Bug #19410 Test 'kill' fails on Windows + SCO
Submitted: 27 Apr 2006 19:58 Modified: 30 Jan 2007 3:54
Reporter: Kent Boortz Email Updates:
Status: Closed Impact on me:
None 
Category:MySQL Server Severity:S2 (Serious)
Version:4.1.20 + 5.0.21 + newer OS:Microsoft Windows (Windows 2003 x64 + SCO OSR 6)
Assigned to: Magnus Blåudd

[27 Apr 2006 19:58] Kent Boortz
Description:
The test 'kill' fails with

 mysqltest: At line 91: query 'reap' succeeded
 - should have failed with errno 1053...

How to repeat:
Run the test on Windows x64
[20 Jul 2006 13:05] Joerg Bruehe
Symptom:

=====
kill                           [ fail ]

Errors are (from /PATH/mysqltest-time) :
mysqltest: At line NNN: query 'reap' succeeded - should have failed with errno 1053...
(the last lines may be the most important ones)
=====

Observed on Windows x64 (5.0) + SCO OSR 6 (4.1).

SCO with low probability only: 
4.1.20 once in "pro", PS protocol,
4.1.21 once in "debug", PS protocol,
so it might be a timing issue ?
[8 Dec 2006 15:45] Bugs System
A patch for this bug has been committed. After review, it may
be pushed to the relevant source trees for release in the next
version. You can access the patch from:

  http://lists.mysql.com/commits/16660

ChangeSet@1.2584, 2006-12-08 16:43:50+01:00, msvensson@neptunus.(none) +1 -0
  Bug#19410 Test 'kill' fails on Windows + SCO
   - Use "mysql_field_count" to determine if there is a need to
     call "mysql_store_result"
[8 Dec 2006 16:09] Bugs System
A patch for this bug has been committed. After review, it may
be pushed to the relevant source trees for release in the next
version. You can access the patch from:

  http://lists.mysql.com/commits/16667

ChangeSet@1.2586, 2006-12-08 17:09:07+01:00, msvensson@neptunus.(none) +1 -0
  Bug#19410 Test 'kill' fails on Windows + SCO
[30 Jan 2007 3:54] Paul Dubois
Noted in 4.1.23, 5.0.36, 5.1.15 changelogs.

mysqltest incorrectly tried to retrieve result sets for some queries
where no result set was available.
[28 Feb 2007 8:01] Bugs System
A patch for this bug has been committed. After review, it may
be pushed to the relevant source trees for release in the next
version. You can access the patch from:

  http://lists.mysql.com/commits/20724

ChangeSet@1.2618, 2007-02-28 09:01:23+01:00, msvensson@pilot.blaudden +1 -0
  Bug#19410 Test 'kill' fails on Windows + SCO
  - Use "mysql_stmt_field_count" to determine if there is a need to
   call "mysql_stmt_store_result"
[28 Feb 2007 8:11] Bugs System
A patch for this bug has been committed. After review, it may
be pushed to the relevant source trees for release in the next
version. You can access the patch from:

  http://lists.mysql.com/commits/20727

ChangeSet@1.2610, 2007-02-28 09:10:38+01:00, msvensson@pilot.blaudden +1 -0
  Bug#19410 Test 'kill' fails on Windows + SCO
  - Use "mysql_stmt_field_count" to determine if there is a need to
  call "mysql_stmt_store_result"
[24 May 2007 23:29] Peter O'Gorman
This test still fails for us on x86_64 red hat enterprise linux 5 with mysql-5.0.38 (yes, I know it is not the latest, but this bug is supposed to be fixed in that version), built with the native gcc-4.1. about 20% of the time.
[25 May 2007 9:07] Magnus Blåudd
Hi,

if you see the problem on a different platform than what was reported for this bug you need to file a spearate bug report.

Please remember to include the exact output so we know where. Preferrably I'd like to see a 100% repeatable test case but if you get it for 1/5 that whould be enough to take a look.

If it is the testcase for bug#14851 that fails, you should note that the testcase contains a "--sleep 1" command which makes it a little bit unreliable. We try to kill a long running query "in the middle" and if you have a fast machine that query could take less than the 1 second + roundtrip time(s) => the query could have suceeded before it was killed.

/ Magnus