume PostgreSQL have some that we can steal but I haven't looked in to the process for importing them in to the MonetDB test suite.
6) I actually want case-insensitive matching. I'm planning to emulate that in my client application using the internationalizationally highly-dodgy '[tT][eE][sS][tT]', but if somebody can suggest a neat way to add an 'isimilar to' feature then I'm interested. My opinion is that it's not a good idea to add exactly that spelling because it's excessively non-standard.
7) I haven't tested the non-PCRE (i.e. libc regexec()) code path at all. I don't even know if it compiles.
8) I'm not clear on the policy which has been used to decide how many MAL functions to add. I've done the exact same entry points as LIKE so that the SQL and optimizer code can be common between the two, but it does look a little excessive versus simply having one entry point which is given all the necessary parameters.
9) I'm not confident about the code I've added to sql_upgrades.c. I need to look more carefully about how that file decides what code to run when. I'm also concerned about what happens if I maintain a local copy of this patch for an extended amount of time before it gets upstreamed by you guys.
10) I haven't implemented a LIKEjoin equivalent. It's not one of my use cases and it doesn't seem like it would be a common use case for other people and I hope (but haven't tested) that queries requiring it would still work, albeit slowly.
11) Since my use case was making things faster, I've also enabled the PCRE JIT. Some quick benchmarking suggests that this is about 55% of the total runtime of the non-JIT. I haven't benchmarked the cost of compiling the JIT and the TLB flush(es) involved. We could benchmark and calculate the threshold of the number of rows which are being scanned to make the overhead worth it. There's a long section in the PCRE docs titled "Availability of JIT support" which points out that a compile-time check is not necessarily easy to do; we could either add it to autoconf or accept that PCRE JIT support has been available for many years and it's unlikely that anybody would switch it off.
12) I've used PCRE_NO_UTF8_CHECK. This gives a 25% speed boost (i.e. 75% of the runtime) for non-JIT and is implied/mandatory for pcre_jit_exec() but comes with the warning "When it is set, the effect of passing an invalid UTF-8 string as a pattern is undefined. It may cause your program to crash or loop.". On the occasions where I've accidentally fed bogus data to MonetDB it has always caught any malformed UTF-8 but I'm not certain that this is absolutely guaranteed everywhere (e.g. select '\201\202').
> Attached file: [similarto.patch](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_6533_similarto.patch_592) (text/plain, 34512 bytes)
> Description: Implementation v1
| SIMILAR TO regular expression matching [patch] | https://api.github.com/repos/MonetDB/MonetDB/issues/6533/comments | 0 | 2020-11-30T16:24:34Z | 2024-06-28T13:39:43Z | https://github.com/MonetDB/MonetDB/issues/6533 | 753,618,181 | 6,533 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-04 18:46:54 +0100
From: @sjoerdmullender
To: SQL devs <<bugs-sql>>
Version: 11.27.11 (Jul2017-SP3)
Last updated: 2018-03-29 15:39:23 +0200
## Comment 26161
Date: 2018-02-04 18:46:54 +0100
From: @sjoerdmullender
## Comment 26162
Date: 2018-02-04 18:53:10 +0100
From: @sjoerdmullender
The following creates a table and insert one row where the first column is filled in using the default value (a sequence) and the other columns come from a file. When doing this, the NULL AS blurb gets ignored.
create table t (id int auto_increment, a clob, b clob);
copy 1 records into t (a, b) from stdin (id, a, b) using delimiters ',','\n','"' null as '';
2,,
select * from t;
this now results in a row with values id=1, a='', b='' instead of id=1,a=NULL,b=NULL.
## Comment 26163
Date: 2018-02-04 18:54:37 +0100
From: @sjoerdmullender
The first column must be in the file, but it must be ignored by the COPY IN.
## Comment 26164
Date: 2018-02-04 22:36:20 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [e2fda4b58c9f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e2fda4b58c9f) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=e2fda4b58c9f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=e2fda4b58c9f)
Changeset description:
Don't skip checking for nullstr when skipping first column.
This fixes bug #6532.
## Comment 26323
Date: 2018-03-29 15:39:23 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| copy into ignore null as directive if first column doesn't come from file | https://api.github.com/repos/MonetDB/MonetDB/issues/6532/comments | 0 | 2020-11-30T16:24:30Z | 2024-06-27T13:03:04Z | https://github.com/MonetDB/MonetDB/issues/6532 | 753,618,137 | 6,532 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-01 12:40:47 +0100
From: @hannesmuehleisen
To: SQL devs <<bugs-sql>>
Version: 11.27.11 (Jul2017-SP3)
CC: @njnes
Last updated: 2018-02-04 10:34:11 +0100
## Comment 26159
Date: 2018-02-01 12:40:47 +0100
From: @hannesmuehleisen
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_13_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/64.0.3282.119 Safari/537.36
Build Identifier:
select 0.7/987654321.0; returns 0.0
It should probably not? More correct result obtained with
select cast(0.7 as double)/987654321.0 ; (7.087499999911405e-10)
Reproducible: Always
## Comment 26160
Date: 2018-02-04 10:34:11 +0100
From: @njnes
this is a cast between 2 decimals (x,1). Automatically casting to doubles is not done currently.
| Precision issues with decimals | https://api.github.com/repos/MonetDB/MonetDB/issues/6531/comments | 0 | 2020-11-30T16:24:27Z | 2024-06-28T06:49:33Z | https://github.com/MonetDB/MonetDB/issues/6531 | 753,618,096 | 6,531 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-01 10:56:01 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-05-17 08:06:02 +0200
## Comment 26156
Date: 2018-02-01 10:56:01 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/64.0.3282.119 Safari/537.36
Build Identifier:
The select query bellow with the NOT IN operator is giving wrong results:
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab0(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab0 VALUES (97,1,99), (15,81,47), (87,21,10);
3. SELECT ALL - ( - COUNT ( * ) ) FROM tab0 AS cor0 WHERE col0 / ( - + col2 ) NOT IN ( + + col0, + col0 * + ( col1 ), + ( + - col0 ) * - 45, + CAST ( NULL AS INTEGER ) );
### Actual Results:
3
## Comment 26158
Date: 2018-02-01 10:59:20 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [165307be5363](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=165307be5363) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=165307be5363](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=165307be5363)
Changeset description:
Added tests for bugs 6529 and 6530
## Comment 26217
Date: 2018-02-18 16:17:38 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [bead8d5e8ca2](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=bead8d5e8ca2) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=bead8d5e8ca2](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=bead8d5e8ca2)
Changeset description:
fixes for bug #6529 and 6530, ie do not set the tnonil flag (nil in gives nil hash values).
## Comment 26312
Date: 2018-03-29 15:39:14 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
## Comment 26452
Date: 2018-05-17 08:06:02 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [1d3a15ef3aee](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=1d3a15ef3aee) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=1d3a15ef3aee](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=1d3a15ef3aee)
Changeset description:
added stable output for bug #6602 (and correct for bugs 6530 and 6594)
| Sqlitelogictest: select query with NOT IN giving wrong results | https://api.github.com/repos/MonetDB/MonetDB/issues/6530/comments | 0 | 2020-11-30T16:24:23Z | 2024-06-27T13:03:02Z | https://github.com/MonetDB/MonetDB/issues/6530 | 753,618,051 | 6,530 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-02-01 10:43:39 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-03-29 15:39:17 +0200
## Comment 26155
Date: 2018-02-01 10:43:39 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/64.0.3282.119 Safari/537.36
Build Identifier:
Mserver gets an assertion error in rel_bin.c with the following query bellow.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab2(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab2 VALUES (64,77,40), (75,67,58), (46,51,23);
3. SELECT * FROM tab2 cor0 WHERE col1 IN ( + CAST ( NULL AS INTEGER ) + - CAST ( NULL AS INTEGER ), col2 * + col1 + - - 30 );
### Actual Results:
The following assertion error:
mserver5: rel_bin.c:575: exp_bin: Assertion `s' failed.
### Expected Results:
An empty result set.
