II12561: THIS APAR DOCUMENTS COMMON ERRORS IN THE IMS/VS DATABASE AREA. COMPID 569517611 565515800 5655B0100 PREV=II11174 NEXT=IIXXXXX

APAR status

• Closed as documentation error.

Error description

• THIS APAR IS BEING CREATED TO ASSIST LVL1/LVL2 AND CUSTOMERS IN
  DIAGNOSING CUSTOMER PROBLEMS, WITHIN IMS, THAT ARE NOT APARABLE
  PROBLEMS SUCH AS USER ERRORS.  THE GOAL IS THAT ANYONE WORKING
  ON ONE OF THESE COMMON PROBLEMS WILL NOT HAVE TO 'RE-INVENT THE
  WHEEL'.  THERE IS A SIMILAR APAR FOR THE IMS SUB-COMPONENTS,
  (DC, DATA BASE, ETC.).  ALL PROBLEM DESCRIPTIONS SHOULD BE AS
  BRIEF AS POSSIBLE.  ALL PROBLEMS SHOULD BE NUMBERED AND SHOULD
  LIST THE APPLICABLE RELEASE LEVELS, REQUIRED MAINTENANCE, AND
  KNOWN RESOLUTION.
  additional keywords:  infoapar db dbinfoapar infoapardb
  --------------  PROBLEM INDEX  --------------------------------
  1.DB open failure MSGDFS0730I V,18 & MSGIEC070I 211-212 PG. E 1
  2.IMS DLI address space fails with ABEND0F8 Subcode 014 PG. E 1
  3.PRELOAD of appl prog can cause STATUSAJ STATAJ AJ     PG. E 3
  4.DFSPCC30 abendU0026                                   PG. E 4
  5.Abend0C4 doing ACBGEN because CA's PDSMAN is enabled  PG. E 5
  6.Abend0C4 in DFSAOSF0 R2 is BAD                        PG. E 6
  7.AbendS0c4 DFSDCAP0 SDBTARG=zero                       PG. E10
  8.AbendS0C4 S0C4 DFSDVSM0 UPAD R10 and R5 bad           PG. E11
  9.S0C4 PSW and R14 in DFSLRH00. Regs look like PSW's.   Pg. E11
  10.DSP0037I CHANGE.RECON UPGRADE for V7 coexistence     PG. E10
  11.PSB and DBD entries not in ACBLIB after SB37 abend   PG. E11
  12.DFSHALDB DSPXPDDU S138 abends                        PG. E12
  13.GSAM dataset appears empty after XRST and ISRT calls PG. E13
  14.ABEND S0C1 WHEN TAKING EXTENT ON A OSAM DATABASE     PG. E14
  15.DLI Address space S0C4 MODID UNKNOWN                 PG. E17
  16.BLSR parm not supported for IMS VSAM datasets.       PG. E18
  17.S0C3 in module DLPXDCD8                              PG. E23
  18.Abend 2103 executing DFSMAID0                        PG. E23
  19.Cond. Code 103 from DFSMAID0                         PG. E24
  20.DFS0730I I,31 after deleting/redefining HALDB part.  PG. E25
  21.S0C4 in DFSXBAT0 when running DLI batch.             PG. E26
  22.Cont. of II11174 entry #23. IMS & PDSE support       PG. E27
  23.DFS871E RC 8 RSN 98 ON ILDS during RELOAD of a HALDB PG. E29
  24.Cancelling a LIST.RECON job results in abend S438    PG. E30
  25.CA11 invalid page free causing IMS S0D3              PG. E31
  26. First x'9904 data capture record not created      PG. E32
  27. After SB37 abend on ACBLIB, re-run of ACBGEN
      did not rebuild PSB                               PG. E32
  28. ABENDS0C1 DFSDBLR00+x'180 label PSBREAD1          PG. E33
  29. Data Capture problem caused by OEM compatitlity   PG. E33
  30. DFS0730I RCB, RSN10 or U3303 in batch on z/OS 1.5 PG. E33
  31. msgdfs0651w error 11 running partition database
  oem product fixes are pdf8106 for ims 8.1,
      pdf7106 for ims 7.1 & pdf9101 for ims9.1
  32. Abendu0845 DFSDVSM0 using hiperbatch .
      The VSAM for IMS db shows Hiperbatch enabled.
      AMB (pointed from BUFC +04) +56, bit 40
      bit 40 (AMBHB) on.
      This occurs with DFSMS R1G0 and up because VSAM PUT
      is not deferred when AMBHB is on at this DFSMSlevel.
      Hiperbatch is not allowed with IMS.  IMS uses VSAM LSR.
      Data sets accessed with VSAM LSR are not eligible
      for Hiperbatch.
  
