Bug #7735 Test failure: 'synchronization' ('CREATE TABLE ...: Table ... already exists')
Submitted: 7 Jan 2005 18:29 Modified: 31 Mar 2005 12:05
Reporter: Joerg Bruehe Email Updates:
Status: Can't repeat Impact on me:
None 
Category:MySQL Server Severity:S3 (Non-critical)
Version:4.1.9 + 5.0.3-pre OS:FreeBSD (FreeBSD 4.7 + MacOSX 6.8)
Assigned to: Ramil Kalimullin CPU Architecture:Any

[7 Jan 2005 18:29] Joerg Bruehe
Description:
With today's pull, a 'classic' build (=without InnoDB, without NDB) failed test 'synchronization' with this symptom:

=== cut ===
Errors are (from /Users/mysqldev/powermacg4/test/mysql-classic-4.1.9-apple-darwin6.8-powerpc/mysql-test/var/log/mysqltest-time) :
/Users/mysqldev/powermacg4/test/mysql-classic-4.1.9-apple-darwin6.8-powerpc/bin/mysqltest: At line 17: query 'CREATE TABLE t2 LIKE t1' failed: 1050: Table 't2' already exists
=== cut ===

Symptom is identical on 'cane', 'cane-linuxthreads', and 'powermacg4',
the test passes on all other platforms (including 'powermacg5'!) (except 'butch': compile failure).

Latest changeset:
  1.2190.1.1 05/01/07 12:29:15 lenz@mysql.com +1 -0
  - Updated Do-pkg to include the MySQL Preference Pane into the Mac OS X
    installation disk images

How to repeat:
Build and run on an affected platform.
[25 Feb 2005 12:50] Joerg Bruehe
Re-occurred on machine 'ds9' (OpenBSD) in 4.1.10 based on changeset
  1.2180 05/02/12 19:17:33 guilhem@mysql.com +1 -0
  new static archive_inited variable, so that archive_db_end() will do something only if archive_db_init() was run before
  (protection against destroying uninited mutex in the case where mysqld fails early at startup (before archive_db_init() was called)
  and then calls archive_db_end() to clean up).

In four (4) test runs (pro-commercial + pro-gpl, both default + '--ps-protocol'), the error happened only in one run, so it seems to be some race condition or similar thing.
After 'synchronization', also 'system_mysql_db' and 'system_mysql_db_fix' failed, but to me these look like consequences.
Messages from the failing run:

synchronization                [ fail ]

Errors are (from ...) :
.../mysql-pro-4.1.10-unknown-openbsd3.6-i386/bin/mysqltest: At line 32: query 'SHOW CREATE TABLE t2' failed: 1146: Table 'test.t2' doesn't exist
(the last lines may be the most important ones)

Ending Tests
Shutting-down MySQL daemon

Master shutdown finished
Slave shutdown finished
Resuming Tests

system_mysql_db                [ fail ]

Errors are (from ...) :
.../mysqltest: At line 14: Result length mismatch
(the last lines may be the most important ones)
Below are the diffs between actual and expected results:
-------------------------------------------------------
*** r/system_mysql_db.result    Fri Feb 18 17:51:55 2005
--- r/system_mysql_db.reject    Thu Feb 24 23:38:52 2005
***************
*** 128,130 ****
--- 128,132 ----
  ) ENGINE=MyISAM DEFAULT CHARSET=utf8 COLLATE=utf8_bin COMMENT='Column privileges'
  show tables;
  Tables_in_test
+ t1
+ t2
-------------------------------------------------------
Please follow the instructions outlined at
http://www.mysql.com/doc/en/Reporting_mysqltest_bugs.html
to find the reason to this problem and how to report this.

Ending Tests
Shutting-down MySQL daemon

Master shutdown finished
Slave shutdown finished
Resuming Tests

system_mysql_db_fix            [ fail ]

Errors are (from ...) :
ERROR 1146 (42S02) at line 28: Table 'test.func' doesn't exist
ERROR 1146 (42S02) at line 30: Table 'test.columns_priv' doesn't exist
ERROR 1146 (42S02) at line 37: Table 'test.tables_priv' doesn't exist
ERROR 1054 (42S22) at line 55: Unknown column 'Grant_priv' in 'where clause'
ERROR 1054 (42S22) at line 109: Unknown column 'Type' in 'columns_priv'
ERROR 1060 (42S21) at line 115: Duplicate column name 'type'
ERROR 1054 (42S22) at line 123: Unknown column 'Show_db_priv' in 'where clause'
.../mysqltest: At line 83: Result length mismatch
(the last lines may be the most important ones)

Ending Tests
Shutting-down MySQL daemon

Master shutdown finished
Slave shutdown finished
Resuming Tests

system_mysql_db_refs           [ pass ]
[28 Feb 2005 19:59] Joerg Bruehe
Same effect in 5.0.3-pre, also on "ds9". Build based on changeset
  1.1849 05/02/25 11:48:46 joreland@mysql.com +1 -0
  Merge mysql.com:/home/jonas/src/mysql-4.1
  into mysql.com:/home/jonas/src/mysql-5.0

Happened only in second test run ('--ps-protocol').