Technical Knowledge Base

Analyze log files

Analyze log files

The following logs are useful in finding issues that may not have been reported as alerts on the cluster. Note, all of these logs will have the newest entry at the bottom.

  1. sel_parsed.txt
    • Name of log bundle.tar.gz\latest_ClusterName\X#-SC#\ipmitool
      • This is the system event log recorded by the BMC. There is a log for each storage controller in the cluster and the XMS. They are found in their respective sub folders. X#-SC# for the storage controller, and XMS for the XMS.
  2. alerts.out
    • Name of log bundle.tar.gz\latest_ClusterName\xms\xms-logs
  3. messages
    • Name of log bundle.tar.gz\latest_ClusterName\X#-SC#\system\logs
  4. crit-messages
    • Name of log bundle.tar.gz\latest_ClusterName\X#-SC#\system\logs\
  5. This contains the output files for all commands run by the gather.
    • Name of log bundle.tar.gz\latest_ClusterName\xms\xmcli
  • showX-Bricks.out
  • Showstoragecontrollersinfo.out
    • Shows part numbers, OS Version, health status and serial numbers of controllers
  • showSSDs.out
    • Shows part numbers and state of drives
  • Showlocaldisks.out
    • This shows your hard drives for your controllers. Including state and part number.
  • ShowBatteryBackupUnits.out
    • BBU Part numbers and status
  • ShowAllInfinibandswitches.out
    • Part numbers for the infiniband ( not power supplies)
  • ShowAllInfinibandSwitchesPSUs.out
    • Infiniband power supply information ( no part number)
    • 071-000-588 ( found on sticker on PSU in Infiniband Switch)
  • .outs show part number information for
  • ShowAllModules.out
    • Gives you detail information on SAS Loop,
  • ShowAllDAEs.out
    • Gives you part number and details of DAEs
  • ShowAllDAEControllers.out
    • Health of DAEs
    • Part Numbers
    • Location
    • LCC Card Part Numbers
  • ShowClustersRaidProperties.out
    • Helps understand how raid is laid out on the DAE drives. Min and Max number.
    • Shows how many drives have to stay up in order for the raid to continue to work