Backtrace:
0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:51
1 0x00007ffff4748381 in __GI_abort () at abort.c:79
2 0x00007ffff473e8fa in __assert_fail_base (fmt=0x7ffff48b9c28 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", assertion=assertion@entry=0x7fffe84cce68 "s", file=file@entry=0x7fffe84ccdf8 "rel_bin.c",
line=line@entry=575, function=function@entry=0x7fffe84cd548 <__PRETTY_FUNCTION__.15580> "exp_bin") at assert.c:92
3 0x00007ffff473e972 in __GI___assert_fail (assertion=0x7fffe84cce68 "s", file=0x7fffe84ccdf8 "rel_bin.c", line=575, function=0x7fffe84cd548 <__PRETTY_FUNCTION__.15580> "exp_bin") at assert.c:101
4 0x00007fffe835963a in exp_bin (be=0x7fffb4002b60, e=0x7fffb40f49f0, left=0x7fffb40fbdf0, right=0x7fffb40fc420, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:575
5 0x00007fffe836292f in rel2bin_project (be=0x7fffb4002b60, rel=0x7fffb40f4a80, refs=0x7fffb40f97f0, topn=0x0) at rel_bin.c:2479
6 0x00007fffe836d73c in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb40f4a80, refs=0x7fffb40f97f0) at rel_bin.c:4856
7 0x00007fffe836285c in rel2bin_project (be=0x7fffb4002b60, rel=0x7fffb40f7910, refs=0x7fffb40f97f0, topn=0x0) at rel_bin.c:2467
8 0x00007fffe836d73c in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb40f7910, refs=0x7fffb40f97f0) at rel_bin.c:4856
9 0x00007fffe836101e in rel2bin_union (be=0x7fffb4002b60, rel=0x7fffb40f6a00, refs=0x7fffb40f97f0) at rel_bin.c:2158
10 0x00007fffe836d6b0 in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb40f6a00, refs=0x7fffb40f97f0) at rel_bin.c:4844
11 0x00007fffe836004f in rel2bin_semijoin (be=0x7fffb4002b60, rel=0x7fffb40f33a0, refs=0x7fffb40f97f0) at rel_bin.c:1937
12 0x00007fffe836d682 in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb40f33a0, refs=0x7fffb40f97f0) at rel_bin.c:4840
13 0x00007fffe836285c in rel2bin_project (be=0x7fffb4002b60, rel=0x7fffb40f3440, refs=0x7fffb40f97f0, topn=0x0) at rel_bin.c:2467
14 0x00007fffe836d73c in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb40f3440, refs=0x7fffb40f97f0) at rel_bin.c:4856
15 0x00007fffe836d9e4 in _subrel_bin (be=0x7fffb4002b60, rel=0x7fffb40f3440, refs=0x7fffb40f97f0) at rel_bin.c:4913
16 0x00007fffe836dafc in output_rel_bin (be=0x7fffb4002b60, rel=0x7fffb40f3440) at rel_bin.c:4936
17 0x00007fffe8389f9b in sql_relation2stmt (be=0x7fffb4002b60, r=0x7fffb40f3440) at sql_gencode.c:532
18 0x00007fffe838a18e in backend_dumpstmt (be=0x7fffb4002b60, mb=0x7fffb4101220, r=0x7fffb40f3440, top=1, add_end=1,
query=0x7fffb40feb30 "select * from tab2 cor0 where col1 in ( + cast ( null as integer ) + - cast ( null as integer ), col2 * + col1 + - - 30 );") at sql_gencode.c:572
19 0x00007fffe838aa06 in backend_dumpproc (be=0x7fffb4002b60, c=0x7fffe970e368, cq=0x7fffb4105a00, r=0x7fffb40f3440) at sql_gencode.c:702
20 0x00007fffe834db2f in SQLparser (c=0x7fffe970e368) at sql_scenario.c:1226
21 0x00007ffff7a0ca7b in runPhase (c=0x7fffe970e368, phase=1) at mal_scenario.c:510
22 0x00007ffff7a0cb84 in runScenarioBody (c=0x7fffe970e368, once=0) at mal_scenario.c:532
23 0x00007ffff7a0ce3d in runScenario (c=0x7fffe970e368, once=0) at mal_scenario.c:569
24 0x00007ffff7a0ed78 in MSserveClient (dummy=0x7fffe970e368) at mal_session.c:514
25 0x00007ffff7a0e8c7 in MSscheduleClient (command=0x7fffb4000b30 "0", challenge=0x7fffcb3b3d9b "U2YlY6LW", fin=0x17fc5f0, fout=0x7fffc4002d90, protocol=PROTOCOL_9, blocksize=8190, compute_column_widths=0)
at mal_session.c:397
26 0x00007ffff7aae933 in doChallenge (data=0x17fc600) at mal_mapi.c:284
27 0x00007ffff7694aca in thread_starter (arg=0x7fffc4004e80) at gdk_system.c:475
28 0x00007ffff4af961b in start_thread (arg=0x7fffcb3b4700) at pthread_create.c:465
29 0x00007ffff482698f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95
## Comment 26157
Date: 2018-02-01 10:59:19 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [165307be5363](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=165307be5363) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=165307be5363](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=165307be5363)
Changeset description:
Added tests for bugs 6529 and 6530
## Comment 26216
Date: 2018-02-18 16:17:37 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [bead8d5e8ca2](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=bead8d5e8ca2) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=bead8d5e8ca2](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=bead8d5e8ca2)
Changeset description:
fixes for bug #6529 and 6530, ie do not set the tnonil flag (nil in gives nil hash values).
## Comment 26221
Date: 2018-02-19 10:46:33 +0100
From: @PedroTadim
This query still crashes when I compile MonetDB without hugeint support.
## Comment 26236
Date: 2018-03-01 19:30:42 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [f40eedc96966](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=f40eedc96966) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=f40eedc96966](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=f40eedc96966)
Changeset description:
more fixes for bug #6529, make sure the identity column expressions are unique
## Comment 26237
Date: 2018-03-01 19:31:06 +0100
From: @njnes
fixed
## Comment 26316
Date: 2018-03-29 15:39:17 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| Sqlitelogictest crash in select query with IN operator and cast | https://api.github.com/repos/MonetDB/MonetDB/issues/6529/comments | 0 | 2020-11-30T16:24:20Z | 2024-06-27T13:03:01Z | https://github.com/MonetDB/MonetDB/issues/6529 | 753,617,991 | 6,529 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-01-31 14:07:13 +0100
From: @drstmane
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @mlkersten, @njnes, @PedroTadim
Last updated: 2019-11-29 17:53:15 +0100
## Comment 26140
Date: 2018-01-31 14:07:13 +0100
From: @drstmane
Management summary:
MonetDB does not yet exploit potential "dead-code/join elimination" as optimization opportunity in queries (over views) with (left) outer foreign key joins.
Details:
Consider a normalized star-like schema with a single central pivot/fact table ("fk") with foreign-key joins into one or more satellite/dimension table(s) ("pk*"):
create table pk1 (
pk1 int generated always as identity not null primary key,
v1 int
);
create table pk2 (
pk2 int generated always as identity not null primary key,
v2 int
);
create table fk (
id int generated always as identity not null primary key,
fk1 int references pk1(pk1),
fk2 int references pk2(pk2)
);
Also, consider a single (non-materialized) view "v" using left outer joins to gather all data into a single de-normalized "universal" table --- for user convenience:
create view v as (
select * from fk
left outer join pk1 on fk.fk1 = pk1.pk1
left outer join pk2 on fk.fk2 = pk2.pk2
);
With this, any query over view "v" that does not access any attributes from (one (or more, or all) of) the satellite/dimension table(s) does not require the respective left outer join(s) to be executed --- assuming my knowledge and understanding of SQL semantics is correct; correct me if not ...
For instance, the number of tuples in the view is (guaranteed to be) identical to the number of tuples in the pivot/fact table "fk". Consequently, queries like
"select count(*) from v",
"select count(*) from fk left outer join pk1 on fk.fk1 = pk1.pk1" or
"select count(*) from fk left outer join pk1 on fk.fk1 = pk1.pk1 left outer join pk2 on fk.fk2 = pk2.pk2"
are bound/guaranteed to return the same result and query
"select count(*) from fk".
Thus, there is no need to perform any of the left outer foreign-key joins to evaluate this query.
I.e., instead of
> plan select count(*) from v;
project (
| group by (
| | project (
| | | left outer join (
| | | | left outer join (
| | | | | table(sys.fk) [ "fk"."id" NOT NULL HASHCOL , "fk"."fk1", "fk"."fk2" ] COUNT ,
| | | | | table(sys.pk1) [ "pk1"."pk1" NOT NULL HASHCOL ] COUNT
| | | | ) [ "fk"."fk1" = "pk1"."pk1" NOT NULL HASHCOL ],
| | | | table(sys.pk2) [ "pk2"."pk2" NOT NULL HASHCOL ] COUNT
| | | ) [ "fk"."fk2" = "pk2"."pk2" NOT NULL HASHCOL ]
| | ) [ "fk"."id" NOT NULL HASHCOL as "v"."id" ]
| ) [ ] [ sys.count() NOT NULL as "L13"."L13" ]
) [ "L13" NOT NULL as "L14"."L13" ]
a simple
> plan select count(*) from v;
project (
| group by (
| | project (
| | | table(sys.fk) [ "fk"."id" NOT NULL HASHCOL ] COUNT
| | ) [ "fk"."id" NOT NULL HASHCOL as "v"."id" ]
| ) [ ] [ sys.count() NOT NULL as "L7"."L7" ]
) [ "L7" NOT NULL as "L10"."L7" ]
or even
> plan select count(*) from v;
project (
| group by (
| | table(sys.fk) [ "fk"."id" NOT NULL HASHCOL ] COUNT
| ) [ ] [ sys.count() NOT NULL as "L3"."L3" ]
) [ "L3" NOT NULL as "L4"."L3" ]
would be sufficient.
I'll prove some more examples in the form of tests in the MonetDB source repository.
Unfortunately, at the time of writing, MonetDB does not yet exploit this "dead-code elimination" as optimization opportunity, but rather canonically executes all joins in the view with any query over the view.
## Comment 26141
Date: 2018-01-31 14:09:52 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [0adae9690ba4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=0adae9690ba4) made by Stefan Manegold <Stefan.Manegold@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=0adae9690ba4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=0adae9690ba4)
Changeset description:
added tests with "desired" sample PLANs & EXPLAINs for feature request / bug #6528
The tests that are still bound to fail until the feature is actually implemented
are (for now) disabled with ("unknown") conditional "FEATURE_REQUEST".
