II13278: OEM BUGS IN VSAM, CAT. & IDCAMS CAUSED BY OEM PROGRAMS OR OTHER IBM PRODUCTS. SEE II10001/11013/11513/12140/12615/14105

APAR status

• INTRAN

Error description

• This APAR continues ii10001, ii11013, ii11513, ii12140 and
  ii12615, in documenting symptoms/bugs related to VSAM, Catalog,
  IDCAMS & RLS caused by various OEM vendor products or other IBM
  products.
  KEYWORDS:  VSAMBUGS INFOAPAR and INFOAPAR5.
  --------------------------------------------------------------
  

Local fix

• 239)  CSMI hung in FCIOWAIT and could not be purged.  CICS
        region had to be cycled.
        IAM zap-id p-70.0134 for v7.0/08p and above dated
        Feb 7, 2002.
        IAM describes problem as follows:  "If a program returns
        through the UPAD exit prior to the ECB being posted, then
        IAM will issue a wait with R1 pointing to the ECB.  The
        code then clears the ECB, however, if R1 is changed by
        the SVC routine, then the ECB is not properly cleared,
        which may cause a subsequent entry through the UPAD to
        falsely think it is posted, causing possible corruption
        of a lock queue."
  240)  MSGIEC036I 002-018 on REPRO from tape to tape.
        The abend002 occurred because the RDW specified
        an LRECL that was larger than specified in the DCB
        for the dataset.  Looks like the DCBPRECL field
        was picked up instead of the DCBLRECL.
            Vendor : Innovation Data Processing
            Product: UPSTREAM ZAP
            fix    : P-31.0067
  241)  ABEND138
        Second enqueue for SYSVTOC from same task.
        INNOVATION (FDR) fix P-53.5162
  #242) OEM Vendor Enet a DB2 & IMS product, RRDF 2.2
        VSI Corruption w/ RPLFDBWD = x'2E080020' Problem
        reported in DFSMS/MVS HDZ11F0 / R1F0 VSI overlay
        was on x'08', contained a address not found in
        dump. fix from Enet is ptf JN00442 which modified
        loadmods E1CMDST and L2URM. Problem gives 0C1
  243 - DB2 / NPI data set having problems extending to a new
        volume, can't seem to allocate an extent to a new volume.
    DSNP007I  *DB3E DSNPXTN0 - EXTEND FAILED FOR
               F1WMPO3G.DSNDBD.DSA2P000.XA2A0003.I0001.A005.
               RC=00D70027
               CONNECTION-ID=DB3E, CORRELATION-ID=010.PFTXTN01,
               LUW-ID=*
        In DFSMS r1F0 and r1G0 there is an EOV control block
        expansion that is different from prior releases and BMC
        products are impacted by this.
  OW53107 corrects Media Manager code for rebuilding EDBs.
  BMC fixes are BPG1802 and BPG1804 for MAINVIEW and STOPx37.
  Customer should apply both fixes.
  #244) Abend0c1 in IDA019R1 + 12B6 Plh chain pointer
        overlaid. Vendor HURON has ticket AA09792
  #245) ABEND0C4 is in csect IDA0200T at uw82653, release R1F0.
        Offset at or near x'10E6' depending on PTF level.
        Sterling Commerce/Connect:Direct ver 4.1.09 using ATTACH
        macro with SZERO=NO. Fix number T028393.
        Sterling Commerce/Connect:Direct ver 4.3.0 fix number
        T028465. 6/19/02.
  #246) 878-04 BMC UltraOpt BSOP VWA CPOOL is BMC
        UltraOpt a VTAM session optimiztion tool
        excessive getmain's of 1K blocks BMC fixed
        in release 41.1(?)SP241 KEY6 ABOVE the line
        (key6=VTAM) SUBPOOL 241 KEY 6 DQE's = 1000
        PMR26172,499
  #247) S0C4 Abend in module IDA019L1. CSECT IDA019C1 at
        x'58C'  UW70910. R4 REG4 is zeros x'00000000'  Error
        code is CEE0374C Application is the PRF report product
        from CA (formerly from Platinum). A DB2 query product.
        CA has fix URFC119,  requires two prerequisite fixes,
        URFC091 and URFC119, Abend only seems to affect Z/OS
        users but could conceivably affect anyone especially
