On Mon, 2002-10-07 at 17:40, Victoria Reznichenko wrote: > MJF> One of the machines is simultaneously slave to one server and master to > MJF> two others; on that machine, if a "RESET MASTER" or "PURGE MASTER LOGS" > MJF> is executed while its slave thread is performing a query, mysqld > MJF> reproducibly dies.
> I also saw the similar bug report, but I couldn't repeat it. Could you > test it with our binaries? As mentioned, the original crash was observed with the stock Solaris 8 3.23.52-Max binary. The result was the same, just without a gdb-able backtrace. Just to confuse matters, I put the install back to 3.23.42 and could still make it fall over with a stream of INSERTs (just input from a local client and not even the slave thread) and a RESET MASTER concurrently. So, either we've been living very, very luckily for about a year, or this is a problem specific to Solaris 8. (The old installation, which suffered from a number of bugs but never this one, was Sol7/3.23.42) My job is currently to fix this, so I'll try to spread some test cases around a selection of machines to find more information. Cheers, -Matt. --------------------------------------------------------------------- Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