## Comment 26142
Date: 2018-01-31 15:07:22 +0100
From: @drstmane
As an example what performance gain would be possible, here are some counts and times to calculate them for some satellite/dimension tables ("PK[012]"), pivot/fact tables ("FK[12]"), and views ("V[12]"):
select count(*) from "PK0"; -> 24,160,999 ( 0.062s)
select count(*) from "PK1"; -> 481,398,012 ( 0.046s)
select count(*) from "PK2"; -> 324,529,050 ( 0.043s)
select count(*) from "FK1"; -> 268,516,267 ( 0.054s)
select count(*) from "FK2"; -> 1,588,447,035 ( 0.043s)
select count(*) from "V1" ; -> 268,516,267 ( 5m:28.835s)
select count(*) from "V2" ; -> 1,588,447,035 (2h:24m:16.584s)
with these schema definitions (omitting satellite/dimension tables "PK[012]"):
CREATE TABLE "FK1" (
"FK1_1" INTEGER REFERENCES "PK1" ("PK"),
"FK1_2" INTEGER REFERENCES "PK1" ("PK"),
"FK1_3" INTEGER REFERENCES "PK1" ("PK"),
"FK1_4" INTEGER REFERENCES "PK1" ("PK"),
"FK1_5" INTEGER REFERENCES "PK1" ("PK"),
"FK1_6" INTEGER REFERENCES "PK1" ("PK"),
"FK1_7" INTEGER REFERENCES "PK1" ("PK"),
"FK2_1" INTEGER REFERENCES "PK2" ("PK"),
"FK2_2" INTEGER REFERENCES "PK2" ("PK"),
"FK2_3" INTEGER REFERENCES "PK2" ("PK"),
"FK2_4" INTEGER REFERENCES "PK2" ("PK"),
"FK2_5" INTEGER REFERENCES "PK2" ("PK"),
"FK2_6" INTEGER REFERENCES "PK2" ("PK"),
"FK2_7" INTEGER REFERENCES "PK2" ("PK"),
);
CREATE TABLE "FK2" (
"FK0" INTEGER REFERENCES "PK0" ("PK"),
"FK1_1" INTEGER REFERENCES "PK1" ("PK"),
"FK1_2" INTEGER REFERENCES "PK1" ("PK"),
"FK1_3" INTEGER REFERENCES "PK1" ("PK"),
"FK1_4" INTEGER REFERENCES "PK1" ("PK"),
"FK1_5" INTEGER REFERENCES "PK1" ("PK"),
"FK1_6" INTEGER REFERENCES "PK1" ("PK"),
"FK1_7" INTEGER REFERENCES "PK1" ("PK"),
"FK2_1" INTEGER REFERENCES "PK2" ("PK"),
"FK2_2" INTEGER REFERENCES "PK2" ("PK"),
"FK2_3" INTEGER REFERENCES "PK2" ("PK"),
"FK2_4" INTEGER REFERENCES "PK2" ("PK"),
"FK2_5" INTEGER REFERENCES "PK2" ("PK"),
"FK2_6" INTEGER REFERENCES "PK2" ("PK"),
"FK2_7" INTEGER REFERENCES "PK2" ("PK"),
);
create view "V1" as (
select *
from "FK1" fk
left outer join "PK1" pk1_1 on fk."FK1_1" = pk1_1."PK"
left outer join "PK1" pk1_2 on fk."FK1_2" = pk1_2."PK"
left outer join "PK1" pk1_3 on fk."FK1_3" = pk1_3."PK"
left outer join "PK1" pk1_4 on fk."FK1_4" = pk1_4."PK"
left outer join "PK1" pk1_5 on fk."FK1_5" = pk1_5."PK"
left outer join "PK1" pk1_6 on fk."FK1_6" = pk1_6."PK"
left outer join "PK1" pk1_7 on fk."FK1_7" = pk1_7."PK"
left outer join "PK2" pk2_1 on fk."FK2_1" = pk2_1."PK"
left outer join "PK2" pk2_2 on fk."FK2_2" = pk2_2."PK"
left outer join "PK2" pk2_3 on fk."FK2_3" = pk2_3."PK"
left outer join "PK2" pk2_4 on fk."FK2_4" = pk2_4."PK"
left outer join "PK2" pk2_5 on fk."FK2_5" = pk2_5."PK"
left outer join "PK2" pk2_6 on fk."FK2_6" = pk2_6."PK"
left outer join "PK2" pk2_7 on fk."FK2_7" = pk2_7."PK"
);
create view "V2" as (
select *
from "FK2" fk
left outer join "PK0" pk0 on fk."FK0" = pk0."PK"
left outer join "PK1" pk1_1 on fk."FK1_1" = pk1_1."PK"
left outer join "PK1" pk1_2 on fk."FK1_2" = pk1_2."PK"
left outer join "PK1" pk1_3 on fk."FK1_3" = pk1_3."PK"
left outer join "PK1" pk1_4 on fk."FK1_4" = pk1_4."PK"
left outer join "PK1" pk1_5 on fk."FK1_5" = pk1_5."PK"
left outer join "PK1" pk1_6 on fk."FK1_6" = pk1_6."PK"
left outer join "PK1" pk1_7 on fk."FK1_7" = pk1_7."PK"
left outer join "PK2" pk2_1 on fk."FK2_1" = pk2_1."PK"
left outer join "PK2" pk2_2 on fk."FK2_2" = pk2_2."PK"
left outer join "PK2" pk2_3 on fk."FK2_3" = pk2_3."PK"
left outer join "PK2" pk2_4 on fk."FK2_4" = pk2_4."PK"
left outer join "PK2" pk2_5 on fk."FK2_5" = pk2_5."PK"
left outer join "PK2" pk2_6 on fk."FK2_6" = pk2_6."PK"
left outer join "PK2" pk2_7 on fk."FK2_7" = pk2_7."PK"
);
## Comment 26143
Date: 2018-01-31 15:17:44 +0100
From: @drstmane
Omitting the "unused" foreign key joins is indeed only possible in these cases:
(1)
If the foreign key attribute is NOT declared to be NOT NULL, the foreign key join must be a one-sided outer join, with outer on the foreign side of the join, i.e., "FK left outer join PK" or "PK right outer join FK".
(2)
If the foreign key attribute IS declared to be NOT NULL, the foreign key join can other be a foreign-side outer join as with (1) or an inner join.
In all other cases, the query result without some or all foreign key joins is not guaranteed to have the same length (i.e., number of tuples) as the result with all foreign key joins.
## Comment 26144
Date: 2018-01-31 15:25:04 +0100
From: @mlkersten
Is there evidence that the dead code produces a significant performance degradation?
## Comment 26145
Date: 2018-01-31 15:31:33 +0100
From: @mlkersten
sorry, some info was provided ;)
## Comment 26146
Date: 2018-01-31 15:41:09 +0100
From: @drstmane
NB,
materializing the view is no option in my case; trying so resulted in
"
GDK reported error: allocating too much virtual address space
allocating too much virtual address space
allocating too much virtual address space
HEAPextend: failed to extend to 6206751880 for 01/75/17511.tail: h->storage == STORE_MEM && can_map && fd >= 0 && HEAPload_intern() != GDK_SUCCEED
"
... after 70 hours of heaving I/O on a 256 GB machine and 30+TB free disk space ...
Yes, my "PK[012]" tables are also quite wide ...
## Comment 26233
Date: 2018-02-28 11:20:10 +0100
From: @njnes
implemented the sql optimizer step. A small step in mal optimization could
still be added.
## Comment 27040
Date: 2019-06-11 11:01:00 +0200
From: @sjoerdmullender
It looks like changeset [78c33accfff9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=78c33accfff9) has fixed this.
Stefan, are you happy with the fix?
| dead-code / join elimination in queries (over views) with (left) outer foreign key joins | https://api.github.com/repos/MonetDB/MonetDB/issues/6528/comments | 0 | 2020-11-30T16:24:16Z | 2024-06-27T13:03:00Z | https://github.com/MonetDB/MonetDB/issues/6528 | 753,617,953 | 6,528 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-01-31 00:15:17 +0100
From: Trevor <<tandreas>>
To: SQL devs <<bugs-sql>>
Version: 11.27.11 (Jul2017-SP3)
CC: @njnes
Last updated: 2018-02-12 16:12:09 +0100
## Comment 26138
Date: 2018-01-31 00:15:17 +0100
From: Trevor <<tandreas>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:58.0) Gecko/20100101 Firefox/58.0
Build Identifier:
A crash occurs when using an alias to a column in a subquery.
See the example below:
create table "t1" ("d1" int);
insert into "t1" values (1), (2), (3), (4), (5);
-- Works
select
"d1" as "value"
from "t1"
order by "value";
-- Works
select
a."value"
from (
select
"d1" as "value"
from "t1"
order by "d1"
) as a;
-- Crashes
select
a."value"
from (
select
"d1" as "value"
from "t1"
order by "value"
) as a;
From merovingian.log:
2018-01-30 17:10:15 ERR test[26999]: could not find L2.value
2018-01-30 17:10:15 ERR test[26999]: t1.d1
2018-01-30 17:10:15 ERR test[26999]: a.value
2018-01-30 17:10:15 ERR test[26999]: *** Error in `/usr/bin/mserver5': double free or corruption (out): 0x00007f95980cb810 ***
Reproducible: Always
### Steps to Reproduce:
Run the queries in the details.
### Actual Results:
Database crashes.
### Expected Results:
Query should complete.
## Comment 26151
Date: 2018-01-31 18:53:29 +0100
From: @njnes
fixed problem with table aliases
## Comment 26152
Date: 2018-01-31 21:53:32 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [f0b5bb2c3ac9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=f0b5bb2c3ac9) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=f0b5bb2c3ac9](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=f0b5bb2c3ac9)
Changeset description:
fixed bug #6527 (be more carefull with order by and table aliases)
| Crash using order by alias in subquery | https://api.github.com/repos/MonetDB/MonetDB/issues/6527/comments | 0 | 2020-11-30T16:24:14Z | 2024-06-27T13:02:59Z | https://github.com/MonetDB/MonetDB/issues/6527 | 753,617,918 | 6,527 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-01-30 21:26:25 +0100
From: Trevor <<tandreas>>
To: SQL devs <<bugs-sql>>
Version: 11.27.11 (Jul2017-SP3)
CC: @njnes
Last updated: 2018-03-29 15:38:57 +0200
## Comment 26137
Date: 2018-01-30 21:26:25 +0100
From: Trevor <<tandreas>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:58.0) Gecko/20100101 Firefox/58.0
Build Identifier:
I am experiencing a crash when attempting to use an aggregate function inside a case statement in a having clause. The server restarts correctly but the crash is reproducible every time.
Below is an example. One use case for this is that I am computing a formula on aggregated data, and want to ensure no divide by zero occurs.
Are case statements in having clauses supported?
create table "t1" (
"id" int,
"c1" varchar(100),
"d1" int,
"d2" int
);
insert into "t1" values
(1, 'A', 50, 80),
(2, 'A', 200, 350),
(3, 'A', 89, 125),
(4, 'B', 4845, 13),
(5, 'B', 194, 597),
(6, 'C', 5636, 5802),
(7, 'C', 375, 3405),
(7, 'D', 365, 0),
(7, 'D', 87, 0);
-- Works
select
"c1",
sum("d1") as "d1",
sum("d2") as "d2"
from "t1"
group by "c1"
having sum("d1") < case when 5 > 10 then 500 else 400 end;
-- Works
select
"c1",
sum("d1")as "d1",
sum("d2") as "d2",
1.0 * sum("d1") / (1.0 * case when sum("d2") > 0 then sum("d2") else null end) as "formula"
from "t1"
group by "c1";
-- Crashes
select
"c1",
sum("d1")as "d1",
sum("d2") as "d2",
1.0 * sum("d1") / (1.0 * case when sum("d2") > 0 then sum("d2") else null end) as "formula"
from "t1"
group by "c1"
having (1.0 * sum("d1") / (1.0 * case when sum("d2") > 0 then sum("d2") else null end)) > 1;
Looking in merovingian.log:
2018-01-29 14:47:15 ERR test[2879]: *** Error in `/usr/bin/mserver5': double free or corruption (out): 0x00007f57200ea4b0 ***
2018-01-29 14:47:18 MSG merovingian[504]: database 'test' (2879) has crashed (dumped core)
2018-01-29 14:49:35 MSG merovingian[504]: database 'test' has crashed after start on 2018-01-29 14:42:36, attempting restart, up min/avg/max: 0s/0s/0s, crash average: 1.00 0.20 0.07 (2-0=2)
2
Reproducible: Always
### Steps to Reproduce:
Run queries provided in details.
### Actual Results:
Database crashes.
### Expected Results:
Query completes and returns correct results.
## Comment 26153
Date: 2018-02-01 10:28:33 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [567582612f6f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=567582612f6f) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=567582612f6f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=567582612f6f)
Changeset description:
Added test for bug #6526.
