MySQL Bugs: #9279: [ERROR] /usr/local/mysql/bin/mysqld: Got error 127 from storage engine

Bug #9279	[ERROR] /usr/local/mysql/bin/mysqld: Got error 127 from storage engine
Submitted:	18 Mar 2005 13:45	Modified:	24 Apr 2005 4:39
Reporter:	Wolfgang Weghorst	Email Updates:
Status:	No Feedback	Impact on me:	None
Category:	MySQL Server: MyISAM storage engine	Severity:	S2 (Serious)
Version:	4.1.10a	OS:	IBM AIX (AIX 5.2 Patchlevel 2)
Assigned to:		CPU Architecture:	Any

Description:
We ware using MySql (4.1.10a, 32 bin) in conjunction with Bugzilla. The attachments-table has reached a somewhat magical size of 1023,9M.
Further inserts into this table are immediately leading to an crash (corruption) of this table.
Even after an Repair, the table crashes again.

Checking the table in MySql-Admin gives the following messages:
------------------------------------------------------------------------------------------------------------------------
Status checked for table kdtest_bugs2173.attachments.
Table is marked as crashed
------------------------------------------------------------------------------------------------------------------------
Status checked for table kdtest_bugs2173.attachments.
Size of datafile is: 1073741312 Should be: 1074331452
------------------------------------------------------------------------------------------------------------------------
Status checked for table kdtest_bugs2173.attachments.
Corrupt

On our Test-System the same operation is possible: I have copied the table to our testsystem, and everything is okay.
The Test-System in running with AIX 5.2 Pachlevel 4 and MySQL, (4.1.10a, 32 bin).

There is more then sufficient space on both machines.

Do you have any advice or idea's related to our problem?

Regards,

Wolfgang Weghorst

How to repeat:
The bug can be repeated on our Production-System if we are inserting
another row in the mentioned table.
The table is in an fine condition before the insert-statement is executed.

hi

is there any way you can provide us with a test-case that reproduces this behaviour ?  error 127 means that the record-file is crashed, and this could have happened because of many reasons.

thanks

No feedback was provided for this bug for over a month, so it is
being suspended automatically. If you are able to provide the
information that was originally requested, please do so and change
the status of the bug back to "Open".