  Error key code x'E4084545' E4084545 will be stored in error
  Reg6 for this type of ABENDU0845 in IMS releases V11 and higher.
  Message DFS1058E - REASON = 045 -  WRITE ERROR  - BUFFER NOT
  WRITTEN, CANNOT FREE BQEL will be written to the SYSLOG.
  Added: November 06, 2013 (cej)
  

Local fix

Problem summary

• ======================UPDATE BY KEN FRIESEN=========09/27/2000=
  1. When IMS allocates a database the following messages are
  produced. MSGDFS0740I V,18 and MSGIEC070I 211-212. Further
  investigation found that the SMS proc had been changed. The
  change included a typing error causing the SMS proc to fail when
  it was started. Correcting the error to the SMS proc and
  starting it resolved the problem.
  Additional keywords: DFS0730I V18 RCV18 IEC070I RC211 RSN212
  ======================UPDATE BY KEN FRIESEN=========09/27/2000=
  2. IMS DLI address space fails with ABEND0F8 Subcode 014
  Analysis shows SVC 114 (x'72) being issued in  module IGG019TV
  which is called from IGG019BA10/18/98/HDZ11B0 NONE. R12 points
  to DFSDBLR0-08/31/99 (base level). The S0F8 with subcode x'14
  which says: The issuer had established an enabled, unlocked
  task (EUT) mode functional recovery routine (FRR) by specifying
  EUT=YES on a SETFRR macro. VSAM L2 findings: When a BSAM check
  was done, that code is going to do an SVC to do I/O. This code
  does not set up any recovery via SETFRR.  So this had to have
  been done prior to calling BSAM CHECK. The problem is caused by
  BMC's Mainview for IMS. The fix is BPI7929. BMC's problem
  description states: PTF BPI7741 (on the BBCAND tape of April 14,
  2000) introduced a problem where an abend S0F8 occurs in DLI/SAS
  address space. This abend occurs only when Workload Monitor,
  Workload Analyzer, or IMSPlex Syste Manager is installed.
  Solution: Obtain from BMC PTF BPI7929 and apply it.
  Additional keywords: S0F8 SC14
  =====================Rick Crandall =================10/09/2000=
  3.Use of PRELOAD for a COBOL application must follow rules
    for REUS and RENT, and for restrictions in an LE environment.
    Since SSA checking occurs early in init processing of the
    tran, a common error is to get a STATUSAJ STATAJ AJ due to
    invalid pointers to SSA. See IMS APPL PRG:DB manual.
  =====================UPDATE BY Ken Friesen============01/01/01=
  4.Running pointer checker utility DFSPCC30 abendU0026. Further
  analysis imodule load is failing for DFSPCRP0. R15 is 02000004
  which says 02 Imodule load, 04 module not found. Customer report
  using EOM products QUICKFETCH and PMO. Maintenance was installed
  on QUICKFETCH but requisite maintenance for PMO was not
  installed at the same time. Installing PMO maintenance resolved
  this. The customer didnot identify specific maintenance.
  =====================UPDATE BY Ken Friesen==========03/21/2001=
  5. ACBGEN, DFSUACB0, utility abends with an abend0c4 because
  CA's PDSMAN has some option enabled causing the error. No doc
  was shipped and the customer could not be contacted for more
  information.
  =====================UPDATE BY Ken Friesen==========04/18/2001=
  6. Abend0C4 in module DFSAOSF0 label EOV1500 instruction
  ST    R7,DEBUCBAD, R2 points to DEB EXTENT that does not exist
  (freed/released) OEM product RESOLVE issued the