## Comment 26210
Date: 2018-02-17 09:27:48 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [4d9b69cdf41d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=4d9b69cdf41d) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=4d9b69cdf41d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=4d9b69cdf41d)
Changeset description:
fixes for bug #6526, handle combination of aggregation in a case statement (in the having part).
## Comment 26211
Date: 2018-02-17 09:29:03 +0100
From: @njnes
fixed. Pass the information of handling the 'having' properly in the case statement
## Comment 26295
Date: 2018-03-29 15:38:57 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| Crash using aggregate function inside a case statement in having clause | https://api.github.com/repos/MonetDB/MonetDB/issues/6526/comments | 0 | 2020-11-30T16:24:12Z | 2024-06-27T13:02:58Z | https://github.com/MonetDB/MonetDB/issues/6526 | 753,617,885 | 6,526 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-01-30 16:36:49 +0100
From: Krunal Maniar <<krunalmaniar13>>
To: SQL devs <<bugs-sql>>
Version: 11.27.11 (Jul2017-SP3)
CC: martin.van.dinther, sagarharish999
Last updated: 2019-02-12 13:16:00 +0100
## Comment 26136
Date: 2018-01-30 16:36:49 +0100
From: Krunal Maniar <<krunalmaniar13>>
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:58.0) Gecko/20100101 Firefox/58.0
Build Identifier:
I have created 3 tables in sys schema in monetdb in one particular database D1.
Now i want to create a view in new schema "stats" which accesses all this 3 tables using joins.
After that i created a new user to whom i have granted the select permission on the view.
When i login with this new user and fire a select statement on the view it is throwing error:
access denied "new user" on the table sys.xyz.
This table is used in the views select statement.
Steps
created 2 tables:
1) sys.xyz
2) sys.abc
CREATE SCHEMA stats;
CREATE USER viewreader WITH PASSWORD '123456' NAME 'viewreader User' SCHEMA stats;
grant select on stats.vw_stats to viewreader;
create view stats.vw_tracking_stats as
select * from sys.abc a left join sys.xyz x on a.id = x.id.
Now i logged in with user: viewreader
and then executed the following statement on the view:
select * from stats.vw_stats.
It is giving me following error message:
access denied "viewreader" on the table sys.abc.
Does anybody have a clue what i have to do so that VIEW is able to access the table sys.abc and sys.xyz but the viewreader user should not be able to query the base table directly?
Reproducible: Always
### Actual Results:
SELECT: access denied for lidl to table 'stats.vw_tracking_stats'
### Expected Results:
I should be able to access the data using the view.
## Comment 26345
Date: 2018-04-05 14:36:41 +0200
From: Martin van Dinther <<martin.van.dinther>>
In SQL you will need to grant SELECT privilege to the viewreader for the tables (or columns) used in de view definition. This is because the view is just a stored query, so when querying the view, you are executing the stored query and thus querying the tables (or specific columns). So for the viewreader user you will need to give it SELECT privileges for tables sys.xyz and sys.abc also.
Tip: If a view accesses only a subset of the columns of a table you could limit SELECT privileges to only those columns of the table as referenced in the view, and thus prevent full access to all columns of the table, so select * from mytable would give a runtime error for that user.
| Give Access to VIEW but not the base table | https://api.github.com/repos/MonetDB/MonetDB/issues/6525/comments | 0 | 2020-11-30T16:24:08Z | 2024-06-28T06:50:19Z | https://github.com/MonetDB/MonetDB/issues/6525 | 753,617,851 | 6,525 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-01-25 13:32:42 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-02-14 10:20:34 +0100
## Comment 26121
Date: 2018-01-25 13:32:42 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36
Build Identifier:
Mserver gets an assertion error on default branch in rel_bin.c while compiling the query bellow.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab2(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab2 VALUES (15,61,87), (91,59,79), (92,41,58);
3. SELECT ALL - col0 AS col1 FROM tab2 AS cor0 GROUP BY col0 HAVING - COUNT ( * ) NOT IN ( - COUNT ( * ), 88, COUNT ( * ) );
### Actual Results:
An assertion error (in default branch):
mserver5: rel_bin.c:575: exp_bin: Assertion `s' failed.
### Expected Results:
An empty result set.
Backtrace:
0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:51
1 0x00007ffff4748381 in __GI_abort () at abort.c:79
2 0x00007ffff473e8fa in __assert_fail_base (fmt=0x7ffff48b9c28 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", assertion=assertion@entry=0x7fffe84ccd48 "s", file=file@entry=0x7fffe84cccd8 "rel_bin.c",
line=line@entry=575, function=function@entry=0x7fffe84cd428 <__PRETTY_FUNCTION__.15580> "exp_bin") at assert.c:92
3 0x00007ffff473e972 in __GI___assert_fail (assertion=0x7fffe84ccd48 "s", file=0x7fffe84cccd8 "rel_bin.c", line=575, function=0x7fffe84cd428 <__PRETTY_FUNCTION__.15580> "exp_bin") at assert.c:101
4 0x00007fffe83598d9 in exp_bin (be=0x7fffb4002b60, e=0x7fffb40f2690, left=0x7fffb40e3060, right=0x7fffb40e4a90, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:575
5 0x00007fffe8362bce in rel2bin_project (be=0x7fffb4002b60, rel=0x7fffb40f27c0, refs=0x7fffb40e1940, topn=0x0) at rel_bin.c:2479
6 0x00007fffe836d9db in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb40f27c0, refs=0x7fffb40e1940) at rel_bin.c:4856
7 0x00007fffe8362afb in rel2bin_project (be=0x7fffb4002b60, rel=0x7fffb40fdb20, refs=0x7fffb40e1940, topn=0x0) at rel_bin.c:2467
8 0x00007fffe836d9db in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb40fdb20, refs=0x7fffb40e1940) at rel_bin.c:4856
9 0x00007fffe83612bd in rel2bin_union (be=0x7fffb4002b60, rel=0x7fffb40fb4d0, refs=0x7fffb40e1940) at rel_bin.c:2158
10 0x00007fffe836d94f in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb40fb4d0, refs=0x7fffb40e1940) at rel_bin.c:4844
11 0x00007fffe83602ee in rel2bin_semijoin (be=0x7fffb4002b60, rel=0x7fffb40f2fb0, refs=0x7fffb40e1940) at rel_bin.c:1937
12 0x00007fffe836d921 in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb40f2fb0, refs=0x7fffb40e1940) at rel_bin.c:4840
13 0x00007fffe8362afb in rel2bin_project (be=0x7fffb4002b60, rel=0x7fffb40ef790, refs=0x7fffb40e1940, topn=0x0) at rel_bin.c:2467
14 0x00007fffe836d9db in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb40ef790, refs=0x7fffb40e1940) at rel_bin.c:4856
15 0x00007fffe8362afb in rel2bin_project (be=0x7fffb4002b60, rel=0x7fffb40f6840, refs=0x7fffb40e1940, topn=0x0) at rel_bin.c:2467
16 0x00007fffe836d9db in subrel_bin (be=0x7fffb4002b60, rel=0x7fffb40f6840, refs=0x7fffb40e1940) at rel_bin.c:4856
17 0x00007fffe836dc83 in _subrel_bin (be=0x7fffb4002b60, rel=0x7fffb40f6840, refs=0x7fffb40e1940) at rel_bin.c:4913
18 0x00007fffe836dd9b in output_rel_bin (be=0x7fffb4002b60, rel=0x7fffb40f6840) at rel_bin.c:4936
19 0x00007fffe838a23a in sql_relation2stmt (be=0x7fffb4002b60, r=0x7fffb40f6840) at sql_gencode.c:532
20 0x00007fffe838a42d in backend_dumpstmt (be=0x7fffb4002b60, mb=0x7fffb4101220, r=0x7fffb40f6840, top=1, add_end=1,
query=0x7fffb40feb30 "select all - col0 as col1 from tab2 as cor0 group by col0 having - count ( * ) not in ( - count ( * ), 88, count ( * ) );") at sql_gencode.c:572
21 0x00007fffe838aca5 in backend_dumpproc (be=0x7fffb4002b60, c=0x7fffe970e368, cq=0x7fffb4105a00, r=0x7fffb40f6840) at sql_gencode.c:702
22 0x00007fffe834db01 in SQLparser (c=0x7fffe970e368) at sql_scenario.c:1226
23 0x00007ffff7a0c9e1 in runPhase (c=0x7fffe970e368, phase=1) at mal_scenario.c:510
24 0x00007ffff7a0caea in runScenarioBody (c=0x7fffe970e368, once=0) at mal_scenario.c:532
25 0x00007ffff7a0cda3 in runScenario (c=0x7fffe970e368, once=0) at mal_scenario.c:569
26 0x00007ffff7a0ecde in MSserveClient (dummy=0x7fffe970e368) at mal_session.c:514
27 0x00007ffff7a0e82d in MSscheduleClient (command=0x7fffb4000b30 "0", challenge=0x7fffcb3b3d9b "ffre5Z6IKuY", fin=0x17fc360, fout=0x7fffc4002d90, protocol=PROTOCOL_9, blocksize=8190, compute_column_widths=0)
at mal_session.c:397
28 0x00007ffff7aae527 in doChallenge (data=0x17fc370) at mal_mapi.c:273
29 0x00007ffff7694515 in thread_starter (arg=0x7fffc4004e80) at gdk_system.c:475
30 0x00007ffff4af961b in start_thread (arg=0x7fffcb3b4700) at pthread_create.c:465
31 0x00007ffff482698f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95
## Comment 26126
Date: 2018-01-25 13:46:45 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [d1c6e7340c11](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d1c6e7340c11) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=d1c6e7340c11](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=d1c6e7340c11)
Changeset description:
Added test for bug #6524 (it only crashes on default branch)
## Comment 26149
Date: 2018-01-31 18:29:44 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [c8b7cdf0221a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=c8b7cdf0221a) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=c8b7cdf0221a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=c8b7cdf0221a)
Changeset description:
fix crash because of reused (later renamed) expression, bug #6524
## Comment 26150
Date: 2018-01-31 18:30:47 +0100
From: @njnes
crash was caused by a reused expression (we now create a reference expression)
## Comment 26167
Date: 2018-02-06 17:50:08 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [63141e9bb22d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=63141e9bb22d) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=63141e9bb22d](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=63141e9bb22d)
Changeset description:
Approve test for bug #6524
## Comment 26198
Date: 2018-02-14 10:20:34 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [9ecef945b19a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9ecef945b19a) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=9ecef945b19a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=9ecef945b19a)
Changeset description:
Approve test for bug #6524 (not done on Mar2018 branch)
| Sqlitelogictest: Crash in aggregation query with IN operator | https://api.github.com/repos/MonetDB/MonetDB/issues/6524/comments | 0 | 2020-11-30T16:24:05Z | 2024-06-27T13:02:56Z | https://github.com/MonetDB/MonetDB/issues/6524 | 753,617,794 | 6,524 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-01-25 11:16:26 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-02-12 16:12:08 +0100
## Comment 26117
Date: 2018-01-25 11:16:26 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36
Build Identifier:
The following select query with a case gets a compilation error.
