IBM Support

PI90147: z/OS WLM health percentage does not affect CPSM routing weight.

A fix is available

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as unreproducible in next release.

Error description

  • WLMHEALTH follow-up, allow health percentage (1-100%) to affect
    routing weight.
    

Local fix

Problem summary

  • ****************************************************************
    * USERS AFFECTED: All users of CPSM V5R4M0.                    *
    ****************************************************************
    * PROBLEM DESCRIPTION: z/OS WLM health percentage does not     *
    *                      affect CPSM routing weight.             *
    ****************************************************************
    * RECOMMENDATION: After applying the PTF that resolves         *
    *                 PI90147, all CMASes and MASes running as     *
    *                 CPSM WLM routing regions must be recycled    *
    *                 to pick up the new code.  Note that the      *
    *                 restarts do not need to occur at the same    *
    *                 time.                                        *
    *                                                              *
    *                 After applying this PTF, you may notice      *
    *                 differences in where work is routed within   *
    *                 an active workload when CICS TS 5.4          *
    *                 regions are restarted.  The z/OS WLM         *
    *                 health percentage may influence the amount   *
    *                 of work a CPSM WLM routing region will       *
    *                 route to a target region.                    *
    *                                                              *
    *                 A value of zero (0) means that no work       *
    *                 will be routed to the region. Any workload   *
    *                 affinities associated with this region       *
    *                 will remain and be honored, but routes       *
    *                 based upon those affinities will fail        *
    *                 while the value is zero. Note that the       *
    *                 route failure seen depends upon how the      *
    *                 route is attempted and whether a custom      *
    *                 WRAM program is in affect. Some users may    *
    *                 receive message EYUWR0003W indicating that   *
    *                 the affinity AOR is not available, while     *
    *                 others may receive a SYSIDERR.               *
    *                                                              *
    *                 A value of 1 to 99 inclusive means that an   *
    *                 additional weight will be added to the       *
    *                 region that is inversely proportional to     *
    *                 the WLMHLTH value. That is, the lower the    *
    *                 WLMHLTH value, the greater the additional    *
    *                 weight assigned. This may have the affect    *
    *                 of increasing or decreasing workload being   *
    *                 received by the region as its WLMHLTH        *
    *                 value increases or decreases, depending      *
    *                 upon the weight assigned to other regions    *
    *                 within the AORSCOPE.                         *
    *                                                              *
    *                 A value of 100 means either that a region    *
    *                 has now achieved a fully open z/OS WLM       *
    *                 health open state, or that the region is     *
    *                 running with WLMHEALTH=OFF in the SIT, or    *
    *                 that the region is running CICS TS V5.3 or   *
    *                 lower. A region with a value of 100 is       *
    *                 treated as healthy and no additional         *
    *                 weight is added for the region.              *
    ****************************************************************
    The z/OS WLM health percentage of a CICS region does not
    affect the likelihood of a CICS region being routed to by
    the CPSM WLM routing calculation.  This means that a region
    with a z/OS WLM health percentage of 1% has the same
    probability of being chosen as a region with a z/OS WLM health
    percentage of 99%.  Consequently, a region that is only just
    starting up may receive more workload than initially expected.
    

Problem conclusion

Temporary fix

Comments

  • 1. The following WUI views have been modified and should
    be reloaded for the changes to be picked up:
    EYUxA03I
    EYUxA1X0
    EYUxA21V
    where "x" is one of E, K or S for English, Kanji (Japanese)
    or Simplified Chinese.
    This can be done without restarting the WUI server by using
    option 5, "Import" of the COVC transaction and specifying the
    appropriate data sets containing the aforementioned members.
    
    2. The CPSM routing algorithm has been changed to take
    account of the z/OS WLM health percentage when deciding the
    target region of a routing decision.
    
    3. The following changes have been made to the CICS TS 5.4
    Knowledge Center:
    
    --------------------------------------------------------------
    A new section is added to the CICS TS 5.4 Knowledge Center as
    per the following:
    Administering
    -> Administering CICSPlex SM
     -> Administering workloads with CICSPlex SM
      -> Managing workloads
       -> Workload routing
        -> Effect of the z/OS WLM health service on CPSM workload
           routing
    
    The following text has been added:
    
    --- START OF TEXT ---
    
    In a CICS® region, if the z/OS® Workload Manager (z/OS WLM)
    health service is active, which is the default, CPSM WLM takes
    account of the region's z/OS WLM health state in the routing
    decision.
    
    For CPSM WLM to take into account the z/OS WLM health state in
    the target regions, both the routing and target regions must
    be at a minimum of CICS TS V5.4 (CICS and CPSM libraries),
    otherwise the z/OS WLM health state will be ignored and the
    target regions will be deemed to be ready to receive work.
    