  following messages just before the abend:
  SVM4000I IMS2DLI, IMS2DLI, ddname, ATTEMPTING SPACVOLA FOR
  dbname, vol
  SVM4400I VOLUME ADDED - OLD VOLUME vol, NEW VOLUME SMS
  Stop-X37/PROSMS cannot manage OSAM/BSAM (JCL allocated) DBs.
  In this case RESOLVE is adding a new volume and an extent.
  IMS fixes UQ42936 or UQ43530 casued this. Contact BMC for fix
  BPG1450 to correct this error.
  See INFOAPAR II11174 items #9 and #10
  for other symptoms. The correct procedures for allocating OSAM
  DBs via JCL (BSAM not VSAM ESDS) can be found as follows;
  IMS/ESA Administration Guide: Database Manager Version 6,
  Chapter 12, Section 12.3.1  Allocating OSAM Data Sets
  Section 12.3.2  Example of Allocating an OSAM Data Set
  Section 12.3.3  Cautions When Allocating OSAM Data Sets
  .
  The customer needs to get a list of OSAM DBs (from the DBAs?)
  and give them to the DASD/MVS Support group and have them put
  in the PROSMS exclusion list so that PROSMS ignores them when
  a new extent/volume is needed.
  Addtl symptom: Abendu0049 abendu049 from SWAREQ macro in
  DFSDBAU0 subroutine DFSGDSNM.
  =====================UPDATE BY Rick Crandall========04/26/2001=
  7.S0C4 in DFSDCAP0 at +'2B4' using an SDBTARG of zero.
    Identified by user as caused by BMC fix 478969 for a
    reorg utility.
  =====================UPDATE BY Rick Crandall========05/03/2001=
  8.S0C4 in DFSDVSM0 UPAD exit UPADBB0 +'6' (sometimes a few
    bytes later). This is because VSAM has returned in 31 bit
    mode, and we expect 24bit mode. Fix is OW45772, with
    follow-on OW46403 also required due to a PE.
  =====================UPDATE BY Rick Crandall========06/05/2001=
  9.AbendS0C4 in DFSLRH00 or other IMS modules. Regs at abend
    are bad and appear to be a series of double word PSW's from
    low core. R14 points to DFSLRH00, but that is the only IMS
    clue.
    Error is the result of PDF V2.2 by NEON. Reference problem
    id PMIF-4X3MQM.
  1) PDF corrected this at  Version 2, Release 2, Modlevel 4.
  2) All PDF V2.2 customers have been sent the tape with a
     cover letter  stating the existence of '0C4 in DLI region'
     problem and the strong  recommendation to migrate to the
     V2.2.4 level.
  3) Error can be confirmed by locating the last dispatched PST
     under the DLI region TCB and finding an invalid Low Save Area
      (LSA) pointer.
  =====================UPDATE BY Rick Crandall========07/10/2001=
  10.DSP0037I CHANGE.RECON UPGRADE for V7 coexistence
     One way this can happen is if the coexistence APAR PQ27643
     is not on the V6 RECON being upgraded using the V7 utility.
     There is a flag in the SSID at x'3D' = x'71' if this fix
     is properly installed. A library concatenation problem
     could exist. One user found vendor product RRDF had been
     concatenated in front of the IMS RESLIBs, and replaced
     DBRC module DSPCINT0 and DSPCEXT0. This caused code for
     PQ27643 coexistence to be bypassed.
  =====================UPDATE BY charles jones========07/31/2001=
  11. User experienced a SB37 abend on the ACBLIB dataset during
  an ACBGEN of IMS710 system at PUT0010.
  Messages issued to the console were:
  DFS0940I PSB BBC4DLP  HAS BEEN REPLACED IN LIBRARY.
  DFS0941I PSB BBC4DLP  IF USING DL/I SEPARATE ADDRESS SPACE,
