Special considerations

When working with Management Central monitors, you need to consider these special points.

Special considerations when working with job monitors

  • The Job Count metric monitors the number of active jobs that match the job selection criteria during a collection interval.
  • The Job Monitor window (Management Central > Monitors > Job > Righ-click a job monitor > Open) shows jobs that meet the criteria even if the jobs are no longer active at the end of the interval. Collection services provides information that determines the job count as well as the jobs to display in the window. This data contains information about all of the jobs that are active during that interval. Nevertheless, it is possible that if a job uses negligible CPU, then information about that job is not passed to the job monitor and so it does not appear in the count or the detail status display.
  • For the metrics Job Status and Job Log Message if a job monitor triggers it continues to display those jobs that created the condition even if a job has ended and is not active during the interval. For this condition the job displays with a gray icon, and continues to be displayed until the trigger resets or the monitor restarts.

Special considerations when working with file monitors

  • The Text metric monitors for a specific text string. When you use this metric, the File Monitor obtains a shared read lock on the files that it is monitoring. Programs which obtain a shared update lock can update files without interfering with the monitor. However, users, programs and commands (such as the Work with Objects using Programming Development Manager (WRKOBJPDM) command or the Start Source Entry Utility (STRSEU) command) that obtain an exclusive lock will interfere with the file monitor and might cause it to either fail or to not be able to monitor the criteria during each interval.
  • A file monitor uses an integrated file system to access the information that it needs about the files that it is monitoring. Integrated file systems treat QSYS physical files as directories. Only the physical file members are actually treated as files. If you want to monitor the size of the entire contents of the QSYS physical file you must monitor all of the members that it contains (typically a single file member).

    For example, to monitor the size of the database file QAYIVDTA in the QMGTC library enter /qsys.lib/qmgtc.lib/qayivdta.file/qayivdta.mbr in the Files To Monitor field (Management Central > Monitors > File > Right-click a monitor > Properties > General tab). You can view the size of the database file from within the System i® Navigator File System.

  • The Text metric is the only valid metric when monitoring the QHST file.

Special considerations when working with system monitors

The V5R3 PTF SI18471 introduced the ability for the central system to try to restart a system monitor regardless of the reason. (Before this PTF, the central system would only restart a system monitor if the failure was due to a connection failure with the endpoint and if the monitor was still in a started status. This meant that only monitors with multiple endpoints that suffered connection failures were restarted.) To use this feature the following conditions must be met:
  • The central system must be running release V5R4 or later. (This capability is also available on V5R3 central systems provided the PTF SI18471 is installed.)
  • The keyword &RESTART is in the name of system monitor.
  • The Management Central property Automatically restart monitors on failed systems is checked. (Right-click Management Central > Properties > Connection tab)