A fix is available
APAR status
Closed as program error.
Error description
You receive message DFHIS0002 A severe error (code X'0812') has occurred in module DFHISAL. This was followed by message DFHIS0001 An abend (code ---/AKEC) has occurred at offset X'247C' in module DFHISAL. After the abends, the IPIC connections to the CTG's stop taking work. You are required to recycle the TOR to clean up the IPIC connections. The dump shows CSMI is the failing task with its stack as follows: KE_NUM @STACK LEN TYPE ADDRESS LINK REG OFFSET ERR NAME 02C0 345A3020 0120 Bot B1B01C00 B1B01FBC 0003BC DFHKETA 02C0 345A3140 0320 Dom B1B1A5F8 B1B1A810 000218 DFHDSKE 02C0 345A3460 0820 Dom B1B42B00 B1B44160 001660 DFHXMTA 02C0 345A3C80 0760 Dom B1BCB6A0 B1BCC68E 000FEE DFHRMUO 02C0 345A43E0 06A0 Sub B1BA9008 B1BA941C 000414 DFHRMLSO 02C0 345A4A80 0250 Dom B29E28A0 B29E34FA 000C5A DFHISJU Int +00078E B29E2ACE 00022E PEFORM_COMMIT Int +000C00 B29E3066 0007C6 TERM 02C0 345A4CD0 02C0 Sub B27286B0 B272908C 0009DC DFHISZA Int +000874 B2728902 000252 DFHISZA_FREE_ISSB 02C0 345A4F90 0980 Sub B27031D8 B2705ECE 002CF6 DFHISAL 02C0 345A5910 0E50 Dom B1B8CF28 B1B90B8C 003C64 DFHMEME Int +003222 B1B8D18E 000266 SEND Int +00146E B1B90220 0032F8 CONTINUE_SEND Int +003B98 B1B8E4A0 001578 TAKE_A_DUMP_FOR_CALLER 02C0 345A6760 0620 Dom B1C2EB88 B1C3008C 001504 DFHDUDU Int +000B26 B1C2ED88 000200 SYSTEM_DUMP Int +001934 B1C2FADC 000F54 TAKE_SYSTEM_DUMP The DFHIS0002 message is issued when we try to free the ISSB: IS 0800 ISAL ENTRY - FUNCTION(FREE_RECEIVE) ISSB_PTR(349C7420) SUSPEND(YES) The trace will have an exception trace entry like: IS 0812 ISAL *EXC* - NOT_ACTIVE_ISSB - FUNCTION(FREE_RECEIVE) ISSB_PTR(349C7420) SUSPEND(YES) The problem is caused by a timing error when one task enters abend processing and then frees the IPIC session. In between, another task uses the same IPIC session and then attempts to end and free the session. This fails as the session has been already been freed. This same underlying problem may also be seen as: DFHIS0001 An abend (code ---/AKEC) has occurred at offset X'10DC' in module DFHISEM. DFHIS0002 A Severe Error (code X'0686') has occurred in module DFHISXM. Additional Symptom(s) Search Keyword(s): KIXREVRJL
Local fix
Problem summary
**************************************************************** * USERS AFFECTED: All CICS users. * **************************************************************** * PROBLEM DESCRIPTION: MSGDFHIS0002 severe error code 0812 in * * module DFHISAL when freeing an ISSB. * **************************************************************** * RECOMMENDATION: * **************************************************************** Message DFHIS0002 severe error code 0812 when freeing an ISSB allocated to a mirror task. After this error occurs, the ISSB chain for inactive ISSBs is corrupted with the top ISSB pointing forward to itself. Prior to this error, a separate mirror task encountered a transaction abend. This mirror task sent an IS7 error to its client then released its ISSB to the free chain. This mirror task then lost control. In this window a new ECI request arrived. The ISSB, which had been used by the abending mirror, got allocated for this new ECI request and a new mirror task was attached. The new mirror task lost control after completing a DFHISIS INITIALIZE_RECEIVER call. The abending mirror regained control and called DFHTFP which called DFHISIS SEND_ERROR to log the transaction abend. The abend was logged but SEND_ERROR processing then inspected the ISSB which the abending mirror task was still referencing. If the ISSB had been on the free chain SEND_ERROR would have completed processing at this point. However, the ISSB was now allocated to a new task and critically the ISSB_STATE had been changed to ISSB_SEND. DFHISIS SEND_ERROR concluded that the mirror task had not yet responded to its client so an IS7 error was now sent. However, as the ISSB was now owned by a different mirror task, this IS7 was inappropriate. After this send completed, the abending mirror freed the ISSB despite the fact that this mirror didn't own the ISSB. Later on the new mirror task encountered problems because it was running with an ISSB which had been freed. This mirror task eventually suffered the DFHIS0002 when it attempted to free the ISSB which was already on the free chain. ADDITIONAL KEYWORDS :- CTG CICS Transaction gateway IPCONN IPIC
Problem conclusion
DFHISIS SEND_ERROR has been changed to ensure the mirror task issuing SEND_ERROR still owns the ISSB it references before it attempts to send a response to the client and release the ISSB to the free chain.
Temporary fix
FIX AVAILABLE BY PTF ONLY
Comments
APAR Information
APAR number
PM22385
Reported component name
CICSTS V3 Z/OS
Reported component ID
5655M1500
Reported release
500
Status
CLOSED PER
PE
NoPE
HIPER
NoHIPER
Special Attention
YesSpecatt / Pervasive
Submitted date
2010-09-13
Closed date
2010-10-21
Last modified date
2010-11-02
APAR is sysrouted FROM one or more of the following:
APAR is sysrouted TO one or more of the following:
PM23942 UK61558
Modules/Macros
DESISIS DFHISIS
Fix information
Fixed component name
CICSTS V3 Z/OS
Fixed component ID
5655M1500
Applicable component levels
R500 PSY UK61558
UP10/10/29 P F010
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.
[{"Business Unit":{"code":"BU058","label":"IBM Infrastructure w\/TPS"},"Product":{"code":"SSGMGV","label":"CICS Transaction Server"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"3.2","Edition":"","Line of Business":{"code":"LOB35","label":"Mainframe SW"}},{"Business Unit":{"code":"BU054","label":"Systems w\/TPS"},"Product":{"code":"SG19M","label":"APARs - z\/OS environment"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"3.2","Edition":"","Line of Business":{"code":"","label":""}}]
Document Information
Modified date:
02 November 2010