  DFS0589I PROCESS COMPLETE FOR PSB-BBC4DLP .  PCB = 0000384,
  DFS0940I DBD DBCMDDA  HAS BEEN REPLACED IN LIBRARY.
  DFS0947I DBD DBCMDDA  BUILD DID NOT FORCE PSB REBUILDS.
  Customer was asked to submit a formal Requirement to address
  this because a resolution would be in the form of an external
  change (user startup parm, etc.) which would not be
  introduced in an apar.
  =====================UPDATE BY ANGELA RAYBORN=======08/01/2001=
  12.DFSHALDB DSPXPDDU S138 abends. Cust was using DFSHALDB to
  convert the IVPDB1 IVP database to HALDB and get frequent S138
  abends.  It is also causing the msg:
  BMC2700I NO VALID DBI PASSWORD FOUND- removing the vendor
  product resolved the S138 abend (abends138).  IMS V7.
  =====================UPDATE BY charles jones========09/05/2001=
  13. After an application abnormally terminated, a GSAM dataset
  was reallocated using IEFBR14 by mistake. A following XRST using
  DISP=OLD for the GSAM file was successful. The application
  continued to insert to the GSAM file and terminated normally,
  but the GSAM dataset was empty.
  The design of GSAM assumes that you are restarting with
  a dataset that hasn't been changed outside of GSAM between
  an update job and the later restart job. GSAM does not restrict
  the altering of datasets between update and restart. However,
  doing so will produce unpredictable results.
  ===================== UPDATE BY Ken Friesen =======09/24/2001=
  14. ABEND SOC1 WHEN TAKING EXTENT ON A OSAM DATABASE
  On the MVS console or in the DLI JOB log are the following
  messages from CA product
  VAM-0617 ---- NONVSAM OUT OF SPACE ON  2 VOLUMES
  VAM-0617 WILL ATTEMPT TO SPAN  3 VOLUMES
  VAM-0617 JOB = dli_job_name STEP = DLI_STEP_NAME
  VAM-0617 PGM = DFSMVRC0
  VAM-0617 DSN = osam_DB_dataset_name
  ...etc...
  The symptom dump shows:
  DFS629I MODID = 37SMST -LO85387   EPA =
  Using the EPA address the module found is
  V37SMST -LO85387 SVD5300 05.34 20010122 (C) 1987-2001
  Computer Associates International, Inc . All rights reserved.
  This is partially resolved by CA apar T25b015. A second fix
  was not identified. Contact CA for assistance.
  Additional keywords: 0C1 S00C1 VAM0617 MSGVAM0617
  PMR: 24482,379,000
  =====================UPDATE BY Ken Friesen ========10/10/2001=
  15.Abend S0C4 in DLI address space. The following message were
  on the MVS console LOG:
  IEC999I IFG0196V,IMSPDLI,IMSPDL
  DUMP TITLE=IEC999I IFG0RR0A,........,IMSPDLI ,IMSPDLI,WORKAREA
              = 7CA1F0
  DFS629I IMS DLI TCB ABEND - SYS 0C4
  DFS629I PSW AT ERROR = 070C1000 80CAB798
  DFS629I MODID = UNKNOWN           EPA = UNKNOWN
  R0 & R1 had the following EBCDIC value "DLI117"
  "C4D3C9F1 F1F710E8". Analysis found that the error occured in
  ACF9096V - 06/09/99 10.40 - ACF2 REL 6.3 COPYRIGHT(C)1993,1998
  COMPUTER ASSOCIATES INTERNATIONAL, INC. -
  R2 to reference something and this address is bad.
  CA had a fix-LO93572-S0C4 ACF9096V+X'798' ACF2 FAILS TO
  RECOGNIZE AN XTIOT for base ACF2 6.3. The fix requires an IPL.
  Additional keywords: S0C4 ABEND0C4 0C4 S00C4 ABENDS0C4
  PMR 72601 B379 C000
  =====================UPDATE BY Rick Crandall========11/13/2001=
  16.BLSR parm not supported for IMS VSAM datasets.
     This parm/product is in conflict with IMS management of
     buffer pools via BLDVRP issued to VSAM based on users
     VSRBF statements in VSMxx member of DFSVSAMP DD card.
     IMS manuals are clear on the IMS requirement, but do not
     attempt to expressly forbid each situations that would
     violate that requirement. BLSR is just one such case.
  =====================UPDATE BY Ken Friesen ========12/10/2001=
  17. S0C3 in module DLPXDCD8 at entry point RESPDIR. The problem
  in BMC product DeltaPlex. Contact BMC for a resolution.
  =====================UPDATE BY Ken Friesen ========12/10/2001=
  18.The follinw message are seen in a job that executes the
  DFSMAID0 utility.
  USER COMPLETION CODE=2103 REASON CODE=00007201
  ACTIVE LOAD MODULE NAME=DFSMAID0
  To resolve the problem add an IBM C language or LE runtime libra
  to the JOBLIB or STEPLIB DD running the DFSMAID0 utility.
  =====================UPDATE BY Ken Friesen ========12/10/2001=
  19.COND CODE 0103 when exxecuting DFSMAID0. The following
  message will also apprear in the sysprint output
  DFSMAID0:  error in writing record statistics to dataspace.
  one cause of this is that the size of the data base or the
  sample size selected is too large. If using SAMPLE= reduce this
  number. If not using the SAMPLE= control card then code a
  SAMPLE= card to reduce the amount of database segments
  being used. Be sure its a 0103 condition code. There are others
  that product message DFSMAID0:  error in writing record
  statistics to dataspace.
  The return code of 103 is also caused by a root segment and
  all its dependents exceeding 16 megabytes. APAR will change
  the return code to a more meaningful message. The APAR number
  is PQ69318.
  =====================UPDATE BY Ken Friesen==========07/21/2002=
  20.DFS0730I I,31, on a HALDB after deleting and redefining
  partitions or partition secondary indexes. Issue DBRC
  CHANGE.DB DBD(myDB) PINIT after deleting and redefining
  and before INIT with DFSUPNT0 of DB and index (if used) and
  load job.Failing to do so will result in the DB and/or secondary
  index to remain empty. Running the load job and DFSUPNT0
  repeatedly without the "Partition Init Needed" flag on will not
  do a thing leaving the DB/index empty..
  RCI RSN31 I 31
  ===================== UPDATE BY Ken Friesen =======08/13/2002=
  21.S0C4 in DFSXBAT0 when running DLI batch. Abend is in
  dfsxcb00+x'730, label XCB00701 attempting to store R1,x'98(,r2).
  R2=000C0000. It was found that an old DFSBSCD0 module was in an
  application library in the steplib above RESLIB.Deleting it
  Additional keywords: ABEND0C4, ABEND00C4
  ===================== UPDATE BY Ken Friesen =======08/13/2002=
  22. See infoapar II11174 entry #23. IMS supports PGMLIB being
  a PDSE.
  IMS PRELOAD of a PDSE is supported for MPP and BMP type regions
  ONLY.  The reason being that for region types other than MPP
  and BMP IMODULE LOAD is used which branch enters into MVS
  fetch to get the module.
  ============ UPDATE BY John Butterweck 2/25/03 ===========
  23. During RELOAD of a HALDB (PHIDAM/OSAM) primary database,
  all ILDSs got message with return code 8 and reason code 98
  DFS871E VSAM ERROR ENCOUNTERED LOADING ILDS WITH DDNAME=
  The unloaded data was create by MIGRATE=YES and options
  INSERT=SEQ and BLDSNDX=NO. RPLERRCD 152(X'98') means not enough
