|
View previous topic :: View next topic
|
| Author |
Message |
ursvmg
New User
Joined: 24 May 2007
Posts: 52
Location: Chennai
|
|
|
|
Hello All,
What does the parameter RESTART=* means in the JOB card?
Thanks in advance |
|
| Back to top |
|
 |
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19243
Location: Inside the Matrix
|
|
|
|
Hello,
If you learn to use the documentation linked to from the "Manuals" link at the top of the web page, you will get much quicker response to a reference question.
RESTART=* basically says to restart at the beginning. The words from the Fine Manual we link to are:
| Quote: |
| Indicates that the system is to restart execution (1) at the beginning of or within the first job step or (2), if the first job step calls a cataloged or in-stream procedure, at the beginning of or within the first procedure step. |
|
|
| Back to top |
|
|
ursvmg
New User
Joined: 24 May 2007
Posts: 52
Location: Chennai
|
|
|
|
| Thanks dick scherrer |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19243
Location: Inside the Matrix
|
|
|
|
You're welcome  |
|
| Back to top |
|
|
dr_te_z
New User
Joined: 08 Jun 2007
Posts: 71
Location: Zoetermeer, the Netherlands
|
|
|
|
The "RESTART=" can be tough to code when you have JCL proc's. In my test JCL I bypass this by always coding like this
| Code: |
//abcd JOB (*,*),'Hello forum',NOTIFY=&SYSUID,
// CLASS=A,MSGCLASS=S,MSGLEVEL=(1,1),RESTART=OVERHER
//OVERHER EXEC PGM=IEFBR14
|
Whenever you want to (re)start you job at antoher step just move that single "OVERHER" card to that position. |
|
| Back to top |
|
|
Bitneuker
CICS Moderator
Joined: 07 Nov 2005
Posts: 1104
Location: The Netherlands at Hole 19
|
|
|
|
@dr_te_z: welkom terug  |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19243
Location: Inside the Matrix
|
|
|
|
Hello,
Does this
| Quote: |
| Whenever you want to (re)start you job at antoher step just move that single "OVERHER" card to that position. |
require that the PROC must be recataloged in order to restart in the middle? |
|
| Back to top |
|
|
dr_te_z
New User
Joined: 08 Jun 2007
Posts: 71
Location: Zoetermeer, the Netherlands
|
|
|
|
| dick scherrer wrote: |
| Whenever you want to (re)start you job at antoher step just move that single "OVERHER" card to that position. require that the PROC must be recataloged in order to restart in the middle? |
Nope, the trick is that the "dummy" iefbr14 step, prior to your real step, is executed first. It acts as a label with a simple syntax.
Otherwise you have to specify exaxtly in step in the procedure where to start |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19243
Location: Inside the Matrix
|
|
|
|
Hello,
Yup, i understand the intent, i just don't understand how to magically get the BR14 to re-position itself after some step in the proc (say a restart was needed in step 4 of a 5-step proc).
If i had a cataloged proc
| Code: |
//myproc proc
//step1 exec pgm=pgm1
//step2 exec pgm=pgm2
//step3 exec pgm=pgm3
//step4 exec pgm=pgm4
//step5 exec pgm=pgm5 |
what would be the restasrt jcl to restart in step4? |
|
| Back to top |
|
|
Bitneuker
CICS Moderator
Joined: 07 Nov 2005
Posts: 1104
Location: The Netherlands at Hole 19
|
|
|
|
Dick,
If I'm not mistaken dr_te_z will get it clear for you. His office in Zoetermeer indicates he's working with IBM. But I might be wrong  |
|
| Back to top |
|
|
ursvmg
New User
Joined: 24 May 2007
Posts: 52
Location: Chennai
|
|
|
|
Dick,
I suppose this will work for your question. In the job card, we have to mention
Please correct me if I'm wrong. |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19243
Location: Inside the Matrix
|
|
|
|
Hello Gopal,
| Quote: |
| Please correct me if I'm wrong. |
Please read the whole topic and you will see that your restart will work in general, but it does not answer the question.
The question deals with how to relocate the IEFBR14 restart (see previous info) without having to re-catalog the PROC. I'm trying to find out what was meant by
| Quote: |
Nope, the trick is that the "dummy" iefbr14 step, prior to your real step, is executed first. It acts as a label with a simple syntax.
Otherwise you have to specify exaxtly in step in the procedure where to start |
Hi George - my guess is that this approach is used for proc/pend testing rather than cataloged procs. We'll see |
|
| Back to top |
|
|
Bitneuker
CICS Moderator
Joined: 07 Nov 2005
Posts: 1104
Location: The Netherlands at Hole 19
|
|
|
|
| Well Dick, not tonight I guess; it's 21.27 right now in Holland so dr_te_z is no longer at the office (neither am I but this is a hobby since april 1) |
|
| Back to top |
|
|
Bitneuker
CICS Moderator
Joined: 07 Nov 2005
Posts: 1104
Location: The Netherlands at Hole 19
|
|
|
|
| From the manual it looks like the * goes not by itself as the examples show. But let's wait what dr_te_z has to tell |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19243
Location: Inside the Matrix
|
|
|
|
From another place in the manual, there is
| Quote: |
This example shows the changes that you might make to the JCL before you resubmit a job for step restart.
| Code: |
//J3412 JOB 386,SMITH,MSGLEVEL=1,RD=R,RESTART=*
//S1 EXEC PGM=MYPROG
//INDATA DD DSNAME=INVENT[,UNIT=TAPE],DISP=OLD,
// [VOLUME=SER=91468,]LABEL=RETPD=14
//REPORT DD SYSOUT=A
//WORK DD DSNAME=S91468,
// DISP=(,,KEEP),UNIT=SYSDA,
// SPACE=(3000,(5000,500)),
// VOLUME=(PRIVATE,RETAIN,,6)
//DDCHKPNT DD UNIT=TAPE,DISP=(MOD,PASS,CATLG),
// DSNAME=R91468,LABEL=(,NL) |
The following changes were made in the example above:
The job name has been changed (from J1234 to J3412) to distinguish the original job from the restarted job.
The RESTART parameter has been added to the JOB statement, and indicates that restart is to begin with the first job step. |
IIRC, some places i've been run all of their production with a restart so that when they really have a restart, the jcl only needs the step added - thereby reducing the chance of error (or so they believed). |
|
| Back to top |
|
|
dr_te_z
New User
Joined: 08 Jun 2007
Posts: 71
Location: Zoetermeer, the Netherlands
|
|
|
|
| Bitneuker wrote: |
| His office in Zoetermeer indicates he's working with IBM. But I might be wrong |
Yep, wrong:
A - Zoetermeer is my home adress
B - IBM closed their office in Zoetermeer several years ago (it was a SNA node)
C - I do not work for IBM
Anyway, do not expect wonders from my little tip. I was just annoyed that when you have a job with e.g. step01, step02, step03 you cannot just adjust the RESTART= parameter. When the step is a "exec pgm=idcams" it's ok, but when you have a stored procedure, it is not. I found that moving the little "restart line" around is more easy and independant of the job.
When you've got to restart in the middle of a stored-proc... then you need the oldfashioned way. |
|
| Back to top |
|
|
Bitneuker
CICS Moderator
Joined: 07 Nov 2005
Posts: 1104
Location: The Netherlands at Hole 19
|
|
|
|
| Quote: |
thereby reducing the chance of error (or so they believed).
|
I missed this one so I joined it  |
|
| Back to top |
|
|
|
|
