Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> LOCK problems on RAC

LOCK problems on RAC

From: Cristian Cudizio <cristian.cudizio_at_yahoo.it>
Date: 8 May 2007 03:19:23 -0700
Message-ID: <1178619563.892454.21240@l77g2000hsb.googlegroups.com>


Oracle Database 10g Release 10.2.0.2.0 - 64bit Production With the Real Application Clusters option

On Linux Suse EL 9

it is 2 node cluster.
This morning in alert.log there is

GES: Potential blocker (pid=28970) on resource TX-0008002C-000126F4;  enqueue info in file /opt/oracle/admin/GEOP/bdump/ geop1_lmd0_16600.trc and DIAG trace file

and
iv'e seen on gv$lock a lot of locks of TYPE='TX' and LMODE=3 on same table used as semaphore
to serialize some operations (is a table with one record). All locks where held by session
making an update on the one record of semaphore table (according to gv $session).
According to what i know it would be
one session with one V$lock record type 'TX' and lmode=6 and the others with lmode=0 and REQUEST=6.
So, what does it mean that? why a lot of locks with lmode=3? I've tested from sql*plus to do
an update on locked table and my session was waing and in v$lock it was with lmode=3.
Excuse me for be confused but i hope someone can help me.

Bye
 Cristian Cudizio

http://oracledb.wordpress.com
http://cristiancudizio.wordpress.com Received on Tue May 08 2007 - 05:19:23 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US