PH48587: THREADPOOLMODULE.PERCENTUSED NOT EXPOSED UNDER THE PROMETHEUS METRICS ENDPOINT

APAR status

• Closed as program error.

Error description

• When using Prometheus in WebSphere Application ServerWAS version
  9.0.5.10 on Linux, the threadPoolModule.percentUsed under Thread
  pool metrics cannot be exposed under the Prometheus metrics
  endpoint.
  

Local fix

• threadPool.percentUsed = (number of concurrently active threads
  / maxPoolSize) * 100
  
  Using the metrics.ear, user can use
  was_threadpool_active_threads for the number of concurrently
  active threads. To find the corresponding max pool size for each
  of the resources, you can go to Servers > All servers > [server
  name] > Additional Properties > Thread Pools in the Admin
  Console, and do a calculation based on that value to get the
  corresponding threadPool.percentUsed using Prometheus queries.
  

Problem summary

• ****************************************************************
  * USERS AFFECTED:  All users of IBM WebSphere Application      *
  *                  Server who use the                          *
  *                  provided metrics.ear application for        *
  *                  outputting Prometheus                       *
  *                  formatted metrics.                          *
  ****************************************************************
  * PROBLEM DESCRIPTION: The thread pool module's "PercentUsed"  *
  *                      statistic from the PMI is not exposed   *
  *                      as                                      *
  *                      a Prometheus metric.                    *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  The thread pool module's "PercentUsed" statistic from the
  Performance Monitoring Infrastructure (PMI) is
  not exposed as a Prometheus metric through the metrics.ear. The
  max pool size is also not
  exposed which prevents users from calculating the percent used
  using the Prometheus Query Language (PromQL).
  

Problem conclusion

• The max pool size for the respective thread pool is exposed as a
  Performance Monitoring Infrastructure statistic. This will then
  be exposed as a Prometheus formatted metric when using the
  metrics.ear. This will allow for calculation of the thread pool
  usage.
  
  The fix for this APAR is targeted for inclusion in fix pack
  9.0.5.21. For more information, see 'Recommended Updates for
  WebSphere Application Server':
  https://www.ibm.com/support/pages/node/715553
  

Temporary fix

• You can find the corresponding max pool size for each of the
  thread pools, by
  going to `Servers` > `All servers` > [server name] > `Additional
  Properties`
  > `Thread Pools` in the Admin Console (from there, a calculation
  can be done
  based on that value to get the corresponding thread pool usage.
  

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

• APAR is sysrouted TO one or more of the following:

Fix information

• Fixed component name

  WEBS APP SERV N

• Fixed component ID

  5724H8800

Applicable component levels