A fix is available
APAR status
Closed as program error.
Error description
PGM JBXDLT11 was created by DRD CREATE command and became available before IMS ABEND. But it became 'NOTINIT-2F-NODMB' status during /ERE COLDCOMM and the following message was issued. DFS563I PSB JBXDLT11 REQUIRES UNKNOWN DMB JBX0BOFL, PSB STOPPED The DMB JBX0BOFL was also created by DRD CREATE command and became available before IMS ABEND. And the AREA in this DB was able to reopen during ERE. << Scenario >> 1. CREATE PGM NAME(JBXDLT11) LIKE(RSC(JBXDLT04)) **** BMP 2. /CHE SNAPQ. 3. CREATE DB NAME(JBX0BOFL) LIKE(RSC(JBX0CMFL)) 4. ACBLIB OLC 5. /STA DB JBX0BOFL 6. /STA AREA BOFL01 7. Update AREA BOFL01 in JBX0BOFL by JBXDLT11 ..... 8. F IMSBA1,STOP 9. S IMSBA1 10. /ERE COLDCOMM. -> The following messages issued. DFS563I PSB JBXAMPP4 REQUIRES UNKNOWN DMB JBX0BOFL, PSB STOPPED : DFS563I PSB JBXDLT11 REQUIRES UNKNOWN DMB JBX0BOFL, PSB STOPPED : DFS3715I DEDB AREA REOPEN PROCESS STARTED, RSN=00 IMP1 DFS994I *COLDCOMM* EMERGENCY START COMPLETED. IMP1 DFS2500I DATASET BOFL0101 SUCCESSFULLY ALLOCATED IMP1 ... 11. /DIS AREA BOFL01 BOFL01 N/A N/A N/A 198 N/A JBX0BOFL BOFL0101 10 N/A N/A N/A N/A N/A ... 12. /DIS PGM JBXDLT11 PROGRAM TRAN TYPE JBXDLT11 BMP NOTINIT ---------------------------------------------------------------- << LOG Records >> 22 log : CREATE PGM JBXDLT11 .. 40 log : SNAPQ CHKPT <= Using CHKPT by /ERE COLDCOMM 4006 : does not have DDIR-JBX0BOFL 4007 : include PDIR-JBXDLT11 .. 22 log : CREATE DB JBX0BOFL ---------------------------------------------------------------- << From the dump, which was taken after restart completed >> DDIR of JBX0BOFL exists and DDIR has the address of DMCB. DDIR does not have DDIRBAD flag. PDIR of JBXDLT05 and JBXDLT11 have PDIRBAD and PDIRBADR=2F ------------------ As above, PDIRs were loaded from type4007 log record. But DDIR of JXB0BOFL did not exist in type4006 log record. Then, DFS563I messages were issued just after IMS read type40 CHKPT log. Later, IMS processed type22 log record for CREATE DB JBX0BOFL, then DMCB for DEDB was loaded and randomizer was loaded as DFS2842I.But it did not clear NOTINIT status of PDIRs. -------------------- This is a similar problem as PK51847. PK51847 corrected DFSPGS00 to reset NOTINIT status on UPDATE PGM START(SCHD) command. But it is very inconvenient to enter UPDATE PGM START(SCHD) command to each PSBs. When we created PGM and DB prior to ABEND, we did not need to enter UPDATE PGM START(SCHD) command at all. But, why we need to enter this command after /ERE.... I think that IMS should reset NOTINIT status without operator intervention.
Local fix
Problem summary
**************************************************************** * USERS AFFECTED: IMS V10 users of dynamic resource * * definition CREATE PGM commands * * and MODBLKS online change for * * databases. * **************************************************************** * PROBLEM DESCRIPTION: A program created by a CREATE PGM * * command and scheduled successfully * * becomes NOTINIT-2F-NODMB after /ERE * * COLDCOMM. * * * * A database created by MODBLKS online * * change that is scheduled successfully * * becomes NOTINIT-2F-NODMB after /ERE, * * if IMS restarts from a checkpoint * * before the online change. * **************************************************************** * RECOMMENDATION: INSTALL CORRECTIVE SERVICE FOR APAR/PTF * **************************************************************** A program is created with a CREATE PGM command. The /CHE SNAPQ command is issued to take a checkpoint. The database referenced by the PSB is created with a CREATE DB command. An ACBLIB online change is performed to add the DMB to ACBLIB. The program is scheduled successfully. IMS abends and then is emergency restarted with an /ERE COLDCOMM command. The program is marked as NOTINIT-2F-NODMB and message DFS563I is issued indicating that the PSB requires an unknown DMB, even though IMS does know about the database. The problem is that the program definition was captured in the checkpoint log records and restart does all of its BLDL processing to find DMBs and PSBs in ACBLIB after processing the checkpoint log records. When IMS restart processes the X'22' log record for the CREATE DB command, it doesn't know that the program referencing that database can now reset the NOTINIT status. The program references a database that doesn't exist yet. The database is added with local or global MODBLKS online change. The program is scheduled successfully. IMS abends and then does an emergency restart. The program is marked as NOTINIT-2F-NODMB, even though IMS does know about the database. The problem is that the BLDL processing for DMBs and PSBs is done after the checkpoint log records are processed, before the x'22' log records are processed. The program is marked as NOTINIT because the databases don't exist yet. When IMS restart processes the x'22' log record for the added database, it leaves the program status as NOTINIT, even though the databases have been added and the program can now be scheduled. Additional keywords for searchability: MSGDFS563
Problem conclusion
AIDS: RIDS/SYS RIDS/CNTRL SYS CNTRL GEN: POSTREQ PK65833 *** END IMS KEYWORDS *** DFSDBS00 is changed in the restart logic that processes the x'2205' create database log record for the CREATE DB command or a database added by MODBLKS online change. For the first x'2205' log record for a database, all of the programs marked bad with a status of NOTINIT-2F are marked good, in case the programs reference the database being created. Programs that do not reference the database being created are marked NOTINIT again the next time scheduling is attempted. Programs that do reference the database being created are able to be scheduled.
Temporary fix
Comments
REPINNED RP08/05/12 (ATXT) TO ADD POSTREQ PK65833 INFO. **** PE08/05/12 PTF IN ERROR. SEE APAR PK65833 FOR DESCRIPTION ×**** PE08/05/12 FIX IN ERROR. SEE APAR PK65833 FOR DESCRIPTION
APAR Information
APAR number
PK59333
Reported component name
IMS V10
Reported component ID
5635A0100
Reported release
010
Status
CLOSED PER
PE
NoPE
HIPER
NoHIPER
Special Attention
NoSpecatt / Xsystem
Submitted date
2008-01-16
Closed date
2008-04-10
Last modified date
2008-10-23
APAR is sysrouted FROM one or more of the following:
APAR is sysrouted TO one or more of the following:
UK35438
Modules/Macros
DFSDBS00 DFSDFDB
Fix information
Fixed component name
IMS V10
Fixed component ID
5635A0100
Applicable component levels
R010 PSY UK35438
UP08/04/18 P F804
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","label":"APARs - z\/OS environment"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"10.1","Edition":"","Line of Business":{"code":"","label":""}},{"Business Unit":{"code":"BU048","label":"IBM Software"},"Product":{"code":"SSCVRBJ","label":"System Services"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"10.1","Edition":"","Line of Business":{"code":"","label":""}}]
Document Information
Modified date:
23 October 2008