IC70909: ONTAPE RESTORE HANGS WITH LOOP IN DOPHYSRECVR WHEN UNEXPECTED LOG NUMBER IS FOUND AFTER THE RESERVED PAGES PART OF THE ARCHIVE

APAR status

• Closed as program error.

Error description

• If a different log is placed in the archive than is expected the
  ontape cold restore will hang in an endless loop.
  
  The symptoms are that after the instance is initially started
  for the physical recovery the physrecover thread will stay
  running and will consume CPU time. If you investigate the stack
  of the CPU VP the physrecover thread is running on (by operating
  system means like procstack, pstack or with a debugger), you
  will see it loop in dophysrecvr() and restore_log_page().
  
  The archive tape with the described inconsistency needs to be
  considered bad in this case, but essentially on the attempt to
  perform a restore from it you should receive a proper error
  message instead of being faced with an endless loop.
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED:                                              *
  * Users of ontape                                              *
  ****************************************************************
  * PROBLEM DESCRIPTION:                                         *
  * If a backup of a logical log is corrupted and contains data  *
  * for a log file other than the one described by the tape      *
  * header, logical log restore will not properly detect the     *
  * corruption and will hang in an infinite loop, if the         *
  * unexpected data is located immediately after the tape        *
  * header.                                                      *
  ****************************************************************
  * RECOMMENDATION:                                              *
  * Upgrade to version 11.50.xC8.  Note, however, that the       *
  * condition indicates corruption during the logical log backup *
  * (for example, chunks were overwritten in the system on which *
  * the backup was taken).                                       *
  ****************************************************************
  

Problem conclusion

• Improved checking for this condition is included in 11.50.xC8
  

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:

Fix information

• Fixed component name

  IBM IDS ENTRP E

• Fixed component ID

  5724L2304

Applicable component levels

• RB15 PSY

     UP