    If the service is disabled, to activate this service for a
    CICS region, you must set the WLMHEALTH system initialization
    parameter.
    
    The z/OS WLM health state of a region is reflected by a z/OS
    WLM health indicator in the range 0 - 100. A value of 1-100
    means that the region is ready to receive work whereas a z/OS
    WLM health indicator of 0 means that it is not ready to
    receive work. When the WLMHEALTH SIT parameter is set to OFF,
    the region is deemed as always ready to receive work.
    
    If the target region has a z/OS WLM health indicator of zero,
    it is assigned a very high weight in the routing algorithm.
    Therefore, the region is less likely to receive work. Note
    that the region remains active, so it might receive work if no
    other healthier regions are available in the same scope. This
    is different from quiescing a target region, which removes the
    region as a candidate target from all workloads, regardless of
    the availability of other target regions. For more
    information, see Quiescing a target region in an active
    workload.
    
    During the CICS warm-up or cool-down process, CICS adjusts the
    z/OS WLM health indicator of the region. As a result, this
    affects the routing decision of CPSM WLM, which restricts
    workload into the region during that process. For more
    information about what happens during CICS warm-up or
    cool-down, see CICS warm-up and cool-down by use of z/OS
    Workload Manager health service.
    
    You can deliberately make a region less favorable in CPSM
    routing decisions by setting its z/OS WLM open status to CLOSE
    or IMMCLOSE in the MVS workload management (MVSWLM) view.
    
    --- END OF TEXT ---
    
    --------------------------------------------------------------
    
    The section:
    Administering
     -> Administering CICSPlex SM
      -> Administering workloads with CICSPlex SM
       -> Example CICSPlex SM workload management tasks
        -> Quiescing a target region in an active workload
    
    Has the text of the 2nd paragraph changed from:
    
    --- START OF TEXT ---
    
    If you want to make a region undesirable in the routing
    decision but keep it active, see Effect of the z/OS WLM health
    service on CICSPlex SM workload routing.
    
    --- END OF TEXT ---
    
    to the new text of:
    
    --- START OF TEXT ---
    
    If you want to prevent work routing to a region, regardless of
    whether work is already running there, you can set z/OS® WLM
    health open status to CLOSE or IMMCLOSE.  For more
    information, see Effect of the z/OS WLM health service on
    CICSPlex SM workload routing.
    
    --- END OF TEXT ---
    
    --------------------------------------------------------------
    
    The section:
    Administering
     -> Starting and stopping CICS
      -> CICS warm-up and cool-down, facilitated by z/OS Workload
         Manager health service
    
    In section 'How use of z/OS WLM health service affects CICS
    and CICSPlex SM components', the first paragraph of the
    CICSPlex SM subsection has been changed to:
    
    --- START OF TEXT ---
    
    CICSPlex SM workload distribution uses the z/OS WLM health
    value of the region in the routing algorithm. When CICSPlex SM
    Workload Manager decides where to route work, a region with a
    health value of zero is not eligible to receive work, and
    Workload Manager routes work only to a region with a non-zero
    health value. The greater the health value, the more favorable
    the region will be in the routing decision. For details, see
    Effect of the z/OS WLM health service on CICSPlex SM workload
    routing.
    
    --- END OF TEXT ---
    --------------------------------------------------------------
    
    The new section 'Factors that contribute to dynamic workload
    routing decisions' is added as per the section 'Advantages of
    WLM' in the page
    Administering
    -> Administering CICSPlex SM
     -> Administering workloads with CICSPlex SM
      -> Managing workloads
    
    The section is added as per the section 'Advantages of
    WLM' with a summary text of:
      CICSPlex SM workload routing makes a routing decision based
      on a combination of factors.
    
    The new section is added as:
    Administering
    -> Administering CICSPlex SM
     -> Administering workloads with CICSPlex SM
      -> Managing workloads
       -> Factors that contribute to dynamic workload routing
          decisions
    
    The text of the new section is:
    
    --- START OF TEXT ---
    
    CICSPlex® SM workload routing makes a routing decision based
    on a combination of factors.
    
    The following factors are taken into consideration in the
    routing decisions:
    
    The number of tasks in the region
      Factored as a percentile value, calculated by dividing the
      target region MAXTASKS setting with the current task count.
    
    The health status of the region
      Factored by assigning arithmetic weights, depending on
      whether the region is short-on-storage, taking a transaction
      dump, taking a system dump, running at its MAXTASKS limit,
      in a non-responsive state, or in a CICSPlex SM stall.
    
