Investigating issues after system changes

When you encounter a problem that occurs during an operation that previously ran successfully, review this list to determine the root cause of the problem.

• The setup has changed.
• A component changed such as the SAP HANA, IBM Storage Protect™, operating system, network, or hardware components.
• Patches or updates to one or more of the components were applied.
• Changes occurred that originated from the system:
  • Check whether the disks are running full with the Linux™ df command.
  • If network performance decreases, check whether there are new hosts, or new applications. Check whether defects in software or hardware have occurred.
  • If IBM Storage Protect server processing decreases, check whether more clients or more operations were added. Information is also available in the IBM Storage Protect server activity log.

• The setup changed.
• One or more of the operating system, network, database, or hardware components changed.
• Patches or updates to one or more of the components were applied.
• Changes occur that originate by the system:
  • Check whether the disks are running full with the UNIX™ or Linux df command.
  • If network performance decreases, check whether additional hosts, additional applications, or defects in software or hardware occurred. Compare the log files. The log file names are provided for reference:
    • brbackup / brrestore log
    • sbtio.log
    • tdpdb2.SID.nodename.log
    • db2diag.log
  1. If IBM Storage Protect server processing decreases, check whether more clients or more operations were added. Information is also available in the IBM Storage Protect server activity log.

If none of these changes caused the problem, view the last modified time stamp of the following configuration files:

• backint.log
• backup.log
• dsm.sys
• dsm.opt

Run the following command to list all files in the /etc directory, that have been modified in the previous five days:

find /etc -type f -ctime 5 -print

If you can identify changes that are made to the system, roll them back one at a time and try to reproduce the problem. This method frequently reveals which change or set of changes caused the problem.