On Sep 21, 2007, at 2:58 PM, Avy Wong wrote:
I am having issues backing up WFSB03. It has Journaling on d
$ . I
have been keeping close watch on the throughput but it looks like
it is
slowing down. Is there a way to find out if the session is still
alive?
The 'IdleW ' state concerns me as this is the 3 time I tried to get
this
node backed up properly.
Avy -
When posting, please supply the full info regarding a problem...
You imply that the prior two attempts failed to back up the node:
what condition or error was encountered in the presumed failures
in the prior attempts?
Consider the amount of inventory information being sent from the TSM
server to the client: The client ostensibly needs to hold that info in
memory, and if this is a modestly implemented computer, then the amount
of real memory is probably limited, with a lot already being consumed
by the OS and other apps on the machine; and if so, then virtual memory
is being increasingly used, and certainly things will slow down. A good
look at the Task Manager on that system may be revealing. This may be
a very imbalanced Windows system, with far more data than it has
realistic memory to deal with it all.
Refer back to Andy's posting from yesterday afternoon. You seem to have
a single Producer session chugging away, doing its best to collect the
Active Files inventory from the server before starting traversal of the
file system seeking backup candidates. And, with over a GB of data
already having been sent from TSM server to client, this doesn't sound
like a JBB to me.
Richard Sims at Boston University