  buffers are available to process your request (shared resources
  only).Increasing the number of buffers for all subpools resolved
  the problem.
  addition keyword: PARTITION DB RSN RSN98 RSN098 RC8 RC08 8 98
  =====================UPDATE BY Ken Friesen==========04/24/2003=
  24.Cancelling a LIST.RECON job results in ABEND438. A CLOSE is
  done from DSPURI20+x'4F4. An ABEND222 occurs. Then a RESERVE is
  done from DSPCABN0+x'4D8. Resulting in ABEND438 ISGGNQDQ+x'1C22
  -UW72168. The reserve was for RECON1 but it closed and
  deallocated already.
  Addition keywords: AbendS438, S438, S0438, S222, ABENDS222,S0222
  ============ UPDATE BY Ken Friesen  06.11.2003 ===========
  25. CA11 module U11D8650+x'598' with a compile date of 02/04/03
  and a FMID of CL73000 is branch entering page free to free
  storage using regs 1 and 2 with value of On branch entry to Page
  Services-bit 0 of reg 1 being off indicates that the caller is
  using REGISTER format. This means reg1 contains the beginning
  address of the page request and reg2 contains the end address.
  Reg3 byte 2 contains x'02' which indicates PAGE FREE. This page
  FREE is spanning from 00000100 to 28130CB8 . CA temporary fix
  number is T3C6005.
  addition keywords: S0D3, abends0d3, abends00d3, abend0d3, 0d3,
  page free, page-free, page fix, page-fix, unpage-fix, IMS
  logger, DFSFDLG0 seq# 18627100
  =====================UPDATE BY Ken Friesen==========06.24.2003=
  26.First x'9904 data capture record not created. BMC module
  DLIODDC7 replaced module DFSDCAP0. Renaming DLIODDC7 corrected
  the problem. Contact BMC for a coding soplution.
  =====================UPDATE BY Catherine BAUE=======07/09/2003=
  27.User experienced an abendB37 on the ACBLIB dataset during
  ACBGEN with BUILD DBD=. After correcting the space problem,
  they submitted the job again.  The joblog showed;
    DFS0947I DBD dbdname BUILD DID NOT FORCE PSB REBUILDS.
  The affected PSB's were not rebuilt. They need to run BUILD PSB=
  for the affected PSB's or BUILD PSB=ALL to force rebuild the
  the affected PSB's.  IMS V7.
  =====================UPDATE BY Ken Friesen =========07/29/2003=
  28.Abends0C1 in module DFSDBLR00+x'180 label PSBREAD1. The
  overlay is 8 bytes starting in the middle of an instruction.
  The overlay in this case was text "EMVTA4  " which is an
  Omegamon/IMS VTAM id. The Omegamon/IMS feature Transaction Trace
  Facility (TRF) caused the overlay.Contact Candle Corp. for PTF
  QIDG059 which correct the coding error.
  Addition keywords: abend0c1 abend00c1
  ======================UPDATE BY KEN FRIESEN=========07/29/2003=
  29 A compatibility problem exists between OEM product EDP v2.4.1
  and the SQData. Each product handles data capture data in real
  time and EDP interfers with SQData's ability to capture all
  data capture changes. Removing EDP resolved the incompatibility.
  =====================UPDATE BY KEN FRIESEN==========03/04/2004=
  30. MSGDFS0730I RCB RSN10 from an online system running on
  z/OS 1.5. Batch jobs abend with U3303. The resolution is to
  install fix UA08649/OA05999.
  additional keyword. DFS0730I, DFS730I B,10 5695DF106 Release 1H0
  

Problem conclusion

• INFO APAR Items
  

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