Running jobs in different VSE guests at exactly the same time

[tony.p...@xchanging.com]

Hi,

I need to be able to run a job in 2 z/VSE guests which are part of the same machine group at exactly the same time.

I have tried using DUETIME, but this can be a second or so out.

Anybody got any ideas how I can achieve this?

The reason I am doing this is to try and debug a very intermittent failure we have with CA-Dynam/D.

Thanks

Tony

[Steve Mondy]

Tony,
Not sophisticated but add a // PAUSE in each job then have two people do 1, 2, 3, EOB. Might get a little closer to the same time.

What is your D/D problem?

Steve

_______________________________________________
VSE-L mailing list
VS...@lists.lehigh.edu
https://lists.lehigh.edu/mailman/listinfo/vse-l

________________________________

NOTICE:
This e-mail is intended solely for the use of the individual to whom it is addressed and may contain information that is privileged, confidential or otherwise exempt from disclosure. If the reader of this e-mail is not the intended recipient or the employee or agent responsible for delivering the message to the intended recipient, you are hereby notified that any dissemination, distribution, or copying of this communication is strictly prohibited. If you have received this communication in error, please immediately notify us by replying to the original message at the listed email address. Thank You.

_______________________________________________
VSE-L mailing list
VS...@lists.lehigh.edu
https://lists.lehigh.edu/mailman/listinfo/vse-l

[daniel....@skynet.be]

Hello Tony,

How are you?

What's your problem with CA-DYNAM/D? Timing?
I don't remember, do you have a scheduling package? If not, you could do in
a first step of one of the job a PUNCH DISP=I of a JCL that will run in the
second CPU, then issue a // PAUSE in the job having done the PUNCH DISP=I
and in the second VSE issue a VSECMD to reply the PAUSE in the first VSE.
Ideally both jobs should run in static partition that will only be used by
this specific jobs around the time they are run to avoid replying to a wrong
pause.

Regards,
Daniel


Daniel de Viron
Independant consultant z/VM and z/VSE
Rue haute 42
1435 Corbais
Belgium
tel +32 474 715 713
Company number BE 0542.346.497

[Duerbusch, Tom]

When I'm trouble shooting a timing problem between VSE images, I stack the same job, 10 times, for the same partition (BG for example) on both systems.  Then let them go.

Eventually, one job in System A, will trip over the other job in System B.

If they all work, stack them all up again and repeat as necessary.

Tom Duerbusch

THD Consulting

--

 

[Avihu Gershoni]

If the VSE images have a shared lock file then you could run a third job in any one of those two VSEs doing a LOCK with CONTROL=E,SCOPE=EXT and then issue pause.

Then you release the jobs you want to check and the first step in those jobs will be will be LOCK with CONTROL=S,SCOPE=EXT to the same resource name as the third job.

They will enter a lock wait, and then you can reply to the pause of the third job, which will release your two jobs immediately to run in parallel.

Avihu Gershoni

Mainframe System Department Manager

Hilan Limited

8 Meitav Street

Tel Aviv

Israel

 

[Mick Poil]

Avihu etc.

I don't think the Lock Wait will work consistently. A VSE system with at least one task in an X-system Lock wait is not told when any external Lock is released (there is no signalling between them), so VSE polls the Lock File on a timer basis to see if the external Lock Wait(s) can be resolved now.

At least, that is how I remember the design.

Mike

[tony.p...@xchanging.com]

Hi,

Thanks for all your suggestions.

Using a lock after several attempts I was able to get the job in sync enough to get one of them to fail.

The issue was actually really simple, my 2 z/VSE guests didn’t have different systems letters assigned in CA-CIS, so there was scope for temporary CA-Dynam/D file being created with the same file-id, causing one of the jobs to fail.

Tony