Interesting, I don't think you need to write that configuration file for "every"
node in your pool. Are you using NFS?? Because if so you can have a centralized
global configuration file and any variances would need to just be added to each
nodes local_config. I never got a chance to test some stuff out to day but here
is my setup if you would like to know how to do any of it i'll be glad to help
Any user to use my condor setup need only be registered on the head node. That
user can then log onto any node in the pool and their home directory will
automatically be mounted and be taken with them. To do th! is you'll need NIS and
NFS up and running. The condor user is not a user than people can log into but
rather I set up a condor group to allow people to use condor's abilities. At the
moment I have each node set up to just run what ever they can get their hands on
(i'm still learning) but will probably change this to prefer dedicated jobs in
the future. Very soon i'm going to be adding different architectures to my pool
and have their binaries stored on the head node as well.
good luck Junaid keep me post on interesting things you find.
New Mexico State University
Condor-users mailing list