Bug #18572 Partitions: crash on ALTER TABLE during stress testing
Submitted: 28 Mar 2006 9:18 Modified: 16 Apr 2006 14:45
Reporter: Matthias Leich Email Updates:
Status: Closed Impact on me:
Category:MySQL Server: Partitions Severity:S1 (Critical)
Version:5.1 OS:
Assigned to: Mikael Ronström CPU Architecture:Any

[28 Mar 2006 9:18] Matthias Leich
I get a crash of the server when doing the stress test, where several
threads do INSERT/UPDATE/DELETE and 
in parallel.
The crash occurs during the execution of the ALTER TABLE statement.
snip of the stacktrace:
0x82358c1 handle_segfault + 437
0xffffe420 _end + -144226064
0x831f060 _Z13rr_sequentialP14st_read_record + 44
0x83452d1 _Z24copy_data_between_tablesP8st_tableS0_R4ListI12create_fieldE15enum_duplicatesbjP8st_orderPyS8_ + 1625
0x8343cca _Z17mysql_alter_tableP3THDPcS1_P24st_ha_create_informationP13st_table_listR4ListI12create_fieldERS6_I3KeyEjP8st_order15enum_dup + 10338
0x8251d73 _Z21mysql_execute_commandP3THD + 7029
0x825a2bd _Z11mysql_parseP3THDPcj + 569
0x824e823 _Z16dispatch_command19enum_server_commandP3THDPcj + 2063
0x824e008 _Z10do_commandP3THD + 570
0x824d07d handle_one_connection + 827
0x40050aa7 _end + 929853303
0x40181c2e _end + 931102974

My environment:
   - Intel PC with Linux(SuSE 9.3)
   - MySQL compiled from source
         Version 5.1 last ChangeSet@1.2231.1.1, 2006-03-27

How to repeat:
Please ask me if necessary, because some special files and the
relative complicated internal QA stress test suite "system_1" 
are needed.
[28 Mar 2006 9:19] Matthias Leich

Attachment: stacktrace (application/octet-stream, text), 4.09 KiB.

[28 Mar 2006 9:32] Matthias Leich
Two additional observations might be of importance:
- There are about 4000 lines
  <local path to mysqld>: missing DBUG_RETURN or DBUG_VOID_RETURN 
                                      macro in function "log_slow_statement"
   within master.err
- the protocol of one thread executing "tb1_eng1_ins.test" 
  (Insert into the table which is also target of the ALTER TABLE command
   via stored procedure containing a cursor)
  shows, that it already suffered from
  Bug#18552: Partitions: SP with cursor, autoincrement PK,  duplicate key on insert
[13 Apr 2006 9:21] Matthias Leich
I checked with MySQL 5.1 last ChangeSet@1.2325, 2006-04-11
and was unable to reproduce the crash.
So it looks like a fix for another bug solved the problem.
[13 Apr 2006 9:38] Mikael Ronström
This bug was fixed by previous patch, most likely the one that fixed
bug #18573, this patch will appear in 5.1.10
[16 Apr 2006 14:45] Jon Stephens
Thank you for your bug report. This issue has been committed to our
source repository of that product and will be incorporated into the
next release.

If necessary, you can access the source repository and build the latest
available version, including the bugfix, yourself. More information 
about accessing the source trees is available at

Additional info:

Documented bugfix in 5.1.10 changelog; closed.