Event duration overview

Edit online

Before tuning event duration time, read these prerequisites and notes.

The following are important to keep in mind when you are working with event duration:

  • The total duration time is calculated differently for "slow" and "fast" cluster events.

    "Fast" events are those that do not include acquiring or releasing resources and normally take a shorter time to complete.

    For "fast" events, the total duration time during which PowerHA® SystemMirror® waits before issuing a warning is equal to Event Duration Time.

    "Slow" cluster events are those that involve acquiring and releasing resources or use application controller start and stop scripts. "Slow" events may take a longer time to complete. Customizing event duration time for "slow" events lets you avoid getting unnecessary system warnings during normal cluster operation.

    For "slow" events, the total duration time before receiving a config_too_long warning message is set to the sum of Event-only Duration Time and Resource Group Processing Time.

  • Remember, you can customize event duration time before receiving a warning for cluster events, not for nodes or specific resource groups in your cluster. Once the Total Event Duration Time is specified, the system waits for the specified period of time and sends a config_too_long message to the node which was affected by this event.

    For example, you have a cluster with five resource groups. A node_down event (a "slow" event) occurs on Node A, which owns some of the resource groups. And, you have previously specified the Event-only Duration Time to be 120 seconds, and the Resource Group Processing Time to be 400 seconds.

    When a node_down event occurs on Node A, a config_too_long message is sent to Node A according to this formula:

    Event Duration Time (120 seconds) + Resource Group Processing 
Time (400 seconds) = 520 seconds (Total Event Duration Time).
    A config_too_long message appears on Node A after 520 seconds.
  • During dynamic reconfiguration events, the Cluster Manager uses the previously specified values of the event duration time until warning. After dynamic reconfiguration is complete and the new values of event duration time get synchronized, the Cluster Manager uses the newly specified values.
To configure event duration, complete the following steps:
  1. From the command line, enter smit sysmirror.
  2. From the SMIT interface, select Custom Cluster Configuration > Events > Cluster Events > Change/Show Time Until Warning, and press Enter.
  3. Modify any of the fields press Enter.