We am having constant index corruption problems since moving from 4.0.17 to 4.1.3. All tables are InnoDB and we're using the file-per-table setting on Solaris9.
Every couple days another few tables will show up corrupt and queries start missing things. The database has never gone down, there have never been any hard errors (drive failures, etc.), no crashes, no messages in the .err file, etc. It stays up the whole time and just keeps getting corrupt tables. I then run CHECK TABLE, which reports them corrupt and puts messages in the .err file saying that the indexes don't contain enough entries. It is not always the same tables, although there are a few that just seem to be doomed and are constantly having the problem. The only way I know to fix the failing tables is to dump and reload them. Since many of these tables have several million rows and are growing quickly, this problem is getting more and more miserable every day. This never happened once with the same tables in 4.0.17, where there was never a table corruption problem. So, I'm having a hard time believing it would have something to do with how we're using our tables. Has this happened to anyone else, or does anyone have a suggestion? Thanks -keith -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