Reproducible: Always
### Steps to Reproduce:
1. SELECT CASE WHEN 18 IN ( AVG ( + 12 ) + - + ( - 29 ) + - SUM ( + + 53 ) * - ( 73 ) ) THEN NULL WHEN ( - COUNT ( * ) * - SUM ( DISTINCT 80 ) ) IS NOT NULL THEN 15 END + CASE 27 WHEN - 68 * + CAST ( COALESCE ( 42, + 57 * 20 ) AS INTEGER ) THEN + 11 * 11 + - CASE - 16 + 25 WHEN 37 + 24 THEN 51 + 41 END / 4 END * + 0 AS col1;
### Actual Results:
The error:
SELECT: subquery result missing
### Expected Results:
A single NULL value.
## Comment 26123
Date: 2018-01-25 13:34:50 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [32dc8a911410](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=32dc8a911410) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=32dc8a911410](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=32dc8a911410)
Changeset description:
Added tests for bugs 6522 and 6523
## Comment 26134
Date: 2018-01-29 06:45:29 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [d8a3ac8e4d55](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=d8a3ac8e4d55) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=d8a3ac8e4d55](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=d8a3ac8e4d55)
Changeset description:
handle multiple aggregates in an in expression in the selection
fixes bug #6523
## Comment 26135
Date: 2018-01-29 06:45:46 +0100
From: @njnes
fixed
## Comment 26165
Date: 2018-02-05 10:00:34 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [fa7d796dadb4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=fa7d796dadb4) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=fa7d796dadb4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=fa7d796dadb4)
Changeset description:
Add test for bug #6523.
| Sqlitelogictest: Case statement subquery missing | https://api.github.com/repos/MonetDB/MonetDB/issues/6523/comments | 0 | 2020-11-30T16:24:02Z | 2024-06-27T13:02:55Z | https://github.com/MonetDB/MonetDB/issues/6523 | 753,617,762 | 6,523 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-01-25 10:57:04 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-02-12 16:12:08 +0100
## Comment 26116
Date: 2018-01-25 10:57:04 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36
Build Identifier:
The IN operator is generating compilation errors in some queries. As the error says, the inner query should output a single column as an union of all expressions there.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab2(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab2 VALUES (64,77,40), (75,67,58), (46,51,23);
3. SELECT ALL * FROM tab2 WHERE + col0 IN ( - 12, col0, - col1, col1 / + col0, col1 );
### Actual Results:
The error:
IN: inner query should return a single column
### Expected Results:
The result set:
--------------------
col0 | col1 | col2
------+------+------
64 | 77 | 40
75 | 67 | 58
46 | 51 | 23
## Comment 26122
Date: 2018-01-25 13:34:49 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [32dc8a911410](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=32dc8a911410) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=32dc8a911410](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=32dc8a911410)
Changeset description:
Added tests for bugs 6522 and 6523
## Comment 26132
Date: 2018-01-28 10:24:51 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [e69686af0fb1](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e69686af0fb1) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=e69686af0fb1](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=e69686af0fb1)
Changeset description:
fixed bug #6522, ie handle constants before correlated in part
| Sqlitelogictest: IN operator return a single column | https://api.github.com/repos/MonetDB/MonetDB/issues/6522/comments | 0 | 2020-11-30T16:23:59Z | 2024-06-27T13:02:54Z | https://github.com/MonetDB/MonetDB/issues/6522 | 753,617,704 | 6,522 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-01-24 17:46:53 +0100
From: @swingbit
To: SQL devs <<bugs-sql>>
Version: 11.27.11 (Jul2017-SP3)
CC: @lsidir, @njnes
Last updated: 2018-06-01 11:36:49 +0200
## Comment 26110
Date: 2018-01-24 17:46:53 +0100
From: @swingbit
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36
Build Identifier:
This is rather difficult to reproduce, I just hope that this description can point at the right direction.
Given the following table (notice the ordered index):
sql>\d all_obj_dict
CREATE TABLE "spinque"."all_obj_dict" (
"id" INTEGER NOT NULL,
"idstr" CHARACTER LARGE OBJECT NOT NULL,
"prob" FLOAT(51) NOT NULL,
CONSTRAINT "all_obj_dict_id_pkey" PRIMARY KEY ("id"),
CONSTRAINT "all_obj_dict_idstr_unique" UNIQUE ("idstr")
);
CREATE INDEX "idx_all_idstr" ON "spinque"."all_obj_dict" ("idstr");
During a join with this table (I don't actually have the exact query), I got the following:
2018-01-23 08:51:08 ERR production01[85]: *** Error in `/opt/monetdb/bin/mserver5': double free or corruption (out): 0x00007fda3826aa30 ***
2018-01-23 08:51:08 ERR production01[85]: ======= Backtrace: =========
2018-01-23 08:51:08 ERR production01[85]: /lib64/libc.so.6(+0x791eb)[0x7fda59b381eb]
2018-01-23 08:51:08 ERR production01[85]: /lib64/libc.so.6(+0x8285a)[0x7fda59b4185a]
2018-01-23 08:51:08 ERR production01[85]: /lib64/libc.so.6(cfree+0x4c)[0x7fda59b4528c]
2018-01-23 08:51:08 ERR production01[85]: /opt/monetdb/lib/libbat.so.15(GDKfree+0x27)[0x7fda5be1bd37]
2018-01-23 08:51:08 ERR production01[85]: /opt/monetdb/lib/libmonetdb5.so.23(freeMalBlk+0x3e)[0x7fda5c17c45e]
2018-01-23 08:51:08 ERR production01[85]: /opt/monetdb/lib/monetdb5/lib_sql.so(+0x3b389)[0x7fda5710d389]
2018-01-23 08:51:08 ERR production01[85]: /opt/monetdb/lib/monetdb5/lib_sql.so(SQLengineIntern+0x1a6)[0x7fda5710e8e6]
Recompiling in debug mode (thus with assertions enabled),
gave the following error:
2018-01-23 08:51:08 ERR production01[85]: could not find all_obj_dict.%idx_all_idstr
and stopped at the following assertion:
[Switching to thread 7 (Thread 0x7f0916fd5700 (LWP 24155))]
0 0x00007f091ad168df in raise () from /lib64/libc.so.6
(gdb) up
1 0x00007f091ad184da in abort () from /lib64/libc.so.6
(gdb) up
2 0x00007f091ad0ed67 in __assert_fail_base () from /lib64/libc.so.6
(gdb) up
3 0x00007f091ad0ee12 in __assert_fail () from /lib64/libc.so.6
(gdb) up
4 0x00007f09182c075d in exp_bin (be=0x7f090018edd0, e=0x7f09002cb6d0, left=0x7f09002cdce0, right=0x7f09002cddb0, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at rel_bin.c:577
577 assert(s);
Now, I find two things pretty strange:
1) The assertion itself (rel_bin.c:577):
if (!s && right) {
fprintf(stderr, "could not find %s.%s\n", (char*)e->l, (char*)e->r);
print_stmtlist(sql->sa, left);
print_stmtlist(sql->sa, right);
assert(s);
return NULL;
}
How can this assert ever succeed, with that if statement? Is this meant to signal "the impossible happened" ?
2) The error "could not find all_obj_dict.%idx_all_idstr"
- Is the % symbol expected?
- even after running analyze on the table, storage() shows no orderidx for it
- the file corresponding to that idx does exist on filesystem
Why was the idx not found?
Reproducible: Always
## Comment 26118
Date: 2018-01-25 11:43:29 +0100
From: @lsidir
Hi Roberto,
orderedidx are "not meant to support" or "do not support yet" string columns.
Hence, in the orderidx creation code if the column type is string, we return succesfully with no index created (this is one of those cases where you ask yourself what is a succesful index creation...). Now, sql thinks that the index is there, and thats where trouble starts.
We need to typecheck/safeguard at the sql level and give a message that string columns are not (yet) supported for orderidx.
## Comment 26119
Date: 2018-01-25 11:52:55 +0100
From: @swingbit
Hi Lefteris,
I was indeed under the impression that ordered idx on strings were already supported (changeset 222c6823450f).
No problem with not using them for now. I agree though that they should not be allowed in SQL if not implemented.
## Comment 26120
Date: 2018-01-25 11:57:38 +0100
From: @lsidir
Interesting... it seems that strings where supported at some point afterall but with no multi-merge parralel step.
I will keep looking into it then.
## Comment 26124
Date: 2018-01-25 13:35:34 +0100
From: @sjoerdmullender
1. CREATE INDEX has nothing to do with ORDERED INDEX. It (may?) create an index bat with some sort of hash.
2. The "%" in the name of the column is MonetDB/SQL's way of finding the (SQL-level) index bat.
3. CREATE ORDERED INDEX is only supported on a limited set of types.
4. ordered indexes can also be created in other cases (e.g. when you use ORDER BY on a column--internally, when a persistent bat is sorted).
5. these internally created ordered indexes are available for all "linear" types (i.e. types on which an order is defined), so including strings.
6. ordered indexes, as long as they are transient, may not make it to disk. If they are small, they fit in memory and are not memory mapped. That is the same for all heaps.
7. the assert has nothing to do with ordered indexes.
8. analyze does not create ordered indexes.
## Comment 26125
Date: 2018-01-25 13:45:20 +0100
From: @swingbit
Sjoerd:
1. The index WAS created with CREATE ORDERED INDEX. But \d displays it as CREATE INDEX
7. The assert itself may not have anything to with the ordered indices, but it was triggered because the ordered index was not found. After dropping the ordered index, the issue was gone.
8. I know that analyze doesn't create the index. But as far as I understand, it is supposed to check for its existence.
## Comment 26127
Date: 2018-01-25 13:58:11 +0100
From: @swingbit
NB: even though it is not easily reproducible, I wanted to report this to keep track of it. Somewhere, there is indeed a bug.
