Re: 'latch free' contention...

NetComrade wrote:
>
> recently I had a spiked load on my db, there is no io problems, or so
> it seems (no iowait in top, which also seems a bit weird, b/c it
> usually states something like 6-7 percent).
> here's what I got from v$system_wait
>
> SQLWKS> select * from v$system_event
> 2> where average_wait>10
> 3> order by lower(event)
> 4>
> EVENT
> TOTAL_WAIT TOTAL_TIME TIME_WAITE AVERAGE_WA
> ----------------------------------------------------------------
> ---------- ---------- ---------- ----------
> buffer busy waits
> 16727 2508 733987 43.880373
> enqueue
> 26599 7215 2335725 87.8125117
> log buffer space
> 86 48 6374 74.1162791
> log file switch (checkpoint incomplete)
> 6605 6540 676617 102.440121
> log file switch completion
> 2217 82 51853 23.3888137
> pmon timer
> 298879 287178 60709272 203.123244
> rdbms ipc message
> 3686847 591493 474271855 128.638876
> smon timer
> 1992 1975 60703882 30473.8363
> switch logfile command
> 11 0 234 21.2727273
> write complete waits
> 7 6 617 88.1428571
>
> The thing that I don't understand the most is those timers.
> Log switch problem has been 'kindof' fixed.. (there were some
> contention on the disk).
> Any hints on what to do here? I am not asking to fix the problem..
> just some hints :) I can't find any useful documentation on these
> issues.
>
> Thanx.
> ---------------
> In case I forgot to mention:
> We use Oracle 7.3.4 and 8.1.6 on Solaris 2.6, 2.7 boxes
> ---------------
> Andrey Dmitriev eFax: (978) 383-5892 Daytime: (917) 750-3630
> AOL: NetComrade ICQ: 11340726 remove NSPAM to email

Most timers you will not need to worry about , since they typically represent various Oracle processes simply waiting for something to do.

HTH

-- 
===========================================
Connor McDonald
http://www.oracledba.co.uk

We are born naked, wet and hungry...then things get worse