A fix is available
APAR status
Closed as program error.
Error description
Customer IPLed the system and started to see abends in z/OS UNIX modules: ABEND378 RSN14 ABENDEC6 RSN0B0F0555, RSN0B5B0557 and RSN0B5E0557 ABEND9C6 RSN00340001 Customer had to Re-IPL to clear the situation ANALYSIS: Reviewing the dumps and logrec records we found: i) the S378 abend occured in BPXOINIT in a waitpid() SYSCALL, invoked by the zombie task, BPXPRWAT was trying to release the KIDC storage during cleanup processing before returning to the caller. However, the storage had already been freed. ii) The AbendEC6 RSN0B0F0555 occured when BPXPRIN2 invoked BPXPRTB1 to get a processid and BPXPRTB1 failed because PPRP addresses were invalid. The PPRA anchor was no longer valid. iii) The ABEND9C6 abends occured in different address spaces (all primary OMVS). They were caused by BPXPRIN4 trying to release the PPRA latch and the latch token from PPRALATCHTOKEN was zero. Further analysis showed that the PPRA had been freed and a PPRT cell pool for address space x'0037' was later obtained in its place. The systrace showed that a SP(2) freemain had been issued by BPXNSKIL KNOWN IMPACT: Unpredictable abends will occur and an IPL is necessary to recover. VERIFICATION STEPS: If any of the listed abends occured and storage release for subpool 2 in the OMVS address space fails because storage had already been released, This APAR may apply. The PPRA resides in subpool 2. Take the address of the PPRA from the OCVEPPRA and verify the "PPRA" eye catcher or content. If the eye catcher or some of the PPRA table is not valid this apar may apply. If the systrace shows a subpool 2 freemain in the OMVS address space from BPXNSKIL, this apar applies. If the sp freemain trace entry is not found, this apar may still apply. Following is what the storage release would look like: BPXNSKIL+1E20 R7C0 PC ... 0 00_1AA88ED0 00311 Storage Release SSRV 133 00000000 00000203 00000000 00000000
Local fix
RECOVERY ACTION: IPL
Problem summary
**************************************************************** * USERS AFFECTED: * * Users running z/OS UNIX System Services on * * z/OS V2R3 (HBB77D0) or above. * **************************************************************** * PROBLEM DESCRIPTION: * * An overlay caused BPXNSKIL to attempt to send a signal to an * * empty process group. In this situation we erroneously obtain * * a zero-length buffer to snapshot member of the process * * group. At the end of processing when we go to free the * * buffer, the zero length results in freeing all of subpool 2. * **************************************************************** * RECOMMENDATION: * * Install this PTF and dynamically activate the service to * * improve z/OS UNIX reliability. * ****************************************************************
Problem conclusion
Checks need to be improved to avoid the inadvertent freeing of subpool 2.
Temporary fix
Comments
APAR Information
APAR number
OA63098
Reported component name
OPENMVS SYS SRV
Reported component ID
5695SCPX1
Reported release
7C0
Status
CLOSED PER
PE
NoPE
HIPER
NoHIPER
Special Attention
NoSpecatt / Xsystem
Submitted date
2022-04-06
Closed date
2022-07-07
Last modified date
2022-08-02
APAR is sysrouted FROM one or more of the following:
APAR is sysrouted TO one or more of the following:
UJ08826
Modules/Macros
BPXNSKIL
Fix information
Fixed component name
OPENMVS SYS SRV
Fixed component ID
5695SCPX1
Applicable component levels
R7D0 PSY UJ08826
UP22/07/20 P F207
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":"BU054","label":"Systems w\/TPS"},"Product":{"code":"SG19M"},"Platform":[{"code":"PF054","label":"z Systems"}],"Version":"7C0"}]
Document Information
Modified date:
02 August 2022