From: jing shen (jshen@christmas.9966.org)
Date: Tue May 10 2005 - 23:18:26 EDT
Hi sunmanagers#,
I installed SEtoolkit 3.4 on Solaris 9(two CPUs) which runs a BIND9.3.1
cache
server.
After putting the system online, I tried to catch CPU load with both
mpstat and SEtoolkit.
I run mpstate as following:
%mpstat 1
CPU minf mjf xcal intr ithr csw icsw migr smtx srw syscl usr sys wt idl
0 0 0 1 415 410 938 7 59 2 0 1554 2 2 0 96
1 0 0 4 224 121 1001 7 61 1 0 1595 2 2 0 96
After modify DELAY macro in pwatch.se to 2 seconds, I run :
% se pwatch.se 276 ( 276 is named process id)
named used 6.5 %CPU, 4.9 %CPU average so far
named used 8.0 %CPU, 4.9 %CPU average so far
I noticed that setoolkit report much more CPU time used than mpstat
reports.
Why? Which data is more reliable ?
thanks
-- -- -- Regards Joe Shen ****************************************** * The sunshine of lifetime is made up of * * little beams which is bright all the * * time. * ****************************************** _______________________________________________ sunmanagers mailing list sunmanagers@sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers
This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 23:30:41 EDT