Re: Old Linklist set use after new set activation

2013-07-25 Thread Peter Relson
The OP's scenario is likely incomplete and/or in some way inaccurate.

If you restart a "task" that has no bearing on anything. If you start a 
new address space, or start a new job in an initiator, that new work unit 
will be using the now-current LNKLST. 

Peter Relson
z/OS Core Technology Design
rel...@us.ibm.com 
 1-845-435-83908+295-8390

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN


Re: Old Linklist set use after new set activation.

2013-07-24 Thread Tom Marchant
On Wed, 24 Jul 2013 14:14:19 -0500, Howard Whitehead wrote:

>After a lead from someone on the CICS-L listserv and doing some research, it 
>appears 
>the z/OS guy may have needed to, but did not do, an LLA REFRESH after doing 
>the 
>ACTIVATE commands.

No.  LNKLST ACTIVATE updates LLA.

I suspect that the sequence of events described in your earlier post is not 
accurate.

-- 
Tom Marchant

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN


Re: Old Linklist set use after new set activation.

2013-07-24 Thread Howard Whitehead
After a lead from someone on the CICS-L listserv and doing some research, it 
appears the z/OS guy may have needed to, but did not do, an LLA REFRESH after 
doing the ACTIVATE commands.

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN


Old Linklist set use after new set activation.

2013-07-24 Thread Howard Whitehead
Hello Everyone,

We had something happen here that led me to do some looking at our
linklists.  To make a long story short, our z/OS guys had to update the
linklist and we ended up with three sets called L0 (the original after the
last IPL), N0 (the 1st new one), and N1 (the 2nd new one).

In the course of researching the first issue I found that we had 3 CICS
regions that were using set L0.  The sequence of events were...

7/7 - IPL with set L0
7/15 - An old SDFHLINK was removed from the linklist and set N0 was
ACTIVATE'd.
7/20 - The 3 subject CICS's were cycled (as they are every week) with
START=COLD.
7/21 - Our original issue was encountered so the old SDFHLINK was put back
in the linklist and set N1 was ACTIVATE'd.

So today I used "D PROG,LNKLST,USERS,NOTCURRENT" and there I find my 3
CICS regions still using set L0.

My understanding of linklist and new set activations is that once a task
is stopped and restarted, the new iteration of the task will start using
the currently active linklist set.  If this is the case, why did the new
iterations of these regions start back up using the original set L0?  What
am I missing?

These are CICS/TS 4.2 regions run as jobs under JES3 under z/OS 1.12.

Thanks,
Howard

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN