Well guy's here's what I did:
I am in the process of investigating if it is feasible to use the BOS
server as a general mean of maintaining/monitoring/starting/stopping
applications (scheduled types and daemon-types) on all or servers and as
a result get a kind of semi-failover toolkit as all servers see the same
tree (AFS) and have access to the same apps etc. I call it the appcon
system.
It 's based on bos and requires afsd/bosserver/perl to exist on all
nodes. I deal with the admin tokens and all that by hiding PW in a
compiled/stripped binary that calls my arbitrary programs that are to be
monitored by bos. In some cases in combination with reauth as well.
Now. What I discovered during the evaluation of this system where a
appcon.check [perl] program that run on all nodes reports the status of
each node to a directory in AFS every 10 minutes, or at some point even
every minute, I saw that a specific file was not the same from an
arbitrary viewpoint. Hence my question.
We do have a crappy network at this time with alot of stupid broadcast
nonsence and I was thinking if this could in some way inhibit the update
process.
I feel that I need to go back to the lab-bench and reinvestigate since
you guy's find my question kind of hard to understand. People I have
talked to also say that is should not be any difference in results from
the two ways to generate a call back. Maybe I'm just messing up things
here.
Thanks guy's
/peo
Ken Hornstein wrote:
>
> >Is there someone out there that can sched some light over the fact
> >that echo -n "" will cause/force a callback, while perl open/close will
> >not?
>
> Um, I guess I don't understand what you mean. Could you provide an
> example?