But it is not urgent for me. The explicit creation of those indices is something that made it to our production pipeline by mistake - it should only have been in development. So by not using them, I'm fine for now.
## Comment 26128
Date: 2018-01-25 17:29:46 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [b6d9a304b2a7](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=b6d9a304b2a7) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=b6d9a304b2a7](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=b6d9a304b2a7)
Changeset description:
Dump correct index type.
This fixes part of bug #6521.
## Comment 26129
Date: 2018-01-25 17:37:00 +0100
From: @sjoerdmullender
Indexes may get removed at the GDK level during operation. That is especially true for ordered and imprints indexes since they don't support updates at all. The problem may well be that the SQL layer created the index, but the GDK layer removed it. There is no way for the GDK layer to tell the SQL layer that it did this, so the SQL layer would have to check and maybe recreate the index.
## Comment 26262
Date: 2018-03-14 12:43:37 +0100
From: @njnes
It seems the relational optimizer made some mistake, ie one of the relational
operators could not find the proper expression in a lower level relational operator. To debug this I really need the query.
## Comment 26265
Date: 2018-03-14 13:29:53 +0100
From: @swingbit
Unfortunately I cannot find it back. I had tried to reproduce it artificially, but with no luck. It happened in a context that is hard to reproduce now, and I wouldn't know how to pull the trigger again.
For what concerns this bug report, I share Sjoerd's last comment: if SQL allows to explicitly create an index, but GDK can remove it without informing SQL, that is a problem. Either find a way to inform SQL, or don't allow explicit creation in the first place, I would say.
## Comment 26474
Date: 2018-06-01 11:36:49 +0200
From: @swingbit
Hi, just wanted to check whether there are news or plan on how to handle this.
In some cases I would like to create an ordered index, but if, as Sjoerd said, GDK can decide to remove it, then it's no useful.
Shouldn't indices that are explicitly created in SQL be not-removable by GDK?
| oderidx exists but cannot be found, and this goes through an always-failing assertion | https://api.github.com/repos/MonetDB/MonetDB/issues/6521/comments | 0 | 2020-11-30T16:23:54Z | 2024-06-27T13:02:53Z | https://github.com/MonetDB/MonetDB/issues/6521 | 753,617,650 | 6,521 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-01-24 15:02:36 +0100
From: @drstmane
To: SQL devs <<bugs-sql>>
Version: 11.27.11 (Jul2017-SP3)
CC: @njnes
Last updated: 2018-02-12 16:12:06 +0100
## Comment 26108
Date: 2018-01-24 15:02:36 +0100
From: @drstmane
This script (DDL + update statement; no data required!):
========
start transaction;
CREATE TABLE "PRIMARY" (
"ID" integer NOT NULL,
"LEN" real NOT NULL,
CONSTRAINT "PK_PRIMARY" PRIMARY KEY ("ID")
);
CREATE TABLE "FOREIGN" (
"ID" integer NOT NULL,
"POS" real NOT NULL,
CONSTRAINT "FK_FOREIGN" FOREIGN KEY ("ID") REFERENCES "PRIMARY" ("ID")
);
alter table "FOREIGN" add column "RelPos" real default NULL;
update "FOREIGN"
set "RelPos" = "POS" / (
select "LEN"
from "PRIMARY"
where "PRIMARY"."ID" = "FOREIGN"."ID"
)
where "POS" between 0 and (
select "LEN"
from "PRIMARY"
where "PRIMARY"."ID" = "FOREIGN"."ID"
)
;
rollback;
========
causes:
========
could not find L10.LEN
FOREIGN.POS
L13.L13
FOREIGN.POS
L13.L13
mserver5: .../Jul2017/source/MonetDB/sql/backends/monet5/rel_bin.c:575: exp_bin: Assertion `s' failed.
Thread 6 "mserver5" received signal SIGABRT, Aborted.
[Switching to Thread 0x7fffd3cc2700 (LWP 30631)]
0x00007ffff44e99fb in raise () from /lib64/libc.so.6
(gdb) bt
0 0x00007ffff44e99fb in raise () from /lib64/libc.so.6
1 0x00007ffff44eb800 in abort () from /lib64/libc.so.6
2 0x00007ffff44e20da in __assert_fail_base () from /lib64/libc.so.6
3 0x00007ffff44e2152 in __assert_fail () from /lib64/libc.so.6
4 0x00007fffeef05b2b in exp_bin (be=0x7fffbc0cb1f0, e=0x7fffbc0d6830, left=0x7fffbc0dded0, right=0x7fffbc0de2b0, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at .../Jul2017/source/MonetDB/sql/backends/monet5/rel_bin.c:575
5 0x00007fffeef055ac in exp_bin (be=0x7fffbc0cb1f0, e=0x7fffbc0d68e0, left=0x7fffbc0dded0, right=0x7fffbc0de2b0, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at .../Jul2017/source/MonetDB/sql/backends/monet5/rel_bin.c:515
6 0x00007fffeef1021f in rel2bin_groupby (be=0x7fffbc0cb1f0, rel=0x7fffbc0d83d0, refs=0x7fffbc0db1e0) at .../Jul2017/source/MonetDB/sql/backends/monet5/rel_bin.c:2753
7 0x00007fffeef19bff in subrel_bin (be=0x7fffbc0cb1f0, rel=0x7fffbc0d83d0, refs=0x7fffbc0db1e0) at .../Jul2017/source/MonetDB/sql/backends/monet5/rel_bin.c:4855
8 0x00007fffeef0ed36 in rel2bin_project (be=0x7fffbc0cb1f0, rel=0x7fffbc0d50b0, refs=0x7fffbc0db1e0, topn=0x0) at .../Jul2017/source/MonetDB/sql/backends/monet5/rel_bin.c:2465
9 0x00007fffeef19ba3 in subrel_bin (be=0x7fffbc0cb1f0, rel=0x7fffbc0d50b0, refs=0x7fffbc0db1e0) at .../Jul2017/source/MonetDB/sql/backends/monet5/rel_bin.c:4847
10 0x00007fffeef0ed36 in rel2bin_project (be=0x7fffbc0cb1f0, rel=0x7fffbc0d5730, refs=0x7fffbc0db1e0, topn=0x0) at .../Jul2017/source/MonetDB/sql/backends/monet5/rel_bin.c:2465
11 0x00007fffeef19ba3 in subrel_bin (be=0x7fffbc0cb1f0, rel=0x7fffbc0d5730, refs=0x7fffbc0db1e0) at .../Jul2017/source/MonetDB/sql/backends/monet5/rel_bin.c:4847
12 0x00007fffeef1746a in rel2bin_update (be=0x7fffbc0cb1f0, rel=0x7fffbc0d50e0, refs=0x7fffbc0db1e0) at .../Jul2017/source/MonetDB/sql/backends/monet5/rel_bin.c:4286
13 0x00007fffeef19cca in subrel_bin (be=0x7fffbc0cb1f0, rel=0x7fffbc0d50e0, refs=0x7fffbc0db1e0) at .../Jul2017/source/MonetDB/sql/backends/monet5/rel_bin.c:4872
14 0x00007fffeef19e4b in _subrel_bin (be=0x7fffbc0cb1f0, rel=0x7fffbc0d50e0, refs=0x7fffbc0db1e0) at .../Jul2017/source/MonetDB/sql/backends/monet5/rel_bin.c:4904
15 0x00007fffeef19f63 in output_rel_bin (be=0x7fffbc0cb1f0, rel=0x7fffbc0d50e0) at .../Jul2017/source/MonetDB/sql/backends/monet5/rel_bin.c:4927
16 0x00007fffeef35270 in sql_relation2stmt (be=0x7fffbc0cb1f0, r=0x7fffbc0d50e0) at .../Jul2017/source/MonetDB/sql/backends/monet5/sql_gencode.c:529
17 0x00007fffeef35479 in backend_dumpstmt (be=0x7fffbc0cb1f0, mb=0x7fffbc1346d0, r=0x7fffbc0d50e0, top=1, add_end=1, query=0x7fffbc104140 "update \"FOREIGN\"\n set \"RelPos\" = \"POS\" / (\n select \"LEN\"\n from \"PRIMARY\"\n where \"PRIMARY\".\"ID\" = \"FOREIGN\".\"ID\"\n )\n where \"POS\" between 0 and (\n select \"LEN\"\n from \"PRIMARY\"\n where \"PRIMARY\".\"ID\" = \"F"...)
at .../Jul2017/source/MonetDB/sql/backends/monet5/sql_gencode.c:569
18 0x00007fffeef35cd8 in backend_dumpproc (be=0x7fffbc0cb1f0, c=0x7ffff0301338, cq=0x7fffbc104440, r=0x7fffbc0d50e0) at .../Jul2017/source/MonetDB/sql/backends/monet5/sql_gencode.c:696
19 0x00007fffeeef9584 in SQLparser (c=0x7ffff0301338) at .../Jul2017/source/MonetDB/sql/backends/monet5/sql_scenario.c:1189
20 0x00007ffff7a131ae in runPhase (c=0x7ffff0301338, phase=1) at .../Jul2017/source/MonetDB/monetdb5/mal/mal_scenario.c:506
21 0x00007ffff7a132f2 in runScenarioBody (c=0x7ffff0301338) at .../Jul2017/source/MonetDB/monetdb5/mal/mal_scenario.c:539
22 0x00007ffff7a13495 in runScenario (c=0x7ffff0301338) at .../Jul2017/source/MonetDB/monetdb5/mal/mal_scenario.c:568
23 0x00007ffff7a1504a in MSserveClient (dummy=0x7ffff0301338) at .../Jul2017/source/MonetDB/monetdb5/mal/mal_session.c:473
24 0x00007ffff7a14b3a in MSscheduleClient (command=0x7fffbc0008d0 "\360\256\f\274\377\177", challenge=0x7fffd3cc1d6b "clN31XaVW", fin=0x7fffbc002970, fout=0x7fffc8002b70, protocol=PROTOCOL_9, blocksize=8190, compute_column_widths=0) at .../Jul2017/source/MonetDB/monetdb5/mal/mal_session.c:364
25 0x00007ffff7ab41b1 in doChallenge (data=0x7fffc80008d0) at .../Jul2017/source/MonetDB/monetdb5/modules/mal/mal_mapi.c:273
26 0x00007ffff768c471 in thread_starter (arg=0x7fffc8004c60) at .../Jul2017/source/MonetDB/gdk/gdk_system.c:485
27 0x00007ffff488f36d in start_thread () from /lib64/libpthread.so.0
28 0x00007ffff45c3e1f in clone () from /lib64/libc.so.6
(gdb) up
1 0x00007ffff44eb800 in abort () from /lib64/libc.so.6
(gdb)
2 0x00007ffff44e20da in __assert_fail_base () from /lib64/libc.so.6
(gdb)
3 0x00007ffff44e2152 in __assert_fail () from /lib64/libc.so.6
(gdb)
4 0x00007fffeef05b2b in exp_bin (be=0x7fffbc0cb1f0, e=0x7fffbc0d6830, left=0x7fffbc0dded0, right=0x7fffbc0de2b0, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at .../Jul2017/source/MonetDB/sql/backends/monet5/rel_bin.c:575
575 assert(s);
(gdb) li
570 s = stmt_project(be, ext, s);
571 if (!s && right) {
572 fprintf(stderr, "could not find %s.%s\n", (char*)e->l, (char*)e->r);
573 print_stmtlist(sql->sa, left);
574 print_stmtlist(sql->sa, right);
575 assert(s);
576 return NULL;
577 }
578 } break;
579 case e_cmp: {
(gdb) p be
$1 = (backend *) 0x7fffbc0cb1f0
(gdb) p *be
$2 = {console = 0, language = 83 'S', mvc = 0x7fffbc00ae20, out = 0x7fffc8002b70, output_format = OFMT_CSV, client = 0x7ffff0301338, mb = 0x7fffbc1346d0, mvc_var = 5, vtop = 1, q = 0x7fffbc104440}
(gdb) p ext
$3 = (stmt *) 0x0
(gdb) p *ext
Cannot access memory at address 0x0
(gdb) p right
$4 = (stmt *) 0x7fffbc0de2b0
(gdb) p *right
$5 = {type = st_list, op1 = 0x0, op2 = 0x0, op3 = 0x0, op4 = {aval = 0x7fffbc0de240, lval = 0x7fffbc0de240, stval = 0x7fffbc0de240, cval = 0x7fffbc0de240, idxval = 0x7fffbc0de240, tval = 0x7fffbc0de240, typeval = {type = 0x7fffbc0de240, digits = 3183328701, scale = 3183328701}, aggrval = 0x7fffbc0de240, funcval = 0x7fffbc0de240, rel = 0x7fffbc0de240}, nrcols = 0 '\000', key = 1 '\001', aggr = 0 '\000',
partition = 0 '\000', flag = 0, nr = 0, tname = 0x0, cname = 0x0, q = 0xbdbdbdbdbdbdbdbd}
(gdb) up
5 0x00007fffeef055ac in exp_bin (be=0x7fffbc0cb1f0, e=0x7fffbc0d68e0, left=0x7fffbc0dded0, right=0x7fffbc0de2b0, grp=0x0, ext=0x0, cnt=0x0, sel=0x0) at .../Jul2017/source/MonetDB/sql/backends/monet5/rel_bin.c:515
515 as = exp_bin(be, at, left, right, NULL, NULL, NULL, sel);
(gdb) li
510 list *l = sa_list(sql->sa);
511
512 for (en = attr->h; en; en = en->next) {
513 sql_exp *at = en->data;
514
515 as = exp_bin(be, at, left, right, NULL, NULL, NULL, sel);
516
517 if (as && as->nrcols <= 0 && left)
518 as = stmt_const(be, bin_first_column(be, left), as);
519 /* insert single value into a column */
(gdb) p be
$6 = (backend *) 0x7fffbc0cb1f0
(gdb) p *be
$7 = {console = 0, language = 83 'S', mvc = 0x7fffbc00ae20, out = 0x7fffc8002b70, output_format = OFMT_CSV, client = 0x7ffff0301338, mb = 0x7fffbc1346d0, mvc_var = 5, vtop = 1, q = 0x7fffbc104440}
(gdb) p at
$8 = (sql_exp *) 0x7fffbc0d6830
(gdb) p *at
$9 = {type = e_column, name = 0x7fffbc0d6890 "LEN", rname = 0x7fffbc0d68a0 "PRIMARY", l = 0x7fffbc0d6740, r = 0x7fffbc0d0590, f = 0x0, flag = 64, card = 3 '\003', tpe = {type = 0x1931940, digits = 24, scale = 0}, used = 0, p = 0x0}
(gdb) p left
$10 = (stmt *) 0x7fffbc0dded0
(gdb) p *left
$11 = {type = st_list, op1 = 0x0, op2 = 0x0, op3 = 0x0, op4 = {aval = 0x7fffbc0dde60, lval = 0x7fffbc0dde60, stval = 0x7fffbc0dde60, cval = 0x7fffbc0dde60, idxval = 0x7fffbc0dde60, tval = 0x7fffbc0dde60, typeval = {type = 0x7fffbc0dde60, digits = 3183328701, scale = 3183328701}, aggrval = 0x7fffbc0dde60, funcval = 0x7fffbc0dde60, rel = 0x7fffbc0dde60}, nrcols = 1 '\001', key = 0 '\000', aggr = 0 '\000',
partition = 0 '\000', flag = 0, nr = 0, tname = 0x0, cname = 0x0, q = 0xbdbdbdbdbdbdbdbd}
(gdb) p right
$12 = (stmt *) 0x7fffbc0de2b0
(gdb) p *right
$13 = {type = st_list, op1 = 0x0, op2 = 0x0, op3 = 0x0, op4 = {aval = 0x7fffbc0de240, lval = 0x7fffbc0de240, stval = 0x7fffbc0de240, cval = 0x7fffbc0de240, idxval = 0x7fffbc0de240, tval = 0x7fffbc0de240, typeval = {type = 0x7fffbc0de240, digits = 3183328701, scale = 3183328701}, aggrval = 0x7fffbc0de240, funcval = 0x7fffbc0de240, rel = 0x7fffbc0de240}, nrcols = 0 '\000', key = 1 '\001', aggr = 0 '\000',
partition = 0 '\000', flag = 0, nr = 0, tname = 0x0, cname = 0x0, q = 0xbdbdbdbdbdbdbdbd}
(gdb) p sel
$14 = (stmt *) 0x0
(gdb)
========
## Comment 26109
Date: 2018-01-24 15:07:32 +0100
From: @drstmane
ps:
It's actually the correlated subquery in the WHERE clause that causes the problem.
The correlated subquery in the SET clause appears to work without the correlated subquery in the WHERE clause.
The assertion / crash also occurs with only the correlated subquery in the WHERE clause, i.e., without the correlated subquery in the SET clause.
## Comment 26113
Date: 2018-01-24 18:34:32 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [49c6abdb8aa5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=49c6abdb8aa5) made by Stefan Manegold <Stefan.Manegold@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=49c6abdb8aa5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=49c6abdb8aa5)
Changeset description:
added test for Bug #6520 "UPDATE with correlated subquery causes assertion (or segfault)"
## Comment 26130
Date: 2018-01-27 14:12:38 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [1a9b08289b80](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=1a9b08289b80) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=1a9b08289b80](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=1a9b08289b80)
Changeset description:
fix bug #6520, add projection in update when needed
## Comment 26131
Date: 2018-01-27 14:13:07 +0100
From: @njnes
fixed by adding proper projection
| UPDATE with correlated subquery causes assertion (or segfault) | https://api.github.com/repos/MonetDB/MonetDB/issues/6520/comments | 0 | 2020-11-30T16:23:51Z | 2024-06-27T13:02:51Z | https://github.com/MonetDB/MonetDB/issues/6520 | 753,617,619 | 6,520 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-01-19 16:48:28 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-02-12 16:12:17 +0100
## Comment 26105
Date: 2018-01-19 16:48:28 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36
Build Identifier:
The following SQL query cannot be executed due to a nonexistent algebra join between BATs of different numeric atom types.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab2(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab2 VALUES (64,77,40), (75,67,58), (46,51,23);
3. SELECT * FROM tab2 WHERE col2 NOT IN ( - + col1, col2 * + col1, + - col0 * + + 36, + + col2, 84 * + + col0 ) AND NOT NULL IS NULL;
### Actual Results:
The following MAL error:
TypeException:user.main[48]:'algebra.join' undefined in: algebra.join(X_98531:bat[:int], X_98533:bat[:lng], nil:BAT, nil:BAT, false:bit, nil:lng);
### Expected Results:
An empty result set.
## Comment 26107
Date: 2018-01-19 17:00:04 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [63944bc2d4a4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=63944bc2d4a4) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=63944bc2d4a4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=63944bc2d4a4)
Changeset description:
Added tests for bugs 6518 and 6519 (enough bugs for some time)
## Comment 26112
Date: 2018-01-24 18:28:46 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [ca3d65de89c2](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ca3d65de89c2) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ca3d65de89c2](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ca3d65de89c2)
Changeset description:
fixes for bugs 6518 and 6519
added type checks in (not) in correlated case (only when needed)
in case of correlated (not) in we should add the identity column (and join on that)
## Comment 26114
Date: 2018-01-24 18:45:53 +0100
From: @njnes
added missing type checks
| Sqlitelogictest: algebra join between lng and int BATs undefined | https://api.github.com/repos/MonetDB/MonetDB/issues/6519/comments | 0 | 2020-11-30T16:23:49Z | 2024-06-27T13:02:50Z | https://github.com/MonetDB/MonetDB/issues/6519 | 753,617,589 | 6,519 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-01-19 16:41:01 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-02-12 16:12:21 +0100
## Comment 26104
Date: 2018-01-19 16:41:01 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36
Build Identifier:
The query bellow gets wrong results.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab2(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab2 VALUES (64,77,40), (75,67,58), (46,51,23);
3. SELECT DISTINCT - 15 * - 19 + - COUNT ( * ) * + 70 FROM tab2 WHERE ( col2 ) NOT IN ( + 12 + col0 );
### Actual Results:
145
### Expected Results:
75
## Comment 26106
Date: 2018-01-19 17:00:03 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [63944bc2d4a4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=63944bc2d4a4) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=63944bc2d4a4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=63944bc2d4a4)
Changeset description:
Added tests for bugs 6518 and 6519 (enough bugs for some time)
## Comment 26111
Date: 2018-01-24 18:28:44 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [ca3d65de89c2](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ca3d65de89c2) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ca3d65de89c2](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ca3d65de89c2)
Changeset description:
fixes for bugs 6518 and 6519
added type checks in (not) in correlated case (only when needed)
in case of correlated (not) in we should add the identity column (and join on that)
## Comment 26115
Date: 2018-01-24 18:46:50 +0100
From: @njnes
fixed by adding an identity join for correlated subqueries (not in)
| Sqlitelogictest: count aggregation with not in operator | https://api.github.com/repos/MonetDB/MonetDB/issues/6518/comments | 0 | 2020-11-30T16:23:45Z | 2024-06-27T13:02:49Z | https://github.com/MonetDB/MonetDB/issues/6518 | 753,617,546 | 6,518 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-01-18 12:13:41 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: martin.van.dinther, @njnes
Last updated: 2019-07-24 17:49:03 +0200
## Comment 26097
Date: 2018-01-18 12:13:41 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36
Build Identifier:
A SELECT with the IN operator fails in this scenario:
- 86 IN ( 47 * - 49 + 22 )
where -2281, the result of (47 * - 49 + 22) as a short, cannot be converted to a bte.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab1(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab1 VALUES(51,14,96);
3. INSERT INTO tab1 VALUES(85,5,59);
4. INSERT INTO tab1 VALUES(91,47,68);
5. SELECT - col1 FROM tab1 WHERE NOT - 86 IN ( 47 * - 49 + 22 );
### Actual Results:
The following error:
overflow in conversion of -2281 to bte.
