PK71336: WRONG PROCESS AND OUTPUT QUEUE TIMES IN MANAGEMENT EXCEPTION REPORT AND RECORDS DROPPED FROM TRANSACTION TRANSIT LOG REPORT

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• There are wrong (large negative) Process and Output queue times
  displayed in the Management Exception report and there are
  records dropped from the Transaction Transit Log report for
  WFI transactions using IMS V10 logs.
  
  The internal mapping of IMS V10 type 31 log record does not
  properly account for QLGUCCBO, causing QLGUSQ6T to be set
  incorrectly.
  
  Additional issue fixed:
  Loop in IPI4PRx1 -
  In storage HWM processing the sequential record chain can be
  corrupted when:
    1. A processing transaction is to be added and there is no
    space available in the processing queue,
    2. The new record, and all other records in the processing
    queue are chained from a single originating transaction.
    3. Flushing occurs, which leaves a pointer to the, now
    invalid originating transaction.
  End result is tight loop in IPI4PR91 or IPI4PRA1 near
  offsets x'F7C', x'F82', x'F86' and x'F8A'.
  KEYWORDS: IMSV10TGT/K
  
  Another symptom of this addtional issue is an abend0c4 in
  the same routine (IPI4PR91) near x'BC2' failing with R7
  containing a bad value of x'7FFFF000'.
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED: All IMS Performance Analyzer V4.1 users.     *
  ****************************************************************
  * PROBLEM DESCRIPTION: Wrong (large negative) Process and      *
  *                      Output queue times in Management        *
  *                      Exception report and records dropped    *
  *                      Transaction Transit Log report for WFI  *
  *                      transactions using IMS V10 logs.        *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  1. The internal mapping of IMS V10 type 31 log record does
  not  properly account for QLGUCCBO, leaving data causing
  QLGUSQ6T to be set incorrectly
  2. In storage HWM processing the sequential record chain can
  be corrupted when:
  a. A processing transaction is to be added and there is
  no space available in the processing queue,
  b. The new record, and all other records in the processing
  queue are chained from a single originating transaction.
  c. Flushing occurs, which leaves a pointer to the, now
  invalid, originating transaction.
  

Problem conclusion

• 1. Correct mapping of IMS V10 type 31 log record into
  internal format.
  2. Storage HWM for processing transactions occurs after
  the record is added, not before
  

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:

  UK39769

Modules/Macros

• IPI#PRSM IPI@PRIM IPIFPROC IPISLCA1 IPISLC91
  IPISMERG IPISVSMG IPI4PRA1 IPI4PR91
  

Fix information

• Fixed component name

  IMS PERF ANALYZ

• Fixed component ID

  5655E1500

Applicable component levels

• R410 PSY UK39769

     UP08/09/24 P F809

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.