PI74676: CPSM EYUXC0023S MAXIMUM DATA CACHE LIMIT FOR TOP CACHE POSSIBLE AFTER APPLYING FIX FOR PI68891 OR PI69978 17/01/25 PTF PECHANGE

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• You apply the fix for APARs PI68891 or PI69978 and then may
  begin experiencing CMAS abends and outages with the following
  message:
  .
  EYUXC0023S Maximum data cache limit has been reached
             for TOP cache.
  EYUXC0024S The CMAS is terminating due to a previous
             error with the Cache component.
  .
  An SVC dump is produced, showing the Topology dataspace filled
  with cache lists.
  .
  .
  Additional Symptom(s) Search Keyword(s): KIXREVSVR
  

Local fix

• Until the fixing PTF (or USERMOD if one is made available) is
  applied, you should monitor the size of the TOP dataspaces
  where the PE PTF is applied. With the PE PTF applied, TOP
  dataspaces may grow until they fill up, and the CMAS will
  abend, and possibly not be able to restart without error. You
  can monitor the current size of all dataspaces for a given CMAS
  with the
  .
    COLU CHE CACHE
  .
  command, either entered on a 3270 session in a CMAS or as a
  modify through the console:
  .
  F <cmasjob>,COLU CHE CACHE
  .
  The output, which appears as a new system-generated SYSOUT DD
  attached to the CMAS job, would look similar to this, though
  the asterisks at the end of the TOP1 and TOP2 lines were added
  to hilight them:
  .
  .
  CMAS Name:  <cmasjob>     Date/Time:  yyyy/mm/dd hh:mm:ss.ttt
  
    Name      Token     ALET    Start     End    Used HWM
  
  DMDSxxxx   FF000050 01010058 00001000 45200000 45121F60
  DAT1xxxx   FF000060 0101005B 00001000 30199000 2FD1C000
  DAT2xxxx   FF000160 0101005C 00001000 00100000 000EC000
  QUE1xxxx   FF000070 0102005A 00001000 11999000 11479000
  QUE2xxxx   FF000170 01020059 00001000 00100000 000F9000
  COM1xxxx   AA000080 0101005F 00001000 00999000 000D0000
  COM2xxxx   AA000180 01010060 00001000 00100000 00001000
  TOP1xxxx   FF000090 01020063 00001000 7F999000 7F997000 ***
  TOP2xxxx   FF000190 01020062 00001000 7F900000 7F8FE000 ***
  RTA1xxxx   AA0000A0 01010066 00001000 00999000 008A8000
  RTA2xxxx   AA0001A0 01010067 00001000 00100000 00001000
  MON1xxxx   AA0000B0 01010064 00001000 00999000 00089000
  MON2xxxx   AA0001B0 01010065 00001000 00100000 00001000
  WLM1xxxx   FF0000C0 0101006A 00001000 00999000 0080A000
  WLM2xxxx   FF0001C0 0101006B 00001000 00100000 00001000
  MAS1xxxx   FF0000D0 0101005D 00001000 03199000 02D5D000
  MAS2xxxx   FF0001D0 0101005E 00001000 00100000 00001000
  BAS1xxxx   AA000100 01010068 00001000 24999000 244B6000
  BAS2xxxx   AA000200 01010069 00001000 00100000 000F9000
  .
  
  where 'xxxx' in each name above is the 4-character CMAS SYSIDNT
  of the owning CMAS. You would pay close attention to TOP1 and
  TOP2. In particular, as TOP2 gets larger and larger, you should
  schedule a restart of the CMAS before it abends. Upon restart
  this gets cleaned up and the slow storage creep may begin
  again.
  

