II14702: AVOIDING MIGRATION ISSUES FOR FASTER MIGRATION HINTS & TIPS TO V10

Subscribe to this APAR

By subscribing, you receive periodic emails alerting you to the status of the APAR, along with a link to the fix after it becomes available. You can track this item individually or track all items by product.

Notify me when this APAR changes.

Notify me when an APAR for this component changes.

APAR status

  • INTRAN

Error description

  • Due to customers running into issues while migrating, this Info
    APAR was developed with the following bullets to aid customers
    to have a faster Migration by avoiding known migration issues.
    .
    o  If CATMAINT | DSNTIJTC fails, please open an SR pmr and list
       the error and provide the appropriate doc for the error -
       when it happens.  L2 may be able to identify the +fix or
       resolution for V10 or recommend Fallback to V8 (asap) and
       then Rebuild, Recovery and/or Repair problem V8 objects
       there.  Especially for DSNxxxxx names objects.
       +  Please use the  1 (800) IBM SERV =1 (800) 426-7378 number
          for escalating, provide the error and error doc asap.
       +  L2 Duty personnel will only get called out if the PMR
          is a P1S1 and resides on our DB2PS,L117 queue.
    .
    o  Having correct migration Maintenance level and/or latest
       Migration fixes applied first may help avoid any issues
       during V10 Migration.  Please review the Enhanced Hold Data
       website for detailed information for planning your migration
       & maintenance strategy and coordinate with your z/OS System
       Programmer (daily, weekly, monthly quarterly.. et.al.).  LINK
       http://service.software.ibm.com/holdata/390holddata.html
    .
       - FIX CATEGORIES IBM.Migrate-Fallback.DB2.V10
         and    IBM.Coexistence.DB2.SYSPLEXDataSharing
       - KEYWORD  DB2MIGV10/K  &  DB2COEXIST/K
       are used to represent fixes that allow prior releases of DB2
       to migrate to or fall back from DB2 V10 and Fixes that enable
       to coexist when in data sharing mode respectively.
       - See more information on FIXCAT at Link:
       http://www-03.ibm.com/systems/z/os/zos/smpe/fixcategory.html
    
       Recent fixes for Migrations include:
       +  UK75727 v10/PM52752  04E E40601 DSNUEXDL+1402  (PE/y fix)
       +  UK76530 v10/PM56449  Multi abends Rollback/(2) DDF
          Threads (Hiper) includes  04E 00C90101 DSNKDLE: 500A
          & DSNB250E w/DSNKUNR2
       +  UK80924 v10, UK80925 v9/PM62119  04E E70014 DSNXLDDL+2234
          when accessing catalog objects or resources.
    .
    o  Dropping the User Defined objects in V8 or V9.  Use this
       query to find them.  (See V10 SMS-Managed bullet below.)
          SELECT * FROM SYSIBM.SYSINDEXES
             WHERE DBID = 0006
             AND OBID >398 AND OBID < 600;
    
       To get a good DSNTIJTC RC:00, and once established on DB2 V10
       CM8 you can elect to Create the ones you need again.
    .
    o  Remember, while in {Shadow like} V10 migration & DSNTIJTC has
       not successfully Run, all Catalog or Directory TBS updates or
       changes are restricted, resulting in 00C900A6 and/or 00D70014
       issues.  Users will only be allowed to do r/o activities.
       +  If you  -Display UTIL and  -Display Thread on V8 - V9 and
          eliminate these users; it will help avoid delays caused by
          r/w requests - errors once DB2 V10 is started.
       +  There should be no Utilities, no DDL, no GRANTs or
          REVOKEs, and no BINDs.
       +  Have Mgmt send out EMails to DBA's & System Programmers
          to ensure they are aware of the migration timeframe.
       +  Ensure no long running Batch jobs have switched times or
          will exist during your planned migration timeframe.
       +  There are a number of vendor products that access the
          catalog and never commit.  These will also interfere with
          the DSNTIJTC and DSNTIJEN jobs.  Such products must be
          terminated before running either of these jobs.
       +  If 00C900A6 with ABEND0C1 DSN9SCN9 & DSNFWTO occurs during
          DSNTIJTC, CATMAINT to be restarted.  Please collect the
          Dump & Logrec for L2 review and confirmation on CA's
          ASR230Z 01/29/07, II14195 - CA Fix Q094419.
       +  If V10 CM9 Catmaint received
          DSNU778I  <ssid DSNUEXDL - ERROR PROCESSING SQL STATEMENT
          SQL CODE IS: -904 , DSNXIDBTB UNAVAILABLE RESOURCE,
          Type 200 for >> DSNATPDB.DSNASDMU << followed by a
          DSNT500I  <ssid  DSNTDMI RESOURCE UNAVAILABLE, 00C900A6
          TYPE 100 & NAME DSNDB06
          To resolve:  Bring DB2 V10 ssid down, switch back to V9
          starting tasks & libraries on V9, bring DB2 V9 up & down,
          then swap V10 JCL & libraries back in, start DB2 V10, and
          rerun CATMAINT.
    .
    o  After migrating to V10.  If you receive error TMD32102I
       Message '(xxxxx) is not setup to monitor SSID at v9.9.9' and
       a disconnection occurs, upgrade your TMON DB2 to V5 as a
       requirement level for DB2 V10.
    .
    o  If DB2 Client or DB2 Connect gets SQL1598N after DB2 V10 CM9,
       review Link of DB2 Connect Documentation to Reactivate the
       server side license again and follow instructions.
       http://www-01.ibm.com/support/docview.wss?uid=swg21589320
    .
    o  If a failed V8 utility was not terminated before DSNTIJTC job
       was started for V10 migration, a Terminate will get this msg:
        DSN
              -TERM UTIL(*)
        DSNU115I  -DB2P DSNUGDIS - UNLOAD UTILITY WITH
                    UTILID = UNLOAD.BCT CAN ONLY BE
                    TERMINATED BY A MEMBER WITH THE
                    SAME RELEASE = 810
        DSN9022I  -DB2P DSNUGCCC '-TERM UTIL' NORMAL COMPLETION
       Based on DSNU115I message, customers need to go back V8 to
       terminate that utility and then continue the migration.  This
       emphasizes the importance of -Display & -Terminate commands.
    .
    o  Having a clear go / no go timeline is also good, but you must
       stick to it.  If you think you can fallback after DSNTIJTC
       fails and do the required recovery actions (rebuild, recover,
       repair, (terminate)) on V8 | V9 ... and then try migrating to
       V10 & run CATMAINT | DSNTIJTC again--do so.  If L2 recommends
       a matching +fix, & you can accept/apply it in time to re-try
       running CATMAINT | DSNTIJTC--do so.
    .
    o  A Soft Fallback when DSNTIJTC fails or even if RC:00 during
       the initial phase of V10 to V10 CM is relatively easy.  Every
       thing it creates or alters is against the real table spaces
       or indexes.  However, V8 and V9 can deal with all of these
       changes with no problem.  It is safe, easy and fast to
       fallback to the original V8 | V9 catalogs.
       +  Remember to Term Util  ( i.e. V10 DSNTIJTC CATMAINT job)
       +  If you ever run into a severe issue after migrating to
          V10 Compat mode,always contact IBM DB2 first.
          +  If the error occurred well after V10 CMx, please open
             a separate SR PMR, include the error and error doc.
    .
    o  If DSNTIJTC Create index receives resource sqlcode-904
       DSNXICX2 with 00E40601 & 00C90305.  The work files appear
       exhausted.  Check and Reset ZPARM MAXTEMPS = 0 if nnn,
       and try rerunning DSNTIJTC again.
    .
    o  Starting DB2 in Maint mode is good, even if it's not in the
       Migration steps.  It limits users to install SYSADM, install
       SYSOPR,and SECADM.  For Data Sharing, it restricts access
       only on the DB2 member you execute the command from.  Other
       members of the group are unaffected.
       +  Most Data Sharing migrations are less time consuming if
          the other members are down.
       +  If data sharing with members on-line (24x7 requirement)
          * must route work away from the DB2 mbr where DSNTIJTC is
            running.
          * must temporarily change workload balancing and
            sysplex routing scheme.
    .
       ** Please review PE/y PI10085 APAR ++HOLD ACTION if missed...
          where single group SDSNLOAD causes Maintenance MISMATCH
          across live members leading to ESCA GROWTH (WLM runaway
          Stored Procedure requests) which could potentially lead to
          group outage.  UK93886 PE/PM77559 SQLCODE905 DSNX9CAC.
    .
       Note:  Stored procedures calls will fail with SQLCODE -471
          reason code 00E79004, when DB2 is started in maintenance
          mode.  Therefore, restart DB2 in normal operating mode
          before attempting to run DSNTIJRV, the validation job
          for DB2-supplied routines.
    .
          DB2 Development is investigating an SQLCODE-117 issue when
          DSNTIJRT is run in Maint Mode where it fails to notify the
          other members to clear their cached values for stored
          procedures.
      + If DSNTIJRT failed with this:
          x FAILED    IX       SYSIBM        DSNMQPLX
          x FAILED    IX       SYSIBM        DSNMQSRX
        In the V9 libraries, job DSNTIJSG was stripped out to drop
        & recreate the DB2MQ objects/functions. DSNTIJRT rerun good.
    .
    o  For S222 issues:  Check IRLM status.  Confirm it has a unique
       naming convention, correct Release Levels & separate unique
       name from DB2.  Size | Space increases are necessary for V10.
    .
    o  Check for an active bad IRLMPROC and disable if you receive
       DSNT803I  > DB2 CANNOT IDENTIFY TO IRLM.
       *DSNV086E > DB2 ABNORMAL TERMINATION REASON=00E30800
       *IEF450I DSNRMSTR DSNRMSTR - ABEND=S04E U0000 REASON=00E30800
    .
    o  For DSNT501I :DB2x DSNIWKFL RESOURCE UNAVAILABLE msg w/REASON
       00C90084, define more workfile 32K table space ..per II14587.
    .
    o  After running DSNTIJTC & CATMAINT CC=0 a DSN1CHKR was run on
       DSNDB01.DBD01 receiving DSN1802I RING/LINK Chain Broken and
       DSN1812I ORPHAN ID's.  Please ensure the pageset for DBD01 is
       completely stopped prior to running DSN1CHKR.
       You should use these steps:
        1.) -STOP DB(DSNDBD01) SPACENAM(DBD01)
        2.) -DIS DB(DSNDBD01) SPACENAM(DBD01)
             status should be STOP
        3.) Rerun DSN1CHKR.
    .
    o  V10 SMS-Managed EA requirements (Extended Addressability)
       With DB2 V10, SMS will manage and control all new catalog
       changes, all the conversions to universal table spaces and
       all new catalog and directory indexes and table spaces.  All
       the new indexes and table spaces being added in DB2 V10 will
       also use Extended Addressability (EA).  If DB2 detects that
       the catalog and directory are not in an SMS-compliant
       environment, an abend or DSNTIJEN or DSNTIJTC job failure
       can occur.  Messages IDC3009I IGG0CLAT 80, DSNTIJEN or
       DSNTIJTC failure with DSNT500I & 00D70025 issues, and/or
       when STOGROUP "00000001" involved with REASON 00E40601.
    .
      + A fast circumvention is to Drop & Recreate the problem
        User-Defined objects and then rerun DSNTIJEN from the start.
      + Any such data sets not changed to SMS management during the
        initial DB2 V10 migration process will be SMS-Managed the
        next time the associated table space is ReORGANIZED -- for
        example, by job DSNTIJEN if the table space is those
        modified during ENFM.
    .
    o  If DSNTIJEN receives IEC143I 213-04, on an ENFM00nn IC step,
       check that your inline image copy DATACLAS is correct and NOT
       created using linear VSAM.
    .
    o  If DSNTIJEN RC 00E40601 & sqlcode -148, review PM48358 & doc
       changes to- Migration step 21: Enable change data capture
       To re-enable change data capture:
         ALTER TABLE SYSIBM.tablename DATA CAPTURE CHANGES;
       Where tablename is the name of the catalog table (with
       DB2 Replication Server z/OS V10.1).  Otherwise, customers
       must handle the Online changing of the ZPARM twice
       RESTRICT_ALT_COL_FOR_DCC parameter from YES to NO after the
       Alter and then NO to YES before starting Replication again.
      + IF customers  are at DPROP R V10+ levels they can use the
        RESTRICT_ALT_COL_FOR_DCC  NO (taking the default) and use
        the statement to make appropriate changes.
      + DPROP V10 (recommends APAR v10.0 PM94614/(UK97279) which
        is a full refresh +Fix  without needing pre-reqs).
    .
    o  APAR PM92287/V10-UK96402,V11-UK96403 fixes the following:
        ABEND04E RC00C90101 at DSNIIMSI:5006
        ABEND04E RC00C202AA in DSNB1PER
        ABEND04E RC00C90101 at DSNIRSVP:1005
       when running into ENFM DBD serialization issues on DSNTIJEN.
    .
    o  APAR PM96115/V10-UI12313 V11-UI12314  CL13/11/08
       During DSNTIJEN STEP ENFM0040 an ABEND04E 00E40601
       DSNUTILA.DSNUECM2+1814 or during DSNTIJTC Change Data
       Capture 00E40601.  Ensure Module: DSNUECM2 is current.
    .
    o  When running DSNTIJCI, and you receive  DSNU3343I DSNUKINM -
       REAL-TIME STATISTICS INFORMATION MISSING FOR TABLESPACE
       DSNDB06.SYSTSPKS , along with DSNU044I DSNUGSRP - ERROR FROM
       SORT COMPONENT RC=16, UTILITY STOPPED, with ABENDED REASON
       00E40005 and 00E40347 :  RTS is disabled requiring RUNSTATS
       to be run first and then re-run DSNTIJCI (Check Index).
    .
    o  Content Manager logon getting CMBException SQLCODE= 725
       SQLSTATE=42721 SQLERRMC=CURRENT SCHEMA via Distributed JDBC
       driver.  Ensure eClient using data source with UPPER CASE
       NAME and the WAS Connection Pooling is disabled.
    .
    o  DSNT408I SQLCODE = -303, ERROR:  A VALUE CANNOT BE ASSIGNED
       TO OUTPUT HOST VARIABLE NUMBER 9 BECAUSE THE DATA TYPES ARE
       NOT COMPARABLE.  An internal change between V8 & V10 with
       data type compatibility and introduction to IMPLICIT CAST
       may affect your application.  Please consider reviewing this
       Cobol Data Type Table Link for application changes on V10.
    .....
    http://publib.boulder.ibm.com/infocenter/dzichelp/v2r2/topic/com
    .ibm.db2z10.doc.apsg/src/tpc/db2z_equivalentcoboldatatypes.htm
    .....
    Link covering Implicit Casting
    .....
    http://www-01.ibm.com/support/knowledgecenter/SSEPEK_10.0.0/com.
    ibm.db2z10.doc.wnew/src/tpc/db2z_10_implicitcasting.dita?lang=en
    .....
    Link covering Application & SQL incompatibilities for migration
    from V9.1
    .....
    http://www-01.ibm.com/support/knowledgecenter/SSEPEK_10.0.0/com.
    ibm.db2z10.doc.inst/src/tpc/db2z_relincompatapplsqlfromv9.dita?l
    ang=en
    .....
    .
    o  UI14060/PI07855 PE/ Hiper  CL14/02/05  F401
       DSNTIJEN FAILS DSNU777I REORG ENFM WITH BROKEN DATABASE
       DSNU016I ABEND04E 00E40600 INVOLVING LOB MERGE with DBD
       anywhere in the VSAM name.  DB2MIGV10/K  MODULE: DSNUECML
    .
    o  During DSNTIJEN's ENFM0027 step, a customer received various
       DTS SW SCC Product msgs ACC20621-E, ACC20210-E, ACC21231-E
       along with insufficient space IGB17272I on the SPT01 Reorg.
       Basically the B37 required more space, but the volume table
       of contents (VTOC) was full.
       Increasing the SYSREC DD Statement SPACE=  setting  allowed
       the step & job to complete.  Due to SPT01 size requirements
       for that ssid, it may take longer (+- 15 minutes).
       //             SPACE=(CYL,(100,20),RLSE)
                  increased to  (500,500)
    .
       Another ENFM0027 issue may occur requiring DB2's assistance
       to identify orphaned rows involving YEARS old packages.  It
       is best to cleanup old packages and limit the number of
       versions that exist if staying on V10 CMx for longer than
       three+ months before migrations to V10 NFM.  Consider
       issuing  FREE PACKAGE xxxxx PLANMGMTSCOPE(INACTIVE) .
    

Local fix

Problem summary

Problem conclusion

Temporary fix

Comments

APAR Information

  • APAR number

    II14702

  • Reported component name

    PB LIB INFO ITE

  • Reported component ID

    INFOPBLIB

  • Reported release

    001

  • Status

    INTRAN

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt

  • Submitted date

    2012-12-06

  • Closed date

  • Last modified date

    2014-07-08

  • 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



Rate this page:

(0 users)Average rating

Add comments

Document information


More support for:

z/OS family

Software version:

001

Operating system(s):

MVS

Reference #:

II14702

Modified date:

2014-07-08

Translate my page

Machine Translation

Content navigation