A fix is available
APAR status
Closed as program error.
Error description
Parts of the raw above-the-bar internal trace table are overlaid. The overlaying data could be Kernel stack storage with eyecatchers like F4SA. It could be CICS control blocks or application storage. . Possibly, most of the raw above-the-bar trace table looks good except for the 1st x'20' bytes of each 4K page. The first x'30' bytes of each 4K page should look something like this: --------------------- ASID(X'006A') ADDRESS(50_00000000.) STORAGE -------- _0000000: 00000050 00001000 00000050 027FF000 _0000010: 00000000 00000000 D42FC2FE 58000000 _0000020: 4C6E018A 00140500 C042177C 007B0015 -------------------------- The first double word should be a pointer to the next 4K page. The next double word should be a pointer to the prior 4K page. The first trace entry begins at +x'20' into the page with the x'4C6E' bytes. . Below is an example where the trace entry looks good starting x'20' into the page, but the 1st x'20' bytes are overlayed. ------------------------------------- ASID(X'006A') ADDRESS(50_02784000.) STORAGE -------- _2784000: F9F9F9F9 F9F9F0C1 F0F0F0F0 F2F9F9F7 _2784010: F4F460F2 F0F0F0F0 D42FC2FE 58000000 _2784020: 4C6E01E2 00101940 C041915C 0067C5E5 ----------------------------- . Sometimes the Trace Domain Anchor block ( TRA ) has an invalid TRA_NAB (which is the double word at +x20 into the TRA.) When invalid, it does not fall within the bounds of the trace table. The double word at TRA +x50 is the beginning of the trace table. The double word at TRA +x'58 is the end of the trace table. . If TRA_NAB is bad like that, then each attempt to write a trace entry can abend0C4 in DFHTRPX and DFHTRPT because TRA_NAB is the address where the next trace entry will be placed. This can result in a kernel stack looking like this: . 0340 Dom 3334EDC8 333504A7 0016DF *Y* DFHTRPT 1290 Dom 333AA3B8 333AB383 000FCB DFHMEME Int +004BA6 00000000 C55C48 TRACE_MEME_PLIST Int +000F8A 333AEF88 004BD0 TRACE_DATA 0340 Dom 3334EDC8 333504A7 0016DF *Y* DFHTRPT 1290 Dom 333AA3B8 333AB383 000FCB DFHMEME Int +004BA6 00000000 C55C48 TRACE_MEME_PLIST Int +000F8A 333AEF88 004BD0 TRACE_DATA 0340 Dom 3334EDC8 333504A7 0016DF *Y* DFHTRPT 1290 Dom 333AA3B8 333AB383 000FCB DFHMEME Int +004BA6 00000000 C55C48 TRACE_MEME_PLIST Int +000F8A 333AEF88 004BD0 TRACE_DATA 0340 Dom 3334EDC8 333504A7 0016DF *Y* DFHTRPT . And ultimately, CICS can abend with a KERNDUMP issued with the following message: . DFHKE0303 DBDCCICS A RECURSIVE ABEND HAS BEEN DETECTED BY THE KERNEL DOMAIN. . This problem is caused by a DFHXMERM domain call in module DFHISXM that does not properly CLEAR the domain call parameter list. This problem will only occur during tasks that are initiated by input from an IPCONN connection. Also, the path of the bad domain call only happens when the task has been purged prior to getting started (for example purged while queued for TRANCLASS reasons.) Additional Symptom(s) Search Keyword(s): KIXREVDWZ DFHXM0001 An abend (code 0C4/AKEA) has occurred at offset X'01AA' in module DFHXMER. AKC3 UAKC3 abendAKC3
Local fix
turn off XM component tracing with STNTRXM=OFF and SPCTRXM=OFF Additional Keywords: overlay MVCL DATA1
Problem summary
**************************************************************** * USERS AFFECTED: All CICS Users. * **************************************************************** * PROBLEM DESCRIPTION: Repeated abend 0C4 in DFHTRPX and * * DFHTRPT. * **************************************************************** Module DFHISXM fails to clear the existence bits on a domain call for inquire_deferred_abend. This can result in the trace routines picking up residual data leading to a loop of abendS0C4s. Keywords: S0C4 MVCL overlay msgDFHKE0303
Problem conclusion
DFHISXM has been changed in refstep isxm_ensure_abend_set to add a clear to the DFHXMERM calls.
Temporary fix
Comments
APAR Information
APAR number
PI97165
Reported component name
CICS TS Z/OS V5
Reported component ID
5655Y0400
Reported release
800
Status
CLOSED PER
PE
NoPE
HIPER
NoHIPER
Special Attention
NoSpecatt / Xsystem
Submitted date
2018-04-24
Closed date
2018-09-20
Last modified date
2019-02-20
APAR is sysrouted FROM one or more of the following:
APAR is sysrouted TO one or more of the following:
UI58607 UI58608 UI58609 UI58610
Modules/Macros
DFHISXM
Fix information
Fixed component name
CICS TS Z/OS V5
Fixed component ID
5655Y0400
Applicable component levels
R000 PSY UI58607
UP18/09/21 P F809
R100 PSY UI58610
UP18/09/21 P F809
R800 PSY UI58608
UP18/09/24 P F809
R900 PSY UI58609
UP18/09/24 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.
[{"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":"5.1","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":"5.1","Edition":"","Line of Business":{"code":"","label":""}}]
Document Information
Modified date:
20 February 2019