Thanks to all who have sent good advice on tracking down
this WAIT state problem I have. It does appear to be disk
related, but we have had the SAN and Switch checked out and
it's only running at 25% throughput.
However I'm now seeing something interesting.
If I do a ps -flem and look at the WCHAN (wait channel) to
see what my processes are waiting on it now says 'event'.
Eg:
F S UID PID PPID %CPU PRI NI RSS WCHAN STARTED TIME COMMAND
8001 S < 0 525089 524289 0.0 0 -44 344K event Nov 25 0:01.12 /usr/sbin/xntpd -g -c /etc/ntp.conf
8001 S 0 525124 524289 0.0 44 0 192K event Nov 25 0:08.62 /usr/sbin/snmpd
8001 S 0 525137 524289 0.0 44 0 168K event Nov 25 0:00.32 /usr/sbin/svrMgt_mib
8001 S 0 525138 524289 0.0 44 0 224K event Nov 25 0:02.05 /usr/sbin/svrSystem_mib
8001 S 0 525139 524289 0.0 44 0 200K event Nov 25 0:02.89 /usr/sbin/os_mibs
8001 S < 0 525140 524289 0.0 43 -1 288K event Nov 25 0:00.04 /usr/sbin/pmgrd
8001 I 0 525141 524289 0.0 44 0 616K event Nov 25 0:01.65 /usr/sbin/clu_mibs
8001 S 0 525145 524289 0.0 44 0 1.2M event Nov 25 0:00.46 /usr/sbin/cpq_mibs
Previously I was just seeing memory addresses as the WCHAN.
Does anyone know what the 'event' label means?
Thanks,
Paul
Received on Thu Nov 27 2003 - 17:55:43 NZDT