Hello again,

and many thanks to everyone that tried to help me with my IPE problem. I've
appreciated that very much. In the meantime, I've been able to track down
the cause of the IPE a little further. So, in case anybody is interested,
I'll give you a short summary of what has happend since my last mail.

The application in question was called in a loop, which means it was called
about once or twice every minute. It seems that (for reasons that don't
matter here) the application did not terminate normally, but did an abort
which left buffers allocated, named pipes open, etc. This continued for
about one or two days, and then the exception happend.

So it looks like an application terminating without properly closing pipes
may crash OS/2 (admittedly, after quite a while). This is also true for
FP22 - I installed that, but the effect stayed exactly the same. 

We are now fixing the surrounding environment so that the frequent aborts
won't happen anymore. We hope that then the IPE will disappear, too. This
has not yet been confirmed, but it looks like a reasonable explanation.

So I hope this thread can be closed (unless someone has a comment about
OS/2 and unclosed pipes, perhaps?). Thanks again to everybody who
participated,

Klaus

--- FleetStreet 1.18 NR