On Mon, 6 Nov 2006 12:33:13 -0500, Gilbert Saint-Flour wrote:
>
> A task doesn't own a DCB per se, but DEBs created by OPEN are placed on a
> chain anchored in the TCBDEB field of the TCB under which OPEN is invoked.
>
> When a task terminates, the Task Close routine (IFGxTCx IIRC) runs the DEB
> chain and tries to close corresponding DCBs and ACBs. If a DCB/ACB is
> inaccessible or appears corrupted, then TC issues IDC999I and S0C3.
>
> I am not aware of SMP (or IEBCOPY) doing anything that would prevent TC from
> doing its job.
>
Just an update. IBM is pursuing this in PMRecord 56321,033,000:
+IBM -5695DF175 - 06/12/22-12:10-
------------------------------------------------------------------------
Tracing through the code I have determine that access methods issued an
IEFSSREQ to unallocate the SYSPRINT ACB and it failed with RC=4.
Looking into reason why this occurred. ...
So, yes, something I like about MVS is the enterprise-class service given
customers. It didn't take them all this time -- a spent a few weeks
intermittently pondering and pruning my test case.
-- gil
--
StorageTek
INFORMATION made POWERFUL
----------------------------------------------------------------------
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
