On Fri, Dec 19, 2008 at 1:17 AM, Roger Bolan <[email protected]> wrote:
>
> I know I'm catching up with an old thread, but I think this approach is REAL
> SCARY.  If I were a user with anything important on a temp disk and I lost
> it because you forced a CP LOGOFF on me, I would be very upset.   At the
> very least, execute a logoff exec that will check for temp disks and do a
> disconnect instead if there are any.
> --Roger

But that was the question in the post. I assume it is for some
specific purpose. You would not do that in a general user program. And
if the OP actually meant to just terminate the REXX program, my code
was inobscure enough to see where to do that.

And many installations (including yours, IIRC) *do* have the policy to
force idle users off the system after some time. Whether to save
resources (always claimed that it would be more effective to force the
busy users) or for security reasons or to simplify software
management.
One installation I worked frequently purged spool files to emphasize
the fact that spool was not being backed up. Users learn (the hard
way) that their environment is volatile. The number of time that I
actually was forced off while typing the next command is statistically
hard to explain...
Or worse: we had the policy that users with no logon over some time
were actually completely removed (including all backups etc). One chap
was out on assignment for 6 months after he worked on his thesis for a
few years...

When SLA does not match the requirements, you can waste a lot of each
others' time in the cat & mouse game. It would be too simple to accept
owning a temporary disk as an excuse not to be forced off. Users come
up with programs to consume weeny bit of resources to appear idle,
service providers enhance their code to check in CMS NUCON for that
program, smart users rename it to something else, etc.

-Rob

Reply via email to