gtpm2m0oMigration Guide: Program Update Tapes

Recommended Migration for an Established TPF 4.1 System User

This section provides information about migrating your existing TPF 4.1 system to program update tape 2 (PUT) 2.

Note:
IBM recommends that you install PUT 2 in its entirety on your TPF 4.1 system rather than installing individual APARs.

There are 2 methods you can use to install PUT 2:

Before You Begin

Before you begin to install PUT 2 on your TPF 4.1 system, please read the following:

Installing PUT 2 on a New Image in an Existing TPF 4.1 System

The following procedure enables you to install PUT 2 without requiring a complex-wide outage.

To Install PUT 2 on a New Image

  1. Enter ZIMAG to define a new image. This image will be referred to as the PUT 2 image throughout this procedure.
  2. Calculate the number of #XPRGn and #PVRn records that are needed for each image that will have PUT 2 loaded to it.
  3. Define the appropriate RAMFIL statements in the SIP input deck.
  4. Create a new FACE table (FCTB) using the updated SIP input deck. (This is the SIP input deck you updated in step 3.) Use a version of FCTBG that is compatible with the TPF 4.1 system at the PUT 1 level.
    Note:
    ISO-C support and APAR PJ13765 made several changes to FCTBG.
  5. Load the new FCTB to the image that is currently active.
  6. Ensure that enough room exists for the CIMR components by running GTSZ and reassembling CTKX.
  7. Reassemble and recompile all required programs against the PUT 2 libraries.
  8. Load the following segments to the image that is currently active:
    CILA CILB CILC CILD CILE
    CILF CILG CILH CIL0 CIL1
    CIL2 CIL3 CIL4 CIL5 CIL6
    CIL7 CLEM CRLW UELI  

    You can load and activate these segments using E-type loaders.

    Note:
    If you use E-type loaders to load these segments, E-type loaders handles unallocated programs for you. If you use another loader, such as TLDR/ALDR, you must allocate these segments (CILE, CILF, CILG, CILH, and CRLW).
  9. Create or update keypoint A for each CPU ID that includes:
  10. Create a new system allocator (SALO) that includes all newly created segments.
  11. Recompile and run SALO against the PUT 2 libraries to create an updated IPAT and TABLE.
  12. Relink all applicable programs.
  13. Define a full load deck for TPFLDR/TLDR that includes the:
  14. Run TPFLDR/TLDR.
  15. Perform a full load to the PUT 2 image.
  16. Enable the PUT 2 image.
  17. Hard IPL (IPL with CLEAR) the TPF 4.1 system and choose the PUT 2 image as the active image.
  18. Hard IPL any other TPF 4.1 systems that are defined in your complex using the PUT 2 image as the active image.

    Before you continue with the remainder of this step, see Pool Expansion (PXP) Support (APAR PJ17912) for more information about pool conversion and pool conversion fallback.

    Stage 1 of pool conversion is complete when the TPF 4.1 system is IPLed successfully. Stage 2 of pool conversion, which is performed by entering ZPMIG CONVERT, may be completed after all processors in your complex are IPLed with the PUT 2 image.

    Attention: If it is necessary to IPL the original image (PUT 1 image) or to return to stage 1 of pool conversion, enter ZPMIG FALLBACK.

    After stage 2 of pool conversion is complete, you can copy the PUT 2 image to the original (PUT 1) image.

  19. PUT 2 migration is now complete.

Summary of Changes to the General File Format for PUT 2

This section provides information about the changes made to the loader general file (LGF) format JCL. These changes are required for the PUT 2 version of TPFLDR/ALDR to complete successfully. Your JCL or procedures may need to change.

Table 50 is for a sample BSS LGF that was built on a 3380 device. Each area is formatted by a separate job stream and each formatted area is now known by a unique DSN.

See TPF System Generation for samples of the FMTR40 jobs produced by the system initialization program (SIP).

Table 50. Sample Changes to the General File Format for PUT 2

Program Name Statement At PUT 1 Level At PUT 2 Level
FMTR40 EXEC L1A2E (L1A2E, L1A3E, and so on)
FMTR40 FMTDD3 DSN=GNFLBSS DSN=(GNFLBSS, GNF2BSS, and so on)
FMTR40 FMTDD1
FMT  L  N 0000001  0000001  3380
FMT  4  N 0000002  0003413  3380
FMT  END
For job L1A2E:
FMT  L  N 0000001  0000001  338L
FMT  END
 
For job L1A3E:
FMT  4  N 0000002  0003413  338L
FMT  END
DYOPM40 DGF DSN=GNFLBSS DSN=GNFzBSS
Notes:
  • Prior to PUT 2, FMTR40 could process all FMT cards in one job stream. After installing PUT 2, FMTR40 allows one FMT and one FMT END statement to be included. Therefore, each area must be formatted by a separate job and each area that is formatted has a unique DSN.
  • If DGF/RGF is located on the LGF, it will have a unique DSN.
  • In this table, z is the next highest number after GNF2BSS.
  • See TPF System Generation for sample JCL.

Problem Determination Table

Table 51 provides information about some common errors that may occur when trying to bring up a TPF 4.1 system after performing the full load.

Table 51. Sample Changes to the General File Format for PUT 2

Program Error Possible Cause Recommended Action
CTIN Failure, PSW=040A0000 00FFFFFF
  • SALO was compiled against a library that contains an old version of C$IDSPAT.
  • CTKA was not updated or was updated improperly.

  1. Create a new IPAT by recompiling SALO against the latest PUT 2 libraries.
  2. Review step 9 of To Install PUT 2 on a New Image.

I000000D6 The PROGn area has a program base incompatible with the IPAT being loaded Create and load the proper IPAT version.
CTIN - FACE ERROR BUILDING PAT The FCTB version is incompatible with the IPAT version. Create and load the proper FCTB or IPAT version.
CTIN0082E - CT00 INSUFFICIENT STORAGE FOR POOL A TPF image without PXP support applied was IPLed but file pool directories are in a converted state.
  1. IPL the PUT 2 image.
  2. Enter ZPMIG FALLBACK.

Note:
Although many problem symptoms may occur if one or more TPF components are not created properly, the previous list presents some common errors.