>> If you can interrogate the server from another process you
>> may also want to see if the MON$ tables can tell you anything
>> about what the frozen process is/was doing when it died.
> Actually this is an area I must start looking at ;)
> Obviously they want the system back up ASAP, so doing much is
> a little problematic. But what would be the best place to
> start? Is there a crib sheet for the MON$ tables, or do I
> have to untangle the release notes ;)

All I've done with it so far is read the release notes so I
cant help... I have been planning on doing more ;-)

> I have exactly the same version of the IBO code working on
> several sites and has been for many years ... it is only this
> year that this problem has started, and the only thing
> different IS the configuration of windows. [...]

Saying the "only thing different" is a pretty big ask over
a period of years... but you know these sites I don't. On
the other hand it would certainly not be the first time that
Windows broke something with an update - or Linux for that
matter (Ubuntu did it to me more than once).

I remember there was one particular set of updates (SP1?) for
XP that broke some local loopback communications - but did not
cause lockup as far as I ever knew.

I'd certainly be looking at all security software (firewall,
AV, HIPS etc) too. For example I have found my system will
almost stop for quite long periods of time when Comodo checks
for virus signature updates (but it does come back again).
Back when I used ZoneAlarm it broke my system a few times with
it's updates. Related to this would be checking any secure
protocols you are using (ipsec/vpn etc) to see what could be
different (or in conflict with recent Windows updates) there.


Pretty much out of ideas otherwise.

Good luck with it.

--
Geoff Worboys
Telesis Computing