      Note
      Health is assessed independently of system availability
      monitoring (SAM) and is not influenced by SAM settings.
    
    
    The speed of the link between the router and the target
      Factored by assigning arithmetic weights, depending on
      whether the target region is linked to the target by an MRO
      connection, an XCF connection, an LU6.2 connection, an IPIC
      connection, or if the target region is the router itself.
    
      Note
      The link speed is not used for link-neutral algorithms.
    
    The z/OS® WLM health state of the region
      Factored by assigning additional weights, depending on the
      z/OS WLM health value (in the range 0 - 100) of a region
      provided by the z/OS WLM health service. The z/OS WLM health
      state is factored in only when the z/OS WLM health service
      is on.
    
      A region with a health value of zero is not eligible to
      receive work. Any workload affinities that are associated
      with this region will remain and be honored; however,
      workload routing based on those affinities will fail while
      the health value is zero, and will result in either message
      EYUWR0003W or a SYSIDERR condition. Message EYUWR0003W
      indicates that the EYU9WRAM program was in effect and the
      affinity AOR was not available. To resolve such routing
      failures, initiate a warm-up for the region so that the
      health value is non-zero, and then rerun your workload.
    
      A region with a health value of 100 is not assigned
      additional routing weights and is fully capable of receiving
      work.  For regions with a health value in the range 1 - 99,
      the greater the health value, the lower the routing weights
      assigned. Therefore, as the health value of a region
      increases, the region might gradually receive a larger
      workload.
    
    Outstanding CICSPlex SM Realtime Analysis (RTA) Events
    associated with the workload, if any
      Factored by assigning arithmetic weights depending on the
      severity of the events outstanding. These events are
      factored in only when Event name is specified in the WLM
      specification for the workload or any Transaction Group
      definitions associated with it.
    
    Any transaction affinities, if any, that are outstanding to
    override the dynamic routing decision
      Regardless of any other factors, if the routing request has
      an outstanding affinity associated with it, that affinity
      always overrides the route decision.
    
    --- END OF TEXT ---
    

APAR Information

  • APAR number

    PI90147

  • Reported component name

    CICS TS Z/OS V5

  • Reported component ID

    5655Y0400

  • Reported release

    10M

  • Status

    CLOSED UR1

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt / Xsystem

  • Submitted date

    2017-11-13

  • Closed date

    2018-10-10

  • Last modified date

    2019-01-25

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

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

    UI59011

Modules/Macros

  • CJC0NAG0 CJC0WNAL CJCTNAGC CJD0NAG0 CJD0WNAL CJDTNAGD CJE0NAG0
    CJE0WNAL CJETNAGE CJF0NAG0 CJF0WNAL CJFTNAGF CJG0NAG0 CJG0WNAL
    CJGTNAGG EYU0DVWF EYU0NAG0 EYU0NQGR EYU0TIML EYU0TIST EYU0WABA
    EYU0WABB EYU0WABC EYU0WASM EYU0WASV EYU0WINT EYU0WNAL EYUA0126
    EYUA2484 EYUA2659 EYUBWSWT EYUC0126 EYUC2484 EYUC2659 EYUE0126
    EYUE2484 EYUE2659 EYUEA03I EYUEA1X0 EYUEA21V EYUK0126 EYUK2484
    EYUK2659 EYUKA03I EYUKA1X0 EYUKA21V EYUL0126 EYUL2484 EYUL2659
    EYUP0126 EYUP2484 EYUP2659 EYUS0126 EYUS2484 EYUS2659 EYUSA03I
    EYUSA1X0 EYUSA21V EYUT0126 EYUT2484 EYUT2659 EYUTNAGH EYUY0126
    EYUY2484 EYUY2659
    

Fix information

  • Fixed component name

    CICS TS Z/OS V5

  • Fixed component ID

    5655Y0400

Applicable component levels

  • R10M PSY UI59011

       UP18/10/12 P F810

Fix is available

  • Select the PTF appropriate for your component level. You will be required to sign in. Distribution on physical media is not available in all countries.

[{"Business Unit":{"code":"BU058","label":"IBM Infrastructure w\/TPS"},"Product":{"code":"SSGMGV","label":"CICS Transaction Server"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"5.4","Edition":"","Line of Business":{"code":"LOB35","label":"Mainframe SW"}},{"Business Unit":{"code":"BU054","label":"Systems w\/TPS"},"Product":{"code":"SG19M","label":"APARs - z\/OS environment"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"5.4","Edition":"","Line of Business":{"code":"","label":""}}]

Document Information

Modified date:
25 January 2019