EYUWI0012E E    date time applid WLM warmstart AOR normalization could not complete for AORSCOPE (csysgrp) in all workloads for CONTEXT (plexname).

Explanation

As part of WLM warmstart, a CMAS performs processing during initialization to verify that the AORSCOPEs it maintains for the workloads it manages match the CPSM Topology for the CICSGRPs named by the AORSCOPEs. If they do not match, then updates are performed to change the WLM view of the AORSCOPEs to match the Topology view of the CICSGRPs.

An error has occurred during this processing, and as a result the CMAS has not been able to normalize the named AORSCOPE for workloads in the named CICSplex.

System action

A exception trace record is written, and if this is the first occurrence of the EYUWI0011E or EYUWI0012E message during this execution of this CMAS, then a dump is taken. The title of the dump will be similar to the following:

                    EYU0XZSD Dump,jobname,cmasname,lpar,CMAS,LPLT,tasknum,XLEV,
                    EYU0WMWS,date,time
                    

The CMAS then continues AOR normalization for the next AORSCOPE associated with the CICSplex.

User response

Collect the dump, the auxtrace data sets, and the complete job output from the CMAS and contact your IBM support center.

Access the Explorer "Workload Targets" view (Window -> Show view -> Other - specify "Workload Targets" in the "type filter text" box and select OK) or the WUI "Active workload target distribution factors" view (EYUSTARTWLMAWAOR.TABULAR2) from sessions connected to all CMASes that manage the named CICSplex. For each workload displayed by the CMAS that received this error, verify that the target members displayed for the workload match the other CMASes that display the workload.

If any discrepancies are found, then the error resulted in AOR normalization not being performed for the CMAS. To resolve the discrepancies, the CMAS can be restarted, and AOR normalization will be attempted again. If the error does not re-occur on the restart, then perform the verification process described above. If the error re-occurs, or the verification continues to indicate discrepancies, then the CMAS, all of its connected MASes, and any active batch API programs that are connected to the CMAS must be terminated at the same time, to allow the WLM dataspaces to be deleted. Note that if a MAS cannot be terminated, then termination of the MAS agent by using the COSH transaction is sufficient.

Once the CMAS, its MASes, and all batch API programs have been terminated, issue the MVS display job command for the ESSS address space on the LPAR to verify that the WLM dataspaces have been deleted. The format of the command is "D J,EYUXvrm", where "vrm" is the CPSM version, release and modification (for example, for CPSM V5R2M0, the command would be "D J,EYUX520"). There are two WLM dataspaces allocated for the CMAS, named WLM1syid and WLM2syid, where "syid" is the CICS SYSIDNT of the CMAS. If the display job output indicates that the WLM dataspaces are still allocated for the CMAS, then at least one connected MAS or batch API program is still active. To identify what MASes or batch API programs are still active, execute the EYU9XENF utility on the LPAR. For information about this utility, please refer to topic "Using the ESSS Information Display Utility (EYU9XENF)" in the CICS Information Center.

Once the MVS display job command is issued and no WLM dataspaces are displayed for the CMAS, the CMAS, MASes and batch API programs can be restarted. Note that for MASes whose MAS agent code was terminated by the COSH transaction, you must use the COLM transaction to restart the MAS agent instead of restarting the region. When the routing regions connect to the CMAS, the workload for the routing regions will be rebuilt properly.

Module

EYU0WMWS

XMEOUT parameters/Message inserts

  1. date
  2. time
  3. applid
  4. csysgrp
  5. plexname

Destination

Console and COLG



EYUWI0012IE.html | Timestamp icon Last updated: Sunday, 4 June 2017