Page 1 of 1

Isilon - Clear celog 8.x

Posted: Fri Sep 25, 2020 8:55 am
by chris
    1. Open an SSH connection to any node in the cluster using the root account.    
    2. If you want to preserve the original files, make a copy of all files listed in Step 4 before proceeding.    
    3. Disable the CELOG service and make sure that you see that this was done:    
    Cluster-1# isi services -a celog disable      
      The service 'celog' has been disabled.    
    Check that the CELOG process no longer runs on any node. Do not issue multiple times the disable command. Please wait for 30-60 seconds and check again that the processes have stopped. The output below indicates no CELOG process is running:    
    Cluster-1# isi_for_array -s "ps -auwx | grep '[c]elog'"      
      Cluster-1 exited with status 1      
      Cluster-2 exited with status 1      
      Cluster-3 exited with status 1    
    4. Delete the CELOG database files:    
    # rm -rf /ifs/.ifsvar/db/celog          
          # rm -rf /ifs/.ifsvar/db/celog_capture          
          # rm -rf /ifs/.ifsvar/db/celog_analysis          
          # rm -rf /ifs/.ifsvar/db/celog_reporting          
          # rm -rf /ifs/.ifsvar/db/celog_alerting          
          # isi_for_array rm -rf /var/db/celog_capture        
        5. Enable the CELOG service and make sure that you see that this was done:    
    Cluster-1# isi services -a celog enable      
      The service 'celog' has been enabled.    
    Check that CELOG processes are running on each node. Using the "wc -l" command will count the processes on each node and can be used to quickly check that all nodes have processes running in large clusters. Do not issue the "enable" command multiple times. Wait and check the process status after 30-60 seconds, to give the service a chance to start. For large clusters, wait up to 5 minutes. If CELOG does not start, contact Dell EMC Technical Support.    
    Cluster-1# isi_for_array -s "ps -auwx | grep '[c]elog' "      
      Cluster-1# isi_for_array -s "ps -auwx | grep '[c]elog' | wc -l "    
    See, the Notes section below for an example of a healthy CELOG service on a 3-node cluster.    
    6. Send a test event to verify that CELOG is working properly:    
    isi event test create "TestCELOG"    
    This will generate a test event that will be listed in the output of the "isi event events list" command:    
    1.267 10/14 07:47 W    1    1027           TestCELOG    
    Please wait 1-2 minutes to verify that the event is listed. If the event is not shown, try again in 2 minutes. If after 10 minutes, the event is not listed, please contact Dell EMC Technical Support.    

See: ... FS_8.x.pdf