1) Unless enqueueing the table updates its stats, I'd rather not depend on TBSTAT. Seems to me it's as likely as anything else that the admin might ~start~ to do an update, and then get distracted and leave his workstation before making any changes.
2) They do use SDSF at this installation. Where would I look up this ENQ command? Is it one of the REXX-interface features?
3) Operator command, now, there's an idea too. We'd have to make sure the admins have the authority to issue that command, but once that question is settled I can write a routine that issues the command, captures the feedback and display the message. But what command would it be? (I was never an operator; their commands are mostly opaque to me.)
/* Cats consent to love us. Dogs beg to love us. -Cathryn Michon, Grrl Genius */
-----Original Message-----
From: TSO REXX Discussion List <
TSO-...@VM.MARIST.EDU> On Behalf Of ITschak Mugzach
Sent: Thursday, March 11, 2021 15:25
There are several easy ways to know who holds the table:
1. ISPEXEC TBSTAT will tell who last updated the table.
2. SDSF ENQ *; filter major eq SPFEDIT (minor is the table library name).
3. operator command
--- On Thu, Mar 11, 2021 at 10:13 PM Bob Bridges <
robhb...@gmail.com> wrote:
> MODIFY: This is the command that the sysprog might use to bump the
> other admin off the system? I'm going to leave that question up to IT
> management, but if ~I~ were IT management I'd have to think about that
> a while before accepting it as a long-term solution.
>
> ISGQUERY: I looked it up, but from what I found, that's an assembler macro.
> No way to call that from REXX, surely?
>
> If I don't mind reïnventing the wheel (and I don't, mostly), I could
> create a PDS and an external REXX routine that depending on the
> argument would write the operator's ACID into a member named after the
> table, remove it afterward, or display the current ACID. I wouldn't
> want the admins to notice a slowdown, though.
>
> -----Original Message-----
> From: TSO REXX Discussion List <
TSO-...@VM.MARIST.EDU> On Behalf Of Seymour J Metz
> Sent: Thursday, March 11, 2021 08:53
>
> You could modify the application to accept and parse a MODIFY command
> and authorize admins to issue the MODIFY.
>
> You could modify the application to issue ISGQUERY and put out a
> better message when TBOPEN gives a return code of 12.
>
> ________________________________________
> From: TSO REXX Discussion List [
TSO-...@VM.MARIST.EDU] on behalf of
> Bob Bridges [
robhb...@GMAIL.COM]
> Sent: Wednesday, March 10, 2021 7:09 PM
>
> The admins of a department I'm doing work for depend on a complex
> REXX/ISPF application that someone wrote for them years ago. The
> author is now departed. The app has a common problem: Sometimes one
> of the admins ties up a table, then forgets and goes to lunch or
> something and the other admins trying to do the same thing get ISPT036
> ("Table in use / TBOPEN issued for table xxxxxxx that is in use,
> ENQUEUE failed.").
>
> Usually when they run across this problem they give it 30 minutes and
> then contact a system programmer to bump the offending party. Mostly
> they hired me to work on other tasks, but they're wondering whether I
> can modify the app to let the admins do the bumping instead. Whether
> to grant that authority is of course a separate issue, not related to
> REXX. But it occurs to me they might be very pleased if I could
> intercept ISPT036 and display a panel that identifies ~who~ has the