Re: High Concurrency Wait

From: Stefan Knecht <knecht.stefan_at_gmail.com>
Date: Wed, 29 Nov 2017 12:25:03 +0700
Message-ID: <CAP50yQ_nUOZQygSwq0iBSgHt5pE9vUWph72ObRKu4C88YajuEQ_at_mail.gmail.com>



If you're seeing that event in such high amounts, this is most likely a bug.

If you are running parallel queries and those queries are using bind variables, the most likely candidate is "Bug 21834574 - Mutex contention and increased hard parse time due to ILM (segment-access tracking) checking with partitioned objects (Doc ID 21834574.8)". We actually discovered this initially and Oracle filed the initial bug on behalf of us. What you're gonna want to do is take consecutive hanganalyze and systemstate dumps. Oracle will need that to fully prove if you're hitting this bug.

If you have a test environment where you can reproduce it, you can apply patch Patch 21834574: MUTEX CONTENTION FOR PARALLEL QUERY WITH BIND VARIABLES and see if the problem goes away. Do not however, that you will always see this event to some degree. It's quite normal to see it appear in an AWR report, but it should be nowhere near the top.

Stefan

-- 
//
zztat - The Next-Gen Oracle Performance Monitoring and Reaction Framework!
Visit us at zztat.net | Support our Indiegogo campaign at igg.me/at/zztat |
_at_zztat_oracle

--
http://www.freelists.org/webpage/oracle-l
Received on Wed Nov 29 2017 - 06:25:03 CET

Original text of this message