No, I meant restore.  These are scheduled restores.

Shawn Drew wrote:

If you want to make sure the schedules don't interfere with each other,
you should probably make 2 node names and 2 schedule services so there is
no conflict.

Also, you meant "backup" instead of "restore", correct?  It was a little
confusion reading this if that isn't the case.  Assuming so... This
doesn't look like standard behavior.  Did this happen just the once? or
something repetitive? Also, when you said "the second restore is
scheduled"  did that mean the scheduler was restarted again after the
first backup?

 It look like it is paying attention to the schedule duration, but
ignoring the start time.  Looks like it is using the current time as the
start time while the schedule is in a "Pending" state.  It then chose a
point 2 hours from the current time to choose the random time.

I'd normally say to upgrade, but I suppose that's not an option.

Regards,
Shawn





Internet
sjharris AT AU1.IBM DOT COM

Sent by: ADSM-L AT VM.MARIST DOT EDU
05/22/2008 09:21 PM
Please respond to
ADSM-L AT VM.MARIST DOT EDU


To
ADSM-L
cc

Subject
[ADSM-L] Schedmode Polling behaviour





I've seen some behaviour with schedmode polling that I don't understand
and
wonder if anyof you can shed some light on it.

One of my clients schedules a restore every day of 1 file from each of two
nodes to the TSM Server.  This is something to do with SOX compliance and
they are fanatical about it.

Server is 5.1 something on windows2000, client is 5.1.0.0  -  Yes, I know
this went out of support in 2005, and if I could walk away from this I
certainly would.

Originally the first schedule was set to run at 14:15 in a five minute
window, and the second at 14:40 in a five minute window.  Most of the
other
data on this serve is SAP backups and so the files are large. The tape
library is as old as the software with two LTO1 drives, and as a result if
housekeeping was running when the first schedule ran, it woulkd wait for a
tape and the second schedule would miss,

Ok.  So after this happened again recently, I increased the window to 4
hours for each restore.  Now there was a panic because the restores were
pending.

The randomize percentage for this server is 50% and the client in question
is in polling mode. so the first restore should start somewhere between
14:15 and 16:15

Looking at the schedule log we see this is so, with the first schedule set
to start at  15:34.  Because of the panic about the schedules being in
pending status, the scheduler service was restarted and after the restart
(at 15:13) we see a randomized start time of 16:23.  This is outside the
time frame that I expect.  Now, the first restore runs at 16:23 for about
90 seconds and the second restore is scheduled.  I would expect it to run
almost straight away as it should run sometime between 14:40 and 16:40,
but
it is not scheduled until 17:18.

Can anyone explain this behaviour? I'm not looking for a fix, just an
explanation as to how the schedule can be set to run outside the first
half
of the window as this does not gel with the explanation in the manuals.

Regards

Steve
Steven Harris
TSM Admin, Sydney Australia


This message and any attachments (the "message") is intended solely for
the addressees and is confidential. If you receive this message in error,
please delete it and immediately notify the sender. Any use not in accord
with its purpose, any dissemination or disclosure, either whole or partial,
is prohibited except formal approval. The internet can not guarantee the
integrity of this message. BNP PARIBAS (and its subsidiaries) shall (will)
not therefore be liable for the message if modified. Please note that certain
functions and services for BNP Paribas may be performed by BNP Paribas RCC, Inc.