Hey, I am running a two node rac on 11gR2 (220.127.116.11) on SLES 11.
One instance is having a load of 0.05 and the other a load of more than 1.00
ps -ef is not showing any procesess, that is consuming a noticeable value of cpu time or memory usuage.
Even when I shutdown this instance, the load keeps at around 1.00
So, even when the sessions switching to the first instance, the load stays at 0.05
When I issue: crsctl stop crs on the second instance then the load is going down !!!!
At the point of starting the clusterware stack (crsctl start crs) the load is increasing back to 1.00
Any idea, what the reason could be ? How to identify a particular procress that is consuming the ressources ?
By load do you mean load average? If yes then can you check are the processes in the top output are consuming high resources. This is where you will need to start your investigation from
However, please raise a SR with support for the same, as the OS statistics and clusterware logs need to be reviewed for such issues.
Vandana - Oracle