        after maintenance.
  248)  ABEND0c4 pic10 in CA sort module CA71B004is fixed by
        CA fix# LO91373 (don't know if that is alpha oh, or zero).
        ABEND0c1 in load modules (?) CA71B004, CA71B016 and
        CA71B019 for FMID CSO7100 (don't know if alpha oh, or
        zero) fixed by QO17740.
  249) Catalog address space hung.
       SVC26 Get/update request issued
       Catalog gets SYSVVDS EXCLUSIVE and returns.
       At this point catalog waits for PUT update/end.
       An asynchronous interrupt occurs and an SVC37(EOV) occurs.
       CA module issues another SVC26 for the same TCB. Request
       cannot complete until the previous request completes but
       that REQUEST cannot complete.  The CA module eyecatcher is
       .........01.V37L
       STC -LO85387 SVD
       5300 05.33 20010
       The CA fixes are LO99563 and QO194480.
  250) ABEND0C4 RC11 in TERASAM or TERRASAM module TS$OPX32
       trying to open a dummy file.  Softworks
       provided fix in ZAP NUMBER: TSM32334.
  251) ABEND0C4 IDA019C1 during SHOWACB, but ACB address
       is in an area freed and unavailable. PROSMS module
       PRS41410 cleared the high order byte of the UCB
       after intercepting the End-Of-Volume call to IDA0557B.
       PROSMS fix BPG1928. 9/9/02.
  #252) MSGIEC070I 104-204 unable to extend error when space
        is available. Using OEM BMC Corp. SRM/Resolve product.
        BMC fix BPG2232. 09/23/02.
        This fix was also given to another customer by BMC to
        resolve a problem with a hung ( hang wait ) task in
        IDA019SE at or near x'3A4'. 10/7/02.
        Hang occurs because PLH is defer / deferred ( PLHDRSSC )
        DIWA + x'49' and the DIWA is owned by same PLH.
        Actual BMC fix to hang is BPG1802.
  #252) ENQ HANG on SYSDSN minor name of backup catalog when
        HSM ARCMDSUV runs. Problem in BMC Inc. product MAINVIEW
        SRM release 6.1.01, fix number BPG2225.
  253) UCB overlay problem:
       BMC causes an overlay in below-the-line private storage of
       a captured UCB. Since a "captured UCB" is really just an
       extra virtual address created for a 24-bit address space,
       and this virtual address is a secondary reference to the
       same real page that contains the "real" UCB with its 31-bit
       virtual address known to the other address spaces, the real
       page is the same for both the "captured UCB" and the "real"
       UCB. Thus overlays to the "captured UCB" are also overlays
       to the real UCB, and the same is also true for any other
       contents on that same real page.
       This overlay always occurs at +F14 into the page, and the
       content of the overlay is either '40000000' or '40800000'.
       Since UCBs and their extensions are of fixed size, the
       overlay damages either the DCEVIB pointer or the second
       word in the UCB common extension.
      **** You can recognize this problem because the failing
       address resides at 'xxxxxF14' into storage and contains
       either of the above two values, '40000000' or '40800000'.
      **** Current known external symptoms:
       abend0C4 in ICVTST00 or CVAF code with DCEVIB pointer
        overlaid.
       abendB00 in IECVTCCW when called by CVAF (ICVDSD03) and
        second word of UCBCX has been overlaid.
       abend800 in IECVTCCW when called by IGG019TV and second
        word of UCBCX has been overlaid.
      **** Contact BMC for fix number as it will vary depending
       on the FMID installed. One FMID fix number is BAY5838,
       another is BPY5733. This is for base MAINVIEW from BMC.
  #254) MSGIGD306I UNEXPECTED ERROR DURING IEFAB4C2 RC24
        while using DTS Software ACC/SRM. OEM ZAP provided
        by DTS:
  
        *
  *  VSAM ADDVOL INPUT IS NOT GETTING ALL VOLUMES IN SYNC
  *
   NAME ACC0000 ACC2630
   IDRDATA DTS31045
   VER 00EA 4710,C118
   REP 00EA 4700,C118
   CHECKSUM 09030912
  NOTE***
  Verify this ZAP with the vendor BEFORE using it.
  10/10/02.
  #255) A CA Job call PLATMAN is using up storage in
        ECSA. Can see this via verbx vsmdata 'owncomm'
        command. job runs about 5/6 times, but last
        time job is still active and using way more
        memory. Fix is C2275KA.(see pmr: 21059,370)
        keywords: owner gone Computer Associates R1F0
  #256) MSGIEC161I 052(015)-084. Using Computer Associates
        MIM serialization product. Fix number TD45041 needed
        from CA. 11/20/02.  TD45072 needed also. 9/26/03.
  #257) ABEND378-14 following freemain in IDA0557A.
        OEM STOPx37 or MAINVIEW redrives the VSAM End-
        of-Volume interface in error on a Verify Refresh.
        BMC Mainview PTF: BPG2201. 12/01/2002.
  #258) Systems hung with catalog enqueued on SYSIGGV2
  and TMSQNAME.  TMS suspended a task that held the
  TMC, during a dismount, which caused a deadly embrace
  when MIMGR passed the ENQ to other sharing systems.
  KEYWORDS:
  NAME=MAJOR=TMSQNAME MINOR=TMS-TMC SCOPE=SYSTEM
  SYSIGGV2 WAIT HANG CA1 CA-VTAPE
  The following ptf was identified by CA:
  QO29006 - Ensure correct Recovery areas and prevent
  double dismount actions.      12/17/2002
  #259) Customer has broken data set.  It was discovered during
        REPRO when they received:
             msgIDC3302i ACTION ERROR ON dsname
             msgIDC3314i **RECORD OUT OF SEQUENCE - KEY FOLLOWS:
        Full story was never provided.  Customer said "This
        problem has tentatively been identified as a problem with
        EMC micro code in combination with a subsystem feature
        that Innovation Compactor exploits...."
        EMC case number is 8300150
        Innovation case number is L5478.
  #260) Two data sets, CICS.TS130.CIPFANT.DFHDMPA and
        CICS.TS130.CIPFANT.DFHDMPB are cataloged with
        invalid device type. (X'000000FF') or
        (x'0000000F').
        SVM4802I NO VOL IN STORAGE GROUP WITH ENOUGH SPACE
        occurring often.
      BMC Recommendation:
       Recommend they remove the dump datasets from all
       STOPX37 type recovery because these files are system
       controlled through CICS control blocks using specialized
       EXCP.  Using SPACVOLA in such a situation is not
       recommended in our doc.
       Please add this statement to the SPACVOLA FLST:
        SET  MODE = INACT,  MSG = E
        INC   DSN = CICS.TS130.CIPFANT.DFHDMP/
  #261) When terminating LE environment receive
            msgIGZ0009C A DELETE MODULE IGZCLNK UNSUCCESSFUL
        Appears that ABEND4094 rc0C also received.
        Program uses HIPERBATCH and after doing a VSAM GET program
        receives control back in Supervisor State Key 0 (key0 ).
        Problem only occurs when HIPERBATCH ( HIPERSPACE ) used.
        System Trace shows:
            SVC   6B 078D0000 ...
            SVCR  6B 070C0000 ...
        SVC 6B is MODESET macro/command and indicates request to
        change to Supervisor State / key0.  It is being executed
        from a VSAM Compression product from EMC ( Softworks )
        with eyecatchers VCPIDA01914 and VCPCOM14.
        EMC resolution is VCP case# 8397726 ZAP VCP14015.
  #262  Abend0c4 in IDAVRIRB(RLS) caused by a bad reg2 used
        by STOPX37(BMC SOFTWARE) hook in IDAV557A.
        BMC FIX:
        Fix BPG2469 was issued by BMC to resolve this
        problem.
        ..
  #263) MSGIEC070I 195-221 trying to extend a VSAM ds
        under STOPx37 direction. OEM product problem
        description:
   BPG2215 - Vsam EOV hook not always calling IDA0557G under
  Z/os 1.3.
        OEM PTF BPG2215 01/21/03.
  #264) MSGIEC070I 203-204, MSGIEC070I 032-202
        MSGIEC070I 022(22080018)-207 followed by a hang
        of the asid ( CICS, IMS, etc ) with TCB in a
        wait issued in VSAM csect IDA019SE at or near +x'3AC'
        with PLH deferred on x'80' resource, but no valid
        resource pointed to by PLH. Using OEM space mgmt
        product SRS from DTS at release 2.1.11 or 2.2.11.
        Solution: Install SRS release 3.1.8. 01/24/03.
  #265) ABEND102 because OEM code issues an SVC2 with an ECB
        address of all zeros. OEM vendor is Softworks or
        Candle. SVC 2 is issued out of HLPREQ30 and there are
        no active VSAM requests at the time.
        Other possible vendor names or searches: abend102 rsn00
        after IDCAMS REPRO Performance Essential product
        Solution: Problem should only show up in 64 bit mode.
        OEM fix is PSP30242.
  #266) MSGIEC070I 104-204 followed by CICS region hang
        with SVC 1 WAIT ( SVC1 ) in IDA019SE at or near
        offset x'994' depending on PTF level of IDA019SE.
        Using SRS ( Space Recovery System ) from DTS
        Software. SRS below release 3.1.8 is NOT compatible
        according to the vendor with DFSMS R1F0 and above.
        Solution: Upgrade SRS to release 3.1.8. 2/13/03.
  #267) Abend0c4 in IDA019R0 at or near offsets x'9C' -
        x'B4' depending upon maintenance level. Reg3 is
        bad, most likely loaded with zeroes, but could be
        any bad address. A prior ABEND0C1 in IOS code caused
        this abend and was itself due to OEM ASTEX product
        corrupting the IOSBABN. ASTEX from Computer Associates
        fix number JC28045 and JC28052. 2/19/03.
  268) IEC070I 032-202 out of IDA0557B have also seen
       IEC070I 022(22080018)-202 Caused by BMC doing a
       double call to IDA0557B. Fix is BPG2392 12/3/02
       (EOV) (NOTE: other EOV was a UMOD BPG2236 9/9/02)
       Will not see this in a dump cause VSAM control
       blocks are over written, either need BMC GTF trace
       or SLIP in IDA0557B to verify this BPG2392 fix.
       this was R1F0 (this was ptf UW92338, but could
       be other IBM releases).
  269) MSGIEC070I IEC070I 203 - 203 RC203 when using BMC
        FASTREORG FAST REORG utility to load a VSAM file
        with no secondary space quantity defined, but with
        CANDIDATE volsers available. FASTREORG uses a license
        to special VSAM I/O driver Media Manager. A function
        of MM is to ignore candidate volsers when no sec
        space is set. THis is WAD from MM, but can confuse a
        user accustomed to regular VSAM I/O and EOV behavior.
  270) Fast Catalog Search utility, IGGCSIVS, catalog
        scan running long. This can occur if the
        DATAAREA is to small.
        In SYS1.SAMPLIB, change IGGCSIVS constant DATAAREA
        from 32000 to 64000.
  271) Physical I/O errors after a dataset has extended with BMC's
       Mainview and was copied with DSS.  Problem is due to BMC
       reducing the secondary amount to a number of tracks that is
       not a multiple of the CA size.  Fix number is BPG2009.
  msgIDC3302I  ACTION ERROR ON REPMV.INFOFC1.TOPIC.D010418.T132446
  msgIDC3350I 03057,13180245,00022DE80000,D,PL6422,A102512A,STEP01
           467,DA,INDD1   ,SFX001,INVALID SUFFIX ,000020EA000E01,V
  272) BMC STOPX37 issuing EOV twice. Corrupted vsam database
       after job issues userabend 01C
       error type = 2 , reg 15 code = 0008 , reason ocde = 001C
       Following error messages seen:
       IEC161I 080-053
       IEC070I 178(8,17235)-040
       STOPX 37 fix is PTF BPG2392 when BPG2200 is installed.
       Additional keywords: ABEND01C MSGIEC161I RC80 RSN53
       MSGIEC070I RC178 RSN40
  273) Performance (high CPU/Loop) symptoms.  Caused by
  Performance Solutions intercepting SVC 121's.
  

Problem summary

Problem conclusion

Temporary fix

Comments

APAR Information

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

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

Fix information

Applicable component levels