[Networker] save sessions slow to show up?

Hi,

Sometimes, when running a given group, I'll notice that at some point,
only a few save sessions are running in the nwadmin window, but if I go
and check the group details, I see that there are still a bunch of
sessions pending. Other times, there are no sessions running in the
admin window and the group details still shows pending sessions. It
seems odd that NetWorker would not launch these as soon as it has
available resources. I can understand if a given client had maxed out on
its client parallelism, or if the server's parallelism would be exceeded
if it tried to launch any more, but neither of these is the case.
Furthermore, there are plenty of available tapes and the clients are
reachable. And worst of all, the tapes that are still mounted, from the
last sessions that completed (same group), still have plenty of space.
I've observed one of two scenarios:

1. NetWorker just never seems to launch the other sessions. Even after
like 30 minutes, the nwadmin window is still ice cold. But, when I come
in the next morning, the group is done. Obviously, at some point over
the night, it decided to launch them and they completed, but who knows
how long it sat before finally starting them. I guess I could check the
logs, but this will only show me when the saveset completed and not how
long after the group started that NetWorker finally decided to get its
act together and start saving those savesets. What is causing it to sit
dormant so long? You'd think it would kick in as soon as a free session
opened up!

2. NetWorker does start some of the pending sessions but not right away.
Maybe after like 5 minutes of just sitting there doing nothing,
sometimes longer. In the meantime, the sessions window only shows a few
sessions running, so you'd think there would be no reason to preclude
NetWorker from getting on with it, but no. It just does nothing for a
while, and then finally you see some of the pending sessions begin to
start. Once again, it's not like the pending sessions are from clients
who are already maxed out on their parellism. In many cases the affected
client had not previously backed up any sessions for the given group on
this day.

Any thoughts? We are running 6.1.1 under Solaris, and we're seeing this
on our storage node server which is running 6.1.1 under Linux Redhat.
Maybe 6.1.3 fixes this?

I guess I could just sit and diligently watch it all night. Gee, kinda
reminds me of that story where the shoemaker and his wife finally decide
to stay up all night and hide and see who really is repairing all the
shoes! Turns out, it's little elves that come out at night to do the
work. Hmmm...

Thanks.

George Sinclair
George.Sinclair AT noaa DOT gov

--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=