RAC Connection Load Balancing Skewed / Incorrect (Doc ID 557061.1)

Last updated on AUGUST 19, 2016

Applies to:

Oracle Net Services - Version 10.2.0.1 to 11.2.0.4 [Release 10.2 to 11.2]
Oracle Database - Enterprise Edition - Version 10.2.0.4 to 10.2.0.4 [Release 10.2]
Information in this document applies to any platform.
***Checked for relevance on 16-OCT-2013***
All connections to RAC cluster are not evenly distributed


Symptoms

Connections to RAC cluster are not balanced between nodes. Often restart of the RDBMS or TNS listener allow connections to be balanced for short while. PMON trace show large time gaps between updates

*** 2008-01-14 12:55:38.967
kmmlrl: CLSPRD goodness 4
kmmlrl: update for service goodness
kmmlrl: 61 processes
kmmlrl: node load 33
kmmlrl: instance load 52
kmmlrl: nsgr update returned 0
*** 2008-01-14 12:59:08.949
kmmlrl: update for session drop delta: 5252578 5252544 52 52 555
kmmlrl: node load 268
kmmlrl: instance load 52
kmmlrl: nsgr update returned 0
*** 2008-01-14 13:01:00.001
kmmlrl: update for session drop delta: 5252612 5252578 52 52 555
kmmlrl: node load 156
kmmlrl: instance load 52
kmmlrl: nsgr update returned 0
*** 2008-01-14 13:04:17.970
kmmlrl: update for session drop delta: 5252648 5252612 54 52 555
kmmlrl: node load 35
kmmlrl: instance load 54
kmmlrl: nsgr update returned 0

Level 16 Oracle Net Listener trace also shows lbscore increasing on some node, but not all.

Cause

Sign In with your My Oracle Support account

Don't have a My Oracle Support account? Click to get started

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms