Re: x$ksqst annomolies

From: Jonathan Lewis <jonathan_at_jlcomp.demon.co.uk>
Date: Tue, 30 Oct 2001 19:21:42 -0000
Message-ID: <1004469541.27976.0.nnrp-12.9e984b29@news.demon.co.uk>

I think you'll find that a WAIT is only counted after a timeout has occurred; so if a process goes into an enqueue wait, and times out 12 times, acquiring the lock on the 13th iteration, then x$ksqst will show one wait.

Since a timeout occurs after about 3 seconds, it might be quite hard to engineer any change in ksqstwat for SQ enqueues.

--
Jonathan Lewis
http://www.jlcomp.demon.co.uk

Host to The Co-Operative Oracle Users' FAQ
http://www.jlcomp.demon.co.uk/faq/ind_faq.html

Author of:
Practical Oracle 8i: Building Efficient Databases

Screen saver or Life saver: http://www.ud.com
Use spare CPU to assist in cancer research.

Kyle Hailey wrote in message
<3bde34e7.19314052_at_newsfeeds-goliath.1usenet.net>...


>

>

>Anyone out there using the  ksqstwat column in x$ksqst? As I

>understand it this should be the waits on the locks. When I have two

>users cause a row lock, then this value increases for TX lock. Fine.

>The problem is for other locks. I was creating sequence cache

>contention, tracing the users with sql trace (event 10046 level 8) to

>get the wait events and the users are waiting on the SQ lock  but the

>wait field, ksqstwat, in x$ksqst never gets incremented?

>

>Best Wishes

>Kyle Hailey

>http://oraperf.sourceforge.net

>

>

>oraperf.sourceforge.net

>___________________________________________________________________________

___


>Posted Via Binaries.net = SPEED+RETENTION+COMPLETION =

http://www.binaries.net

Received on Tue Oct 30 2001 - 13:21:42 CST