An undetermined stall condition

Ask these questions when you are dealing with a stall condition.

Questions to ask
  • Did the stall occur during initialization?
    • How far did initialization progress?
    • Were there any definition or setup errors reported?
  • Did the stall occur during operation?
    • Are the necessary communication links between CMASs and MASs available?
    • What type of request was being processed?
    • How big was the CICSplex involved?
    • How many CMASs and MASs were involved?
    • What types of monitoring, real-time analysis, and workload management were active?
  • Did the stall occur during termination?
  • Did the stall occur in a CMAS?
    • Did the request time out with an EYUEInnnn message?

      The local CMAS may be waiting for one or more CICS® systems (or their CMAS) to return requested data. A CICSPlex SM view does not return until all the expected data is collected.

    • Did the request time out with a CICS message?
  • Did the stall occur in a MAS?
    Try stopping the MAS agent code (using the STOP action command from the MAS view), then evaluate the underlying CICS system.
    • Is the CICS system taking an SDUMP?
    • Is the CICS system looping or hung?
    • Did the request time out with a CICS message?
    • Is the CICS system experiencing a short on storage (SOS) condition, or has it reached its MAXTASK level?
    Any one of these conditions could prevent some types of CICSPlex SM requests from completing.
Documentation to collect
  • System console log and EYULOG
  • CMAS job logs
  • Unformatted SDUMP from the affected address spaces (TSO, CMAS or MAS)


eyus1f3.html | Timestamp icon Last updated: Monday, 5 June 2017