Shelling the Debugger (AREV Specific)
At 29 MAY 1998 03:54:45PM Victor Engel wrote:
I would like to shell the debugger in order to be able to notify an external program (probably a Windows process) when an automated process fails in such a way that it goes to the debugger. I'd appreciate it if someone can post the calling syntax for RTP25. Any caveats or special handling I should be aware of?
At 29 MAY 1998 04:33PM John Bouley wrote:
Victor, I don't know the correct syntax for RTP25 but I once used a replacement debugger called Reveal by Chris Herbert. This debugger was much more sofisticated than the standard AREV and it supported hooks.
Hope this helps.
At 04 JUN 1998 03:01PM Dan wrote:
Ask McCauley - he used to enjoy knowing this kind of thing.
At 04 JUN 1998 07:56PM Andrew P McAuley wrote:
Why not just replace it? My recollection is that RTP25 and $Debugger rely on Syscom so nothing is passed. Just do what you need to then Abort All!
World Leaders in all things RevSoft
At 16 JAN 2003 09:31AM Hippo wrote:
May I ask which variables does $DEBUGER/RTP25 use.
I am planning do similar thing as Victor does.
(Send an e-mail/SMS with as small error report as possible whenever AREV goes to debuger).
… therefore I need the debgger logic to generate the report.
Victor … what was your solution?
Thanks
At 16 JAN 2003 11:24AM Victor Engel wrote:
I don't recall why I wanted to do this, but I don't think I ever went through with it. Given the timeframe (1998), we probably had some batch jobs from HR-1 that were failing overnight causing interfaces to other systems to fail. We did implement a blat interface from the receiving end of the interface to set off our pagers. This probably worked well enough that I didn't persue the angle of intercepting the debugger.
Besides, maybe the reason the debugger was invoked was because of some network problem that would have caused blat not to work anyway, if called from HR-1. Sorry I couldn't be more helpful.
At 28 FEB 2003 07:54AM Hippo wrote:
Are these syscom variables somehow accessible through R/Basic?
I have studied R/Basic manual carefully and didn't find a note.
Does there exist some more detail documentation anywhere?
At 28 FEB 2003 01:10PM [email protected] wrote:
Syscom is not available to us mer mortals ..
[email protected] onmouseover=window.status=the new revelation technology .. a refreshing change;return(true)"
David Tod Sigafoos ~ SigSolutions
Phone: 971-570-2005
OS: Win2k sp2 (5.00.2195)
OI: 4.1.2
At 04 MAR 2003 04:28AM Hippo wrote:
Thanks for answer, I expected that.
From the discussion here, it seems to me, that knowing how to change
$program is a bit illegal, so reprogramming debugger myself is illegal?
At 04 MAR 2003 09:21AM [email protected] wrote:
There was also a dbug15(?) which was freeware. I might have a copy of this around ..
[email protected] onmouseover=window.status=the new revelation technology .. a refreshing change;return(true)"
David Tod Sigafoos ~ SigSolutions
Phone: 971-570-2005
OS: Win2k sp2 (5.00.2195)
OI: 4.1.2
At 04 MAR 2003 10:28AM Cameron Christie wrote:
I shelled our debugger on a 200-user system both to provide reporting on the number of crashes over time (you'd be amazed how often users don't report problems), but primarily for security reasons. It would take a switched on user, granted, but from the debugger you can edit the environment, reactivate TCL, and do all sorts of mischief. My shell does a check; if I don't recognise an "authorised" user I record the error and then force a logoff.
FWIW,
Cameron