Problem summary

• ****************************************************************
  * USERS AFFECTED: All CICSPlex SM V5R2M0 and V5R3M0 Users      *
  ****************************************************************
  * PROBLEM DESCRIPTION: After applying PTF UI42500 (CPSM 5.2)   *
  *                      or PTF UI42501 (CPSM 5.3), you may      *
  *                      experience CMAS abend with the          *
  *                      following messages:                     *
  *                                                              *
  *                      EYUXC0023S Maximum data cache limit has *
  *                                 been reached for TOP cache.  *
  *                                                              *
  *                      EYUXC0024S The CMAS is terminating due  *
  *                                 to a previous error with the *
  *                                 Cache component.             *
  *                                                              *
  *                      Please note that it is possible for the *
  *                      CMAS to encounter a different data      *
  *                      space related problem before the TOP    *
  *                      data spaces fill up.  In that case,     *
  *                      instead of the EYUXC0023S message       *
  *                      indicated above, one of the following   *
  *                      messages can be issued before the       *
  *                      EYUXC0024S:                             *
  *                                                              *
  *                      EYUXC0016S Maximum space allocation has *
  *                                 been reached for Data Cache  *
  *                                 Management data cache.       *
  *                                                              *
  *                      EYUXC0021S <Total | CPSM> auxiliary     *
  *                                 storage limit has been       *
  *                                 reached.                     *
  *                                                              *
  *                      EYUXC0022S Maximum aggregate data cache *
  *                                 limit has been reached.      *
  ****************************************************************
  * RECOMMENDATION: After applying the PTF that resolves this    *
  *                 APAR, all CMASes and MASes must be           *
  *                 restarted. Note that the restarts do not     *
  *                 need to occur at the same time.              *
  ****************************************************************
  In PTF UI42500 and UI42501, to support scope resolution for
  resource tables, WLMAWAOR, WLMWATOR and WLMAROUT, method
  EYU0WABA, EYU0WABT and EYU0WABU were modified to call method
  TMRL (EYU0TMRL - CICS TS Topology Region Lookup) to get a cache
  list of CICS system name(s). However, those cache lists are not
  deleted properly after using. As a result, the storage for the
  cache lists is not freed.
  
  Eventually, the storage in TOP data spaces is filled up, and
  subsequent allocation of the cache list fails. The CMAS region
  abends with message EYUXC0023S and EYUXC0024S issued.
  

Problem conclusion

• To free up the storage for the cache lists, the following
  changes have been made:
  
  - Method EYU0WABU, EYU0WABA and EYU0WABT have been updated to
    delete the cache lists properly after using.
  
  - Method EYU0WABB has been updated to free the storage allocated
    via XSCG call.
  
  - Abend recovery routine has been added to method EYU0WABA,
    EYU0WABT, EYU0WABU and EYU0WABB to recovery registers and
    free up allocated resources properly.
  
  In addition, to improve the serviceability of cache lists, the
  following changes have been made:
  
  - Copybook EYURXCLT (Data Cache Search List Table) and EYURXCAI
    (Data Cache Alternate Index Table) have been updated to
    include time stamp, system ID, information of the caller etc.
  
  - Method EYU0XCIN, EYU0XCLB and EYU0XCLX have been updated to
    set the related information properly.
  

Temporary fix

Comments

APAR Information

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

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

  UI44175 UI44176

Modules/Macros

• EYU0UCIN EYU0UCLB EYU0UCLX EYU0WABA EYU0WABB EYU0WABT EYU0WABU
  EYU0XCIN EYU0XCLB EYU0XCLX EYU9WMP3 EYU9WMP4 EYU9WMPU EYU9WMR3
  EYU9WMR4 EYU9WMRU EYUA2484 EYUA2485 EYUC2484 EYUC2485 EYUE2484
  EYUE2485 EYUEA1WZ EYUEA1X0 EYUEA1X1 EYUKA1WZ EYUKA1X0 EYUKA1X1
  EYUL2484 EYUL2485 EYUP2484 EYUP2485 EYUSA1WZ EYUSA1X0 EYUSA1X1
  EYUT2484 EYUT2485 EYUY2484 EYUY2485
  

Fix information

• Fixed component name

  CICS TS Z/OS V5

• Fixed component ID

  5655Y0400

Applicable component levels

• R00M PSY UI44176

     UP17/01/26 P F701 {

• R90M PSY UI44175

     UP17/01/26 P F701 {

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.