### Expected Results:
The result set: -14, -5, -47
## Comment 26099
Date: 2018-01-18 12:37:33 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [ca5cee4ff0fd](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ca5cee4ff0fd) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ca5cee4ff0fd](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ca5cee4ff0fd)
Changeset description:
Added tests for bugs 6516 and 6517
## Comment 26101
Date: 2018-01-18 21:37:54 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [9ddf37267b1c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9ddf37267b1c) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=9ddf37267b1c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=9ddf37267b1c)
Changeset description:
fixes for bugs 6516 and 6517, improved type checking in the 'IN/NOT IN' cases
## Comment 26102
Date: 2018-01-18 21:38:12 +0100
From: @njnes
fixed type checking
| Sqlitelogictest overflow in conversion during MAL plan execution | https://api.github.com/repos/MonetDB/MonetDB/issues/6517/comments | 0 | 2020-11-30T16:23:43Z | 2024-06-27T13:02:48Z | https://github.com/MonetDB/MonetDB/issues/6517 | 753,617,509 | 6,517 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-01-18 11:53:09 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-02-12 16:12:05 +0100
## Comment 26096
Date: 2018-01-18 11:53:09 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36
Build Identifier:
There are still missing bat.append operations between different atoms. In this case appending a 70 bte to an integer BAT from an IN operation is failing.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab1(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab1 VALUES(51,14,96);
3. INSERT INTO tab1 VALUES(85,5,59);
4. INSERT INTO tab1 VALUES(91,47,68);
5. SELECT DISTINCT * FROM tab1 AS cor0 WHERE ( col0 ) NOT IN ( + col0, 70 );
### Actual Results:
A MAL error:
'bat.append' undefined in: bat.append(X_42:bat[:int], X_40:bte, true:bit);
### Expected Results:
An empty result set
## Comment 26098
Date: 2018-01-18 12:37:32 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [ca5cee4ff0fd](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ca5cee4ff0fd) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ca5cee4ff0fd](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ca5cee4ff0fd)
Changeset description:
Added tests for bugs 6516 and 6517
## Comment 26100
Date: 2018-01-18 21:37:53 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [9ddf37267b1c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9ddf37267b1c) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=9ddf37267b1c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=9ddf37267b1c)
Changeset description:
fixes for bugs 6516 and 6517, improved type checking in the 'IN/NOT IN' cases
## Comment 26103
Date: 2018-01-18 21:38:30 +0100
From: @njnes
fixed type checking
| Sqlitelogictest unknown bat append operation | https://api.github.com/repos/MonetDB/MonetDB/issues/6516/comments | 0 | 2020-11-30T16:23:39Z | 2024-06-27T13:02:47Z | https://github.com/MonetDB/MonetDB/issues/6516 | 753,617,466 | 6,516 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-01-16 17:59:34 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-02-12 16:12:13 +0100
## Comment 26084
Date: 2018-01-16 17:59:34 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36
Build Identifier:
Inserting a second interval value in a table results in a 0 value instead of NULL.
This happened in MonetDBLite-Java tests (after the merge) and it's happening as well in the latest development version of Jul2017 branch. (I am looking into this as well)
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE testnullsb (g INTERVAL minute to second);
2. INSERT INTO testnullsb VALUES (NULL);
3. SELECT g FROM testnullsb;
### Actual Results:
0 value
### Expected Results:
NULL value
## Comment 26085
Date: 2018-01-16 18:21:39 +0100
From: @PedroTadim
From hg bisect, the commit 64922:b1efe52241ae created the bug.
## Comment 26086
Date: 2018-01-17 10:51:50 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [4b923b30c82c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=4b923b30c82c) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=4b923b30c82c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=4b923b30c82c)
Changeset description:
Added test for bug #6515
## Comment 26090
Date: 2018-01-18 08:07:48 +0100
From: @njnes
handle null in calc.second_interval
## Comment 26093
Date: 2018-01-18 08:08:07 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [248aa8f21695](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=248aa8f21695) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=248aa8f21695](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=248aa8f21695)
Changeset description:
Fixes for bugs 6510, 6512, 6514 and 6515
6510 start sum's at 0 instead of nil in case of count/sum 2 step aggregation
6512 make sure we only call bat.single on constants (ie not on bats).
6514 handle NULL (type void) on both sides of an compare operator, ie cast to a valid type.
6515 handle null in calc.second_interval
| Insert null second interval value results in 0 | https://api.github.com/repos/MonetDB/MonetDB/issues/6515/comments | 0 | 2020-11-30T16:23:37Z | 2024-06-27T13:02:46Z | https://github.com/MonetDB/MonetDB/issues/6515 | 753,617,421 | 6,515 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-01-12 15:04:44 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2018-02-12 16:12:04 +0100
## Comment 26078
Date: 2018-01-12 15:04:44 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36
Build Identifier:
A select query between ranges of two NULL values results in a type exception. MonetDB should do the comparison using other type than "void".
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab1(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab1 VALUES(51,14,96);
3. INSERT INTO tab1 VALUES(85,5,59);
4. INSERT INTO tab1 VALUES(91,47,68);
5. SELECT ALL col1 FROM tab1 WHERE NULL <= NULL;
### Actual Results:
The followin type exception:
TypeException:user.s6_1[16]:'calc.<=' undefined in: X_26:bit := calc.<=(X_24:void, X_25:void);
### Expected Results:
An empty result set.
Other sqlitelogictests are following for other predicate algebra operators: >=, <, >, ==, !=
## Comment 26080
Date: 2018-01-12 15:12:44 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [9a05add39d7b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9a05add39d7b) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=9a05add39d7b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=9a05add39d7b)
Changeset description:
Added tests for bugs 6513 and 6514
## Comment 26089
Date: 2018-01-18 08:07:07 +0100
From: @njnes
handle NULL (type void) on both sides of an compare operator, ie cast to a valid type.
## Comment 26092
Date: 2018-01-18 08:08:05 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [248aa8f21695](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=248aa8f21695) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=248aa8f21695](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=248aa8f21695)
Changeset description:
Fixes for bugs 6510, 6512, 6514 and 6515
6510 start sum's at 0 instead of nil in case of count/sum 2 step aggregation
6512 make sure we only call bat.single on constants (ie not on bats).
6514 handle NULL (type void) on both sides of an compare operator, ie cast to a valid type.
6515 handle null in calc.second_interval
| Sqlitelogictest: Range query between NULL values not possible | https://api.github.com/repos/MonetDB/MonetDB/issues/6514/comments | 0 | 2020-11-30T16:23:34Z | 2024-06-27T13:02:45Z | https://github.com/MonetDB/MonetDB/issues/6514 | 753,617,388 | 6,514 |
[
"MonetDB",
"MonetDB"
] | Date: 2018-01-12 14:52:00 +0100
From: @PedroTadim
To: SQL devs <<bugs-sql>>
Version: -- development
Last updated: 2018-03-29 15:39:22 +0200
## Comment 26076
Date: 2018-01-12 14:52:00 +0100
From: @PedroTadim
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36
Build Identifier:
MonetDB generates a nonexistent MAL function call in select query, followed by an overflow. It is not possible to obtain a long integer with a product of two byte atoms, so the correct MAL call should be generated.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tab1(col0 INTEGER, col1 INTEGER, col2 INTEGER);
2. INSERT INTO tab0 VALUES(97,1,99);
3. INSERT INTO tab0 VALUES(15,81,47);
4. INSERT INTO tab0 VALUES(87,21,10);
5. SELECT DISTINCT - + col2 * 68 * - + 22 FROM tab0 AS cor0;
### Actual Results:
The following error, followed by an overflow:
TypeException:user.main[14]:'calc.*' undefined in: X_1008:lng := calc.*(X_1006:bte, X_1007:bte);
ERROR: 22003!overflow in calculation 68*-22.
### Expected Results:
A single column result set with values 148104, 14960, 70312.
## Comment 26077
Date: 2018-01-12 14:56:49 +0100
From: @PedroTadim
First step is wrong, sorry:
1. CREATE TABLE tab0(col0 INTEGER, col1 INTEGER, col2 INTEGER);
## Comment 26079
Date: 2018-01-12 15:12:42 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [9a05add39d7b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9a05add39d7b) made by Pedro Ferreira <pedro.ferreira@monetdbsolutions.com> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=9a05add39d7b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=9a05add39d7b)
Changeset description:
Added tests for bugs 6513 and 6514
## Comment 26083
Date: 2018-01-16 15:17:31 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [ff833040fa32](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ff833040fa32) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ff833040fa32](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=ff833040fa32)
Changeset description:
Enabled the symbol FULL_IMPLEMENTATION in gdk_calc.c.
And made the newly available interfaces accessible through MAL.
This fixes bug #6513.
## Comment 26154
Date: 2018-02-01 10:30:53 +0100
From: @sjoerdmullender
Works in default branch.
## Comment 26322
Date: 2018-03-29 15:39:22 +0200
From: @sjoerdmullender
The Mar2018 version has been released.
| Sqlitelogictest: Wrong MAL plan generation for column product | https://api.github.com/repos/MonetDB/MonetDB/issues/6513/comments | 0 | 2020-11-30T16:23:31Z | 2024-06-27T13:02:44Z | https://github.com/MonetDB/MonetDB/issues/6513 | 753,617,338 | 6,513 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.