Description:
This happens on Solaris with 5.5.3-m3 when 'datatype test (uses BLOB's and TEXT) is used to run in SystemQA upgrade/downgrade framework.
mysql_upgrade.out says -
Running 'mysql_fix_privilege_tables'...
mysql_upgrade.err says -
ERROR 2013 (HY000) at line 153: Lost connection to MySQL server during query
ERROR 2006 (HY000) at line 154: MySQL server has gone away
ERROR 2006 (HY000) at line 156: MySQL server has gone away
ERROR 2006 (HY000) at line 158: MySQL server has gone away
ERROR 2006 (HY000) at line 159: MySQL server has gone away
ERROR 2006 (HY000) at line 160: MySQL server has gone away
In server log at that point of time:
------------------------------------------
100403 9:38:08 [Note] /export/home/tmp/nidhi/mysql_553m3/bin/mysqld: ready for connections.
Version: '5.5.3-m3' socket: '/tmp/mysql.sock' port: 50004 MySQL Community Server (GPL)
100403 9:38:36 InnoDB: ERROR: the age of the last checkpoint is 9442510,
InnoDB: which exceeds the log group capacity 9433498.
InnoDB: If you are using big BLOB or TEXT rows, you must set the
InnoDB: combined size of log files at least 10 times bigger than the
InnoDB: largest such row.
100403 9:39:32 [Note] /export/home/tmp/nidhi/mysql_553m3/bin/mysqld: Normal shutdown
100403 9:39:32 [Note] Event Scheduler: Purging the queue. 7 events
100403 9:39:34 [Warning] /export/home/tmp/nidhi/mysql_553m3/bin/mysqld: Forcing close of thread 5 user: 'root'
100403 9:39:37 InnoDB: Starting shutdown...
100403 9:39:39 InnoDB: Shutdown completed; log sequence number 70350530
100403 9:39:39 [Note] /export/home/tmp/nidhi/mysql_553m3/bin/mysqld: Shutdown complete
How to repeat:
Need to analyze.