On Wed, Oct 20, 2010 at 11:07 AM, Pete Muir <pmuir@redhat.com> wrote:
Ondrej raised the question of whether this is a per node event (it is). If we want per cluster, we have to talk to Paul F. Which were you after?

I guess it really depends on the function of the startup routine. I would say that both are needed, but it's hard to anticipate. A startup event per node might be doing something like flushing out a temp directory, seeding a in-memory database or starting other services. A cluster-wide startup event might be populating or sorting out records in a shared database or starting up shared services. Obviously the second case you only want performed by one of the nodes (or else they are stepping on each other's toes).

-Dan

--
Dan Allen
Principal Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597

http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen