ARFSET, what tells this definition where to restart the job?

[Anon Anon]

I have CA7 and CA11 in my shop.  recently I asked for the following definition, but I don't understand how or if it will perform the function that I want.  How does this ARFset get my jobs restarted in the step that caused the exception?

 

 

LARF,ARFSET=OSRESTRT,LIST=ALL                                                 

ARFSET=OSRESTRT LIST=ALL                               DATE=15.195    PAGE 0001

                                                                               

NAME: OSRESTRT  UID: 003  CHG: 07/14/2015-07:55  DEFS:  1/ 2 RSPID: MASTER     

                                                                               

                                                                               

INTERCEPT TYPE: JC INDEX:  1                                                   

        APPLICABILITY TESTS:                                                   

        SYS EQ *         SID EQ   0  RSTC LT   5  EM EQ ANY                    

        FROM: 01/01/1975 00:01     TO: 12/31/2074 23:59                        

            TESTS RUN DURING INTERCEPT PROCESSING:                             

            STEP EQ *         PROC EQ *         PGM EQ *                       

            USER EQ ****    SYS  EQ 0***    CC   GE 0001                       

            JCL  EQ                                                            

            RESPONSE SET:                                                      

               AM,CM=H,M=ARF9999 ARF IS RESTARTING &JOBNAME IN 2 MIN.          

               AM,CM=H,M=ARF9999 OPS DO NOTHING.                               

               AW,TIME=0002                                                    

        DISP=R  CA-11=Y  BYPGDG=N  USAGE=   PROCESS=   CC=                     

        STEPS:   START=                   END=                                 

                                                                               

INTERCEPT TYPE: JC INDEX:  2                                                   

        APPLICABILITY TESTS:                                                   

        SYS EQ *         SID EQ   0  RSTC EQ   5  EM EQ ANY                    

        FROM: 01/01/1975 00:01     TO: 12/31/2074 23:59                        

            TESTS RUN DURING INTERCEPT PROCESSING:                             

            STEP EQ *    PROC EQ *         PGM EQ *                       

            USER EQ ****    SYS  EQ 0***    CC   GE 0001                       

            JCL  EQ                                                            

            RESPONSE SET:                                                      

               AM,CM=H,M=ARF9999 &JOBNAME HAS BEEN RESTARTED 5 TIMES.  SEE PR+ 

                         OSE!!!                                                

        DISP=N  CA-11=Y  BYPGDG=N  USAGE=   PROCESS=   CC=                     

        STEPS:   START=                   END=                                 

                                                                               

SLIW-00 REQUEST COMPLETED AT 11:20:13 ON 15.195                               

[Marysue]

Since you are using CA 11, the RESTART will use whatever is in the CMT for restart information (e.g., what step to restart in).

[Anon Anon]

soooooo, I could assign this particular ARFset to any job and CA-11 would retry the first step with a non-zero return code 5 times before posting to CA7 as an abnormal end? 

[Melvin Henderson]

If a job meet its ARF criteria and that job is also in the CA 11 database, the job will go through RESTART processing based on the most probable step from the point of failure.

[Anon Anon]

soooo, how can I see what is in the CMT?  I had this definition installed, activated and assigned to a job- OS9PGTRN.  the job thru a return code 40 and ARF doesn't seem to have intercepted it at all.  certainly didn't attempt a restart or send any messages to the console.

 

14.30.31 JOB13194 ---- WEDNESDAY, 15 JUL 2015 ----                             

TM00001I THRUPUT MANAGER V6 R2.0 IS ACTIVE FOR THIS JOB                        

TM00036I JOB OS9PGTRN CA/11 Schedule Enabled, Dynamic Bind and Affinity Set    

TM00004I JOB OS9PGTRN IS BEING PROCESSED BY CA7 RULES                          

14.30.36 JOB13194  ICH70001I OS@@001  LAST ACCESS AT 14:25:20 ON WEDNESDAY, JULY

14.30.36 JOB13194  $HASP373 OS9PGTRN STARTED - INIT 18   - CLASS M        - SYS

14.30.36 JOB13194  IEF403I OS9PGTRN - STARTED - TIME=14.30.36                  

14.30.37 JOB13194  U11-001 CA-11 - PARM RECEIVED FOR JOB OS9PGTRN - PARM='P'   

14.30.37 JOB13194  U11-046 U11RMS JOB OS9PGTRN WILL INTERFACE WITH DBAS SUBSYSTE

14.30.37 JOB13194  U11-010 CA-11 - 'P' PROCESSING COMPLETE FOR JOB OS9PGTRN    

14.30.37 JOB13194  U11-646 U11TRT JOB OS9PGTRN IS BEING TRACKED VIA DBAS SUBSYST

14.30.37 JOB13194  -                                                         --T

14.30.37 JOB13194  -JOBNAME  STEPNAME PROCSTEP PROGRAM   RC     EXCP   CONN    C

14.30.37 JOB13194  -OS9PGTRN CA07RMS  RMS@B0   U11RMS    0000     28      6   .0

14.30.37 JOB13194  -OS9PGTRN OS9PGTRN CLEANUP  IEFBR14   0000     16      7   .0

14.30.41 JOB13194  -OS9PGTRN JS010    ZDBMCPDU ADUUMAIN  0000    536    186   .0

14.30.41 JOB13194  -OS9PGTRN JS015    ZDBMCPDU ADUUMAIN FLUSH      0      0   .0

14.33.06 JOB13194  -OS9PGTRN JS020    NATDBPR2 IKJEFT01  0040    687    153   .0

14.33.06 JOB13194  IEF404I OS9PGTRN - ENDED - TIME=14.33.06                    

14.33.06 JOB13194  U11-626 JOBNAME=OS9PGTRN,PRODJOB#=0000,TERMINATED SUCCESSFULL

14.33.06 JOB13194  -OS9PGTRN ENDED. CPU TIME 00:00:05.04 ELASPED TIME 00:02:29.6

14.33.06 JOB13194  $HASP395 OS9PGTRN ENDED                                    

 

********************************* Top of Data **********************************

*-*-*-*-* CA07RMS  RMS@B0   RMSRPT                                             

DATES=07/15/2015.196/14:30:37    *           

CA WA RESTART OPTION REPORT 01 - (ARTS) ------ AUTOMATED RERUN AND TRACKING SYST

                                                                               

U11-001 CA-11 - PARM RECEIVED FOR JOB OS9PGTRN - PARM='P'                      

U11-046 U11RMS JOB OS9PGTRN WILL INTERFACE WITH DBAS SUBSYSTEM CAL7 RELEASE 11.0

U11-031 * PROC-STEP=OS9PGTRN STEP=CLEANUP                                      

U11-031 * PROC-STEP=JS010    STEP=ZDBMCPDU                                     

U11-070 SCRATCH/UNCATALOG PROCESSING IGNORED FOR DSN=OS.TEMP.BN.TRAIN.EXTRACT  

U11-070 SCRATCH/UNCATALOG PROCESSING IGNORED FOR DSN=OS.TEMP.BN.TRAIN.EXT.STATS

U11-031 * PROC-STEP=JS015    STEP=ZDBMCPDU                                     

U11-031 * PROC-STEP=JS020    STEP=NATDBPR2                                     

U11-070 SCRATCH/UNCATALOG PROCESSING IGNORED FOR DSN=OS.TEMP.BN.GRNTLU.TRAIN.DAT

U11-010 CA-11 - 'P' PROCESSING COMPLETE FOR JOB OS9PGTRN                       

         END OF REPORT 01  -  TOTAL ITEMS LISTED=        10                    

******************************** Bottom of Data ********************************

 

suggestions?

[Melvin Henderson]

In order to determine the status of the CMT, you would need to issue the following CA 7 top line command:  ARTS

Depending on the configuration of your CA 7/CA 11 interface setup, you may be required to supply credentials.   Once

you are logged into the product, you can enter the LSTP JOBNAME  command.  This command will reflect the last run

of the job by denoting restart step information.

You can also open a support case and someone can call you directly and review you source data.  We can also perform

a webex and directly assist in get your job setup with a arfset and provide the knowledge transfer needed to move further

with setting up arf conditions under CA 11.

 

thanks

Melvin

[Anon Anon]

howdy Melvin,

I'd be delighted to place a support call but I'm not the direct customer with the license.

 

My company contracts out the mainframe operations to a 3rd party who uses CA7/CA11 .   Said 3rd party is telling me that I cannot do what I am trying to do and so far, they have NOT placed a support call to my knowledge.  So, I'm stuck culling the community and mainframe forums trying to figure it out because I don't believe that something so logical is not possible with today's Workload Automation.    I'm going to have to tell the 3rd party how to do it before they will even believe that it can be done.

 

I have 600+ IBM z/OS batch jobs.  The restart instructions for about 10% of them is simply, 'restart in abending step up to 5 times regardless of the abending step'.  'abending' usually means anything that is not a zero return code, not necessarily a true abend.  Is there a way to set up a single ARFSET that just says 'restart in abending step up to 5 times regardless of the abending step'?  maybe someone else can suggest a different technique?    

 

I had another population that were just 'force complete' so I set up a generic ARFSET for that action, but I'd like to exploit workload automation to handle the auto restarts too.    So far I've tried a couple of things, but ARF isn't even trapping the exception at the moment.

 

Best Regards,

Donna

[Anon Anon]

any idea why this is just doing nothing at all?  should I just leave START:  completely blank?  I've tried it with wildcard '*' with no luck, so I'm trying with a bad stepname just to see if I get some sort of error message.

 

[Marysue]

look for messages in the BROWSE file (DD name BROWSE (usually) in the CA7ONL task).  This is a wrap-around file so the messages may be overwritten, but you can recreate the browse message in 7 with history reporting--SASSHIS8 08 report (see the CA 7 Reports Guide Chapter 3).

 

There are many reasons why an ARFSET would not work, but I suggest that you start with the browse messages during the timeframe that a job with this ARFSET completed and should have executed the ARF RESPONSES.

[Anon Anon]

Thank you Marysue,

If I want this definition to trigger the restart in the first step that doesn’t get a 0 return code(which CA11 should know what is the restart step), should I leave the ‘START’ parameter of the FINAL disposition blank?  I tried the wild card (*) but that didn’t seem to work.

 

Best Regards,

Donna

[Marysue]

Just use the FINAL DISP of R and CA11 ? Y and leave the rest to default.

[Marysue]

Also, be sure that the RESPONSE ID on the AR.3 screen has the authority to do the topline RESTART command