Hello everyone,
Does anyone here have any experience/exposure to RLS exit XFCRLSCO.?

We run in a CICSPLEX (two TORs, six AORs and a rolling FOR) across two
LPARs.
We have been using the RLS exit without incident for a few years now. So
long that I don't even think of it any more.
We use RLS for our VSAM CICS resources. We do not run anything as batch RLS.
We do not use Transactional VSAM.

Anyway, we are going down the path to establish replication of our VSAM
file changes to Kafka topics in a distributed tier.

We designed/developed and implemented our pilot into our development
environment and everything ran smoothly. Really all we did was run an ALTER
of the VSAM files to enable replication logging to a Coupling Facility
defined LOGSTREAM.
We deployed to our production environment, and after a few hours we ran
into an issue indicating there were duplicate writes attempted to the
logstream. So we backed everything out.

That was about three weeks ago. Since that time we have been unable to
identify any root cause. The dev environment is very similar but not
identical to production.
We run identical software, same program loads etc. But these are different
SYSPLEXes and different CICSPLEXes, and different CECs etc.

Now we are looking at XFCRLSCO to see if that could be a concern.
>From what we can see in the JES Logs, all of our CICS regions had quiesced
the specific VSAM file prior to the Batch JOB attempt to write to the file.
The batch JOB failed. Return Code in hex 00000024; Reason Code in hex
73050091
"A logging error occurred while VSAM was disconnecting from the CICS VR
address space."

Does anyone have any suggestions as to how I could determine if this error
relates to leveraging the RLS exit XFCRLSCO?

I can't see any issue with the application. Same code in Development and
test.
The same RLS exit runs in both environments.
We have the same CICS TS levels in both environments. CICS TS 5.2 with the
latest in PTFs.
We are running CICSVR 5.2 with the latest PTFs in both environments.
We are running z/OS 2.2 in both environments.

XFCRLSCO could be a red herring here. I can't see how we would be able to
OPEN a VSAM to Batch if CICS already had it OPEN for write, and I can't see
how we could OPEN the file to CICS if a batch JOB already had it OPEN for
write.
The batch JOB failed, so I am assuming that somehow CICS was writing to the
VSAM file, even though, the batch JOB successfully Opened the file for
output.

I am feeling a bit lost and not sure where to continue to the search.
XFCRLSCO? CICS?Batch JOB? CICSVR?

Thanks in advance for any suggestions.

.......Cameron

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to