I've upgraded a database server from:
2 x pIII 700MHZ 784Mb
to
2 x pIII 1ghz 2gig ecc reg
and have been having some sick database lockups since.
9:10pm up 6:11, 3 users, load average: 167.29, 108.42, 49.33 -
unbelieveable load average.
top shows 137 mysql 'threads?'
CPU0 states: 99.26% user, 0.15% system, 0.0% nice, 0.0% idle
CPU1 states: 99.4% user, 0.37% system, 0.0% nice, 0.0% idle
both cpu's are pegged
Mem: 2061960K av, 1427272K used, 634688K free, 0K shrd, 5952K buff
Swap: 522104K av, 0K used, 522104K free 1311104K cached
Memory looks..ok.
The disk is a raid 5 scsi array off of a adaptec(dpt) 2100S..the old raid
controller was a 3200S (dual channel version)
The array was swapable..so its the exact same data. I just moved the array
to the new machine...and i dont see any pattern as to when it
locks...seems a query just gets hung...and then stuff just keep stacking
up till its completely locked. The previous machine/setup ran
perfect...never anything like this at all. I'm using the same version of
slackware on both machines...Started with 2.4.16, tried 2.4.17 and am
using 2.4.12 kernel now (2.4.12 is the same kernel as the old
machine) Same version of mysql as well.... I need to exhaust all my
options before moving it back to the old machine... Anyone with some
ideas? possible bad ram?? maybe corrupt datasomewhere? What is the best
check practice for me..I did a myisamck -r on all tables..and sometimes
some say is usable but needs fixing....... Thanks.
---------------------------------------------------------------------
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
