OY62627: SESSIONS HUNG IN PCDINIT AT INTERCHANGE NODE

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• A network node was inactivated, then re-initialzed, then
  an interchange node in the same APPN network had hung
  searches for resources located on the network node.
  IST1278I CD DSEARCH PENDING FROM A01N TO C255N FOR NETA.ECHO
  IST1051I EVENT CODE = 0201
  IST1062I EVENT ID = 000000010001000000FF000502010881862700C4
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED: All                                          *
  ****************************************************************
  * PROBLEM DESCRIPTION: The sessions were hung in PCDINIT due   *
  *                      to a Directory Services (DS) search     *
  *                      concentration error.  The hung searches *
  *                      were concentrated behind a "scout"      *
  *                      search for the Destination Logical Unit *
  *                      (DLU), but were not released when the   *
  *                      scout search returned.                  *
  *                                                              *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  The problem came about because a "flooded" search was
  outstanding along with a "scout" search.  When a
  "flooded" search returned to the interchange node
  (ICN), ISTDRSBC assumed that it was the "scout"
  search.  ISTDRSBC removed the scout from the concentration
  list, so that when the real scout returned the concentrated
  list was not processed.
  
      SSCP-
      SSCP
     session         CP-CP session
   SA=========ICN-------------------------NN(network node)
      Req#1            Scout Req#1
    --------->    ------------------------->
      Req#2
    --------->   (Concentrated)
  
      Reply#1      Scout reply #1 (not found)
    <--------     <------------------------
                      Flooded Req #2
                  ------------------------->
      Req#3           Scout Req #3
    --------->    ------------------------->
      Req#4
    --------->   (Concentrated)
  
     Reply#2        Flooded reply #2 (not found)
    <---------    <------------------------
     Reply #3       Scout reply #3 (not found)
    <---------    <------------------------
  
  
  (Request #4 hung)
  

Problem conclusion

• ISTDRSBC is now modified to ensure that the network broadcast
  reply being processed is actually the "scout" search before
  removing it from the concentration list.  This has the effect
  of disabling the concentration code while processing "flooded"
  replies.
  
  During testing of this fix, some related anomalies were
  discovered and fixed in this APAR.
  
  ISTDRDQI would still return a negative cache entry if no time
  had elapsed between the last broadcast reply for the resource
  and the next database query for the resource.  The check has
  been altered to never return negative cache entries.
  
  ISTDRSCC, ISTDRSVR and ISTDRSCU have been modified to better
  release concentrated searches after the scout search is
  abnormally terminated.  ISTDRSBC and ISTDRSYB will now be called
  out of ISTDRSCU to free any concentrated searches.  This
  processing is only pertinent if the scout search is being
  terminated.
  

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:

  UY92804

Modules/Macros

• ISTDRDQI ISTDRSBC ISTDRSCC ISTDRSCU ISTDRSVR
  

Fix information

• Fixed component name

  V4R1 MVS/ESA VT

• Fixed component ID

  569511701

Applicable component levels

• R101 PSY UY92804

     UP93/04/16 P F304

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.