VM65776: VARY PROC LOCK HANG PREVENTS MCW002 ABEND

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• System hang because unresponsive processor handler is prevented
  from generating ABENDMCW002 due to a Vary Proc Lock hang.
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED: All users of z/VM.                           *
  ****************************************************************
  * PROBLEM DESCRIPTION:                                         *
  ****************************************************************
  * RECOMMENDATION: APPLY PTF                                    *
  ****************************************************************
  z/VM OPERATOR console shows repeated MSHCPMPG9152E messages for
  the same unresponsive processor and the system stays up but
  becomes unusable because it is hung. If Monitor was running then
  it probably stopped generating sample records around the time
  of the first MSHCPMPG9152E message.
  
  When the unresponsive processor detection front-end (HCPMPGUP)
  finds an unresponsive processor, it stacks a call to the
  unresponsive processor recovery (HCPMPCPR) to determine whether
  the processor can be restarted to recover from the error, or to
  generate ABENDMCW002 and restart the system.  That process
  requires the Vary Processor Lock (HCPRCCVA).  A lock hang on
  that lock will prevent the recovery action from being performed
  and the system will eventually hang.  There are cases where the
  unresponsive processor may prevent a task that holds HCPRCCVA
  from completing.  In that case HCPRCCVA is never released and
  the unresponsive processor recovery is not able to run and
  generate ABENDMCW002 and restart the system.
  
  When the unresponsive processor recovery is blocked because of
  an HCPRCCVA hang, the system hangs and becomes unusable. The
  customer impact of the outage is greater because the system is
  essentially unavailable while hung even though it is running.
  
  The problem of the HCPRCCVA hang preventing the unresponsive
  processor detection function from running is not the cause of
  the outage.  The cause of the outage is the function that
  caused a processor to become unresponsive.  However, the hang
  on the HCPRCCVA lock prevents the unresponsive processor
  recovery from running and also prevents the generation of
  ABENDMCW002.  Without a dump of the system it is not possible
  to determine the cause of the unresponsive processor.
  
  This APAR addresses system availability and FFDC (first-failure
  data capture) aspects of the problem.
  

Problem conclusion

• The APAR fix adds the capability to detect a Vary Proc Lock
  (HCPRCCVA) hang in unresponsive processor recovery prior to
  where it attempts to acquire HCPRCCVA.  If it detects a lock
  hang on HCPRCCVA then an ABENDMPC008 dump is generated.  This
  satisfies the FFDC concern by generating a dump as close to the
  point of failure as is reasonable.  The dump allows the cause
  of the unresponsive processor to be diagnosed.  The fix also
  improves availability by detecting a permanent error closer to
  the point it occurs and forcing an abend dump and re-IPL rather
  than allowing the system to remain up in an unusable state.
  
  Changed parts:
  - HCPRCC ASSEMBLE
  - HCPMPC ASSEMBLE
  - HCPLCK ASSEMBLE
  - HCPSGP ASSEMBLE
  - HCPMTC ASSEMBLE
  - HCPCCF ASSEMBLE
  
  
  SRL changes:
  GC24-6270-01 CP messages and Codes - z/VM Version 7 Release 1
  - Page 107 - add the MPC008 abend information.
  - This is Chapter 2. System Codes - CP Abend Codes
  GC24-6177-12 CP messages and Codes - z/VM Version 6 Release 4
  - Page 86  - add the MPC008 abend information.
  - This is Chapter 2. System Codes - CP Abend Codes -
    Abend Codes A - M
  
  
  MPC008
  Explanation: This module is distributed as object code only;
         therefore, no source program materials are available.
  User response: See z/VM: Diagnosis Guide for information on
         gathering the documentation you need to assist IBM in
         diagnosing the problem; then contact your IBM Support
         Center personnel.
  

Temporary fix

• FOR RELEASE VM/ESA CP/ESA R640 :
  PREREQ: VM65988 VM66105
  CO-REQ: NONE
  IF-REQ: NONE
  FOR RELEASE VM/ESACP/ESAR710 :
  PREREQ: NONE
  CO-REQ: NONE
  IF-REQ: NONE
  

Comments

APAR Information

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

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

  UM35376 UM35377

Modules/Macros

• HCPCCF   HCPLCK   HCPMPC   HCPMTC   HCPRCC   HCPSGP
  

Fix information

• Fixed component name

  VM CP

• Fixed component ID

  568411202

Applicable component levels

• R640 PSY UM35376

     UP19/06/19 P 2001 ¢

• R710 PSY UM35377

     UP19/06/19 P 1902 ¢

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.