RE: Causing (and detecting) a workstation failure

> From: David Morris
>
> You could send a scope message to *EXT, which should allow you an
> opportunity to cleanup. You can also register a return handler using
> ceertx. Would you be interested in an example of either of these?

Thanks, David.  I've coded CEERTX myself - it is very cool.  I haven't tried
scope messages yet, but I'd be interested in the concept.

I guess I was mistaken in thinking that you could handle this
programmatically.  I know I'm getting old and the gray matter is getting a
bit mushy, but I had thought for certain that there was a way of checking
for an error.

What I'm trying to do is figure out how to handle the situation where a
browser session times out.  Since my product supports the INFDS, I had
thought it would be as easy as setting a status code in the INFDS.  I
honestly thought that this was the way OS/400 handled it, but I haven't
coded a WS failure routine is a looooong time (back in my S/38 days).

There are plenty of status codes that indicate hardware errors - for
example, look here specifically:

http://publib.boulder.ibm.com/cgi-bin/bookmgr/BOOKS/QB3AUK00/APPENDIX1.5.8?S
HELF=&DT=19970312124213&FS=FALSE

Check out error 8197:

8197
Description: An error condition that is not recoverable was detected at the
device. An example of such an error is the user turning off the display
device.

This leads me to believe that RPG should be able to intercept this sort of
I/O error.  How did we do it before CEERTX and scope messages?

Mushy brain, mushy brain.