Topic
IC4NOTICE: developerWorks Community will be offline May 29-30, 2015 while we upgrade to the latest version of IBM Connections. For more information, read our upgrade FAQ.
11 replies Latest Post - ‏2012-11-01T17:46:46Z by davidknibb
SystemAdmin
SystemAdmin
124 Posts
ACCEPTED ANSWER

Pinned topic CICS TS 5.1 open beta - new beta driver now available

‏2012-10-12T10:49:50Z |
All - an updated version of the CICS TS 5.1 open beta code has just been made available on the beta download site.

We encourage you all to update with this latest version of the open beta code as it has a significant number of new features and functions.

Any questions please get back to us

Thanks

Francis
Updated on 2012-11-01T17:46:46Z at 2012-11-01T17:46:46Z by davidknibb
  • dysonbags
    dysonbags
    16 Posts
    ACCEPTED ANSWER

    Re: CICS TS 5.1 open beta - new beta driver now available

    ‏2012-10-15T10:12:54Z  in response to SystemAdmin
    Hi Francis,

    I downloaded the new beta code. I decided to install it over the old beta code, maybe not what you envisaged? But I get the following messages during the receive:

    HCB6800 RE-RECEIVED FUNCTION SMPTLIB LOADED
    HCIZ300 NOT RECEIVED FUNCTION REWORK LEVEL IS NOT GREATER THAN THE VERSION ALREADY RECEIVED
    H0B5110 NOT RECEIVED FUNCTION REWORK LEVEL IS NOT GREATER THAN THE VERSION ALREADY RECEIVED
    H0B7110 NOT RECEIVED FUNCTION REWORK LEVEL IS NOT GREATER THAN THE VERSION ALREADY RECEIVED
    H0Z2110 NOT RECEIVED FUNCTION REWORK LEVEL IS NOT GREATER THAN THE VERSION ALREADY RECEIVED
    JCB680D RE-RECEIVED FUNCTION SMPTLIB LOADED
    JCB680L RE-RECEIVED FUNCTION SMPTLIB LOADED
    JCB680M RE-RECEIVED FUNCTION SMPTLIB LOADED
    JCB680W RE-RECEIVED FUNCTION SMPTLIB LOADED
    JCB6801 RE-RECEIVED FUNCTION SMPTLIB LOADED
    JCB6802 RE-RECEIVED FUNCTION SMPTLIB LOADED
    JCB6803 RE-RECEIVED FUNCTION SMPTLIB LOADED

    I'm assuming/hoping that the "not received" messages are due to there being no updates to those particular FMIDs and it is safe to continue with the apply?

    Cheers
    Graham
    • SystemAdmin
      SystemAdmin
      124 Posts
      ACCEPTED ANSWER

      Re: CICS TS 5.1 open beta - new beta driver now available

      ‏2012-10-15T10:24:33Z  in response to dysonbags
      Hi Graham,

      Great to hear you have downloaded and started the install. The following FMIDs relate to the following:

      CICS Service Flow Feature: HCIZ300
      CICS REXX Common: H0Z2110
      CICS REXX Runtime: H0B5110
      CICS REXX Dev: H0B7110

      These are the same as the last open beta and have not been updated. So the SMP/e receive errors are unfortunately expected.

      Cheers

      Adam Coulthard
    • SystemAdmin
      SystemAdmin
      124 Posts
      ACCEPTED ANSWER

      Re: CICS TS 5.1 open beta - new beta driver now available

      ‏2012-10-15T11:11:37Z  in response to dysonbags
      Hi Graham

      Thanks for the information and sorry we caused you problems.

      As Adam pointed out these FMIDS have not changed as they are the REXX for CICS anbd Service Flow Feature FMIDS - they are not needed for CICS TS testing. For the SMP/E Apply & Accept step please remove them and you should be ok.

      For the Beta we always ask that you use a new SMP/E environment.

      If you find you are still having problems please let me know.

      Thanks

      Francis
      • dysonbags
        dysonbags
        16 Posts
        ACCEPTED ANSWER

        Re: CICS TS 5.1 open beta - new beta driver now available

        ‏2012-10-15T15:23:59Z  in response to SystemAdmin
        Hi Francis,

        No problem at all. Trust me to do things differently. :-) Anyway I've tested the upgrade path for you (as opposed to the clean install). Regions are up and running. No obvious problems yet.

