Yves,
There's always more than one way to get the job done. What I suggested
earlier about using an IRB and post/wait logic was my initial response to
your query. Which was how to stop a TCB that won't respond to STATUS STOP.
I think it should work. I would encourage you to 'give it a shot'. There's
probably going to be more to consider than just scheduling the IRB, and
waiting for a post from it. The IRB should probably have an estae
protecting it, and the UTL code as well. In addition, the UTL code probably
ought to get a stimer running. Set the stimer time for some reasonable
value you think is appropriate. Add another ecb to the list for the stimer.
Use a stimer exit to post this ecb. If the stimer expires before the post
comes in from the IRB code, it means something isn't right, and you have to
take alternative action. You were smart enough to figure out how to use
STATUS, and you aren't afraid of looking at a dump. If you haven't done
this type of coding before, don't worry about it. Take it one step at a
time. Get one part working before moving on to the next. Use WTO's from
within the different pieces of code to debug the thing. Doing this you
probably will learn something. Put it on a test system, and test it till
you're absolutely sure you've covered all the bases. Then move it from a
test system to a development system where there's more work getting done.
Don't move it to a production system till it's run on development for some
time. Along the way, you may have to adjust the code some, because you
found out something you didn't know before you started.
Another approach to accomplishing what you want inside the exit might
be to use your security system. Create some resource rules that control
which jobs can keep running and which cannot. Check the rule from within
the exit. Or, create a table of jobs that are allowed to keep running.
Assemble and link the table into a library that is in the link list, such
that the exit can find the table by doing a LOAD for it. Surely your
installation has some rules set up on job naming, such that your exit could
identify a class of work some par of it's jobname. Hope this helps.
--Dave Day
--- Original Message -----
From: "Yves Colliard" <[EMAIL PROTECTED]>
Newsgroups: bit.listserv.ibm-main
To: <[email protected]>
Sent: Monday, January 15, 2007 5:03 AM
Subject: Status Stop and REXX
Hello everyone!
I'm trying within IEFUTL Exit to stop the address space (all other TCBs)
from working; because in certain circumtances I'm asking the operators
(WTOR) to decide if the address space have the right to use more CPU.
To do this I use the status stop macro and it works fine... but not with a
batch TSO Rexx programm; this one (only this TCB) is running and running.
Using a SVC Dump I've seen, that the TCB Stop is delayed.
Does someone have an idea how to even stop such a TCB?
Regards, Gruesse, salutations
Yves Colliard
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
