Charles Mills wrote:
I had two thoughts in the middle of the night last night - forgive me if
these have been covered - I haven't read every reply you've gotten (I've
read most).
1. Does the problem occur if you close the files WITHOUT the error that
drives your recovery routine? If you just set a branch that "pretends" the
error occurred?
2. What is the error that drives the recovery routine? You're not out of
space for the file, are you? Is CLOSE trying to write a last block and
hitting the same space problem?
Charles
Lets see if I can cover everything...
The ACB should be unique for that task. I have not found any logic that
suggests we setup more than one ACB for the task for this file. Short
answer - this should be the only way to access this file.
Yes the TIOT offset is large. There are quite a few DD cards for the region.
As for question 1 above... I would have to say no it does not. A
different program task can start and use the files, then close them
without errors. Of course, in those instances the recovery hasn't been
triggered.
For question 2 above... The error that drives recovery is a S0C4 abend
(its an addressing problem in a module that I need to fix eventually
too). I have had other abends S33E, etc,. that have exhibited the same
problem as I am having.
I have a PMR open with IBM. I hope to have this cleared up soon. It
would appear that some other IBM customers have noted a similar
problem. It appears that I'm not alone in this.
Mark
----------------------------------------------------------------------
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