        Cheers
        Graham
        • dysonbags
          dysonbags
          16 Posts
          ACCEPTED ANSWER

          Re: CICS TS 5.1 open beta - new beta driver now available

          ‏2012-10-16T08:24:31Z  in response to dysonbags
          Hi Adam, Francis,

          Just noticed an error in the new beta. I'm getting these messages on startup.

          09.19.31 STC14955 +GSVC999I CA SYSVIEW for CICS terminated
          09.19.51 STC14955 +DFHFC0951 Z000CICS 821
          821 Non-RLS OPEN of file NDVRSYM failed. DSNAME not available from JCL or
          821 resource definition. Module DFHFCFS.
          09.19.53 STC14955 +DFHAP0001 Z000CICS An abend (code 0C4/AKEA) has occurred at offset X'000001F6' in module EZACICME.
          09.19.53 STC14955 +DFHME0116 Z000CICS 823
          823 (Module:DFHMEME) CICS symptom string for message DFHAP0001 is
          823 PIDS/5655BTA00 LVLS/680 MS/DFHAP0001 RIDS/DFHSRP PTFS/Beta12 AB/S00C4
          823 AB/UAKEA RIDS/EZACICME ADRS/000001F6
          09.19.53 STC14955 +DFHDU0205 Z000CICS A SYSTEM DUMP FOR DUMPCODE: AP0001 , WAS SUPPRESSED BY THE DUMP TABLE OPTION FOR THIS
          DUMPCODE
          09.19.53 STC14955 +EZY1288E 16/10/12 09:19:53 CICS/SOCKETS MODULE EZACICME ABEND ASRA
          09.19.54 STC14955 +DFHFC0952 Z000CICS 826
          826 Dynamic allocation of Non-RLS file PROTCPF failed. Return code
          826 X'0004',X'0210' in module DFHFCFS.
          09.19.54 STC14955 +DFHFC0955 Z000CICS Associated data set is ISVI.CAI.TOOLS.V8R5.PROTCPF. Module DFHFCFS.
          09.20.09 STC14955 +DFHAP0001 Z000CICS An abend (code 0C4/AKEA) has occurred at offset X'000001F6' in module EZACICME.
          09.20.09 STC14955 +DFHME0116 Z000CICS 829
          829 (Module:DFHMEME) CICS symptom string for message DFHAP0001 is
          829 PIDS/5655BTA00 LVLS/680 MS/DFHAP0001 RIDS/DFHSRP PTFS/Beta12 AB/S00C4
          829 AB/UAKEA RIDS/EZACICME ADRS/000001F6
          09.20.09 STC14955 +DFHDU0205 Z000CICS A SYSTEM DUMP FOR DUMPCODE: AP0001 , WAS SUPPRESSED BY THE DUMP TABLE OPTION FOR THIS
          DUMPCODE
          09.20.09 STC14955 +DFHAP0001 Z000CICS An abend (code 0C4/AKEA) has occurred at offset X'000001F6' in module EZACICME.
          09.20.09 STC14955 +DFHME0116 Z000CICS 832
          832 (Module:DFHMEME) CICS symptom string for message DFHAP0001 is
          832 PIDS/5655BTA00 LVLS/680 MS/DFHAP0001 RIDS/DFHSRP PTFS/Beta12 AB/S00C4
          832 AB/UAKEA RIDS/EZACICME ADRS/000001F6
          09.20.09 STC14955 +DFHDU0205 Z000CICS A SYSTEM DUMP FOR DUMPCODE: AP0001 , WAS SUPPRESSED BY THE DUMP TABLE OPTION FOR THIS
          DUMPCODE
          09.20.09 STC14955 +EZY1288E 16/10/12 09:20:09 CICS/SOCKETS MODULE EZACICME ABEND ASRA

          Looking at an old log for the previous beta, I get the following at the same point:
          10.52.18 STC14301 +DFHFC0951 Z000CICS 005
          005 Non-RLS OPEN of file NDVRSYM failed. DSNAME not available from JCL or
          005 resource definition. Module DFHFCFS.
          10.52.20 STC14301 +EZY1258I 02/10/12 10:52:20 EZACIC01 ENTRY POINT IS 0FDFD938
          10.52.20 STC14301 +EZY1258I 02/10/12 10:52:20 TRU00000 ENTRY POINT IS 0FDF95E8
          10.52.20 STC14301 +EZY1291I 02/10/12 10:52:20 LISTENER TRANSACTION= CSKL TASKID= 0000052L ACCEPTING REQUESTS VIA PORT 22602
          10.52.20 STC14301 +DFHFC0952 Z000CICS 009
          009 Dynamic allocation of Non-RLS file PROTCPF failed. Return code
          009 X'0004',X'0210' in module DFHFCFS.

          Would you like me to collect any other info?

          Cheers
          Graham
          • davidknibb
            davidknibb
            12 Posts
            ACCEPTED ANSWER

            Re: CICS TS 5.1 open beta - new beta driver now available

            ‏2012-10-16T14:56:42Z  in response to dysonbags
            Hi Graham,
            Based on this log, you are having two failures to open files, and these are being followed by 0C4s in EZACICME. This module is part of z/OS TCPIP, so you may need assistance from the TCPIP support team to complete debugging this. We have not been able to recreate the 0C4s here in Hursley.

            To deal with the file control errors:
            09.19.51 STC14955 +DFHFC0951 Z000CICS 821 
            821 Non-RLS OPEN of file NDVRSYM failed. DSNAME not available from JCL or 
            821 resource definition. Module DFHFCFS. 

            This indicates that you are trying to open file NDVRSYM, but no data set name has been provided. I do not think it will be possible to open this file without providing the data set name.
            09.19.54 STC14955 +DFHFC0952 Z000CICS 826 
            826 Dynamic allocation of Non-RLS file PROTCPF failed. Return code 
            826 X'0004',X'0210' in module DFHFCFS. 

            In this case, the x'0210' response from dynamic allocation suggests that the data set is open to another job. You should review other jobs using this data set.
            If you would like us to to attempt to see what's going on with the 0C4s in EZACICME then we would need to see a dump. From the log you have supplied so far it appears that the dump may be being suppressed. If that is the case then you would need to remove this suppression, and recreate the problem. Once you have the dump please terse and FTP it as follows:
            ftp to the following site:
            ftp.hursley.ibm.com
            login as anonymous
            p/w = internet email ID
            cd to directory /pub/import/cicsbeta/openbeta
            Then do a binary PUT of your tersed dataset
            Once this is complete, can you please reply with the name of the dump/s you have sent in.

            Thanks, David
            • dysonbags
              dysonbags
              16 Posts
              ACCEPTED ANSWER

              Re: CICS TS 5.1 open beta - new beta driver now available

              ‏2012-10-16T15:24:11Z  in response to davidknibb
              Hi David,

              Sorry, I know that the file open errors are irrelevant to this. Shouldn't have included them.

              I've FTPed a tersed dump to the directory you mentioned. It's called beta12.ezacicme.sdump. I have 2 other dumps if you need them.

              Cheers

              Graham
              • davidknibb
                davidknibb
                12 Posts
                ACCEPTED ANSWER

                Re: CICS TS 5.1 open beta - new beta driver now available

                ‏2012-10-17T14:24:52Z  in response to dysonbags
                Hi Graham,
                I started to look at your dump yesterday. I noticed that internal trace is turned off, and the trace table is very small anyway (only 64K). For a beta I would suggest turning on the internal tracing to aid problem determination.

                I found out this morning that we have now been able to reproduce this internally. It appears that there may be a relation to threadsafety. From our testing it seems that if you remove OTE=YES from the EZACONFG file, that will provide a circumvention.

                We will continue to investigate this to determine the real cause, but in the mean time you may like to apply this circumvention to prevent the 0C4 being issued.

                Thanks, David
                • dysonbags
                  dysonbags
                  16 Posts
                  ACCEPTED ANSWER

                  Re: CICS TS 5.1 open beta - new beta driver now available

                  ‏2012-10-17T15:28:08Z  in response to davidknibb
                  Hi David,

                  Yes, setting OTE=NO does stop the problem. I've bumped up the internal trace table size as you suggested too.

                  Cheers
                  Graham
                  • davidknibb
                    davidknibb
                    12 Posts
                    ACCEPTED ANSWER

                    Re: CICS TS 5.1 open beta - new beta driver now available

                    ‏2012-10-17T15:57:50Z  in response to dysonbags
                    Hi Graham - glad to hear that's resolved the problem for now. I'll let you know how things develop on this issue.
                    Cheers, David
                    • davidknibb
                      davidknibb
                      12 Posts
                      ACCEPTED ANSWER

                      Re: CICS TS 5.1 open beta - new beta driver now available

                      ‏2012-11-01T17:46:46Z  in response to davidknibb
                      Hi Graham,
                      This problem has been found and corrected. The fix will be included in the GA product release.
                      Thanks, David