I got confused with log option setting in cluster.conf. I specified log_level 3 in tag rm, verify with ccs_test, but does not look like it works as expected. here are commands I run and the results: <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"> <meta name="Generator" content="Kate, the KDE Advanced Text Editor"> <pre>[root@node02 log]# ccs_test connect Connect successful. Connection descriptor = 29700 [root@node02 log]# ccs_test get 29700 //rm/@log_level Get successful. Value = <3> [root@node02 log]# clulog -s 5 -p progName "severity 5 should not be shown" [root@node02 log]# tail -1 /var/log/messages Jul 24 14:52:08 node02 clulog[22819]: <notice> severity 5 should not be shown [root@node02 log]# clulog -s 5 -l 5 -p progName "severity 5 should show" [root@node02 log]# tail -1 /var/log/messages Jul 24 14:54:04 node02 clulog[23000]: <notice> severity 5 should show [root@node02 log]# clulog -s 5 -l 4 -p progName "severity 5 should not show, filter is 4" [root@node02 log]# tail -1 /var/log/messages Jul 24 14:54:04 node02 clulog[23000]: <notice> severity 5 should show </pre> My question is why severity 5 is shown in log file even if I did not specify filter in my command, clulog should uses the service manager's assigned log level that is 3. The cluster has 2 nodes, and all nodes are rebooted and restarted before run the commands. Any explanation ? Thanks,