I did some tests backing up/restoring a 4GB file with an AIX TSM client
and TSM server and found the following:
Backup to diskpool unencrypted: 400 seconds
Backup to diskpool encrypted: 600 seconds
Restore unencrypted file from tape: 400 seconds
Restore encrypted file from tape: 600 seconds.
IBM
Just to be sure I am not missing something obvious:
With server-to-server configured, is it true that you can only have one
diskpool or tapepool on the remote server as the destination for the
virtual volumes?
I have several diskpools on my source server that get backed up to local
copypools an
source server.
Hope this helps.
Joerg Pohlmann
604-535-0452
"Pugliese, Edward" <[EMAIL PROTECTED]> Sent by: "ADSM: Dist Stor
Manager" 2005-06-30 15:13 Please respond to
"ADSM: Dist Stor Manager"
To
ADSM-L@VM.MARIST.EDU
cc
Subject
Virtual volumes and
You could use server to server communication and then the remote TSM
server is the one with the library.
Local server sends virtual volumes to remote server.
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
bob molerio
Sent: Wednesday, July 06, 2005
The faults heading in vmstat is not paging. It refers to
trap/interrupts. The sy column refers to system calls.
(man vmstat)
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
PAC Brion Arnaud
Sent: Wednesday, July 20, 2005 11:06 AM
To: ADSM-L@VM.
Instead of multiple versions on the same server, could you run different
versions on separate LPARS in a library client/manager arrangement ? I
do run multiple servers on one AIX partition but they are on the same
level.
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PRO
To tape:
backup db devclass=tape_class type=full scratch=yes wait=yes
To disk:
you could set up a device class file_devclass where type=file
then run the same backup db with the new devclass
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On
I have a local primary storage pool that gets backed up to a copy
storage pool that gets sent off site. In addition, I have started
backing up the primary storage pool to virtual volumes at our DR
location. This is working fine and I was able to backup all previous
tapes in the primary pool to
I am having a problem deleting some old volhist entries where the target
was a device class of file.
"q volhist type=dbb" output shows 3 old entries from when I was playing
around with "File Device Class" and did a few "backup db
devclass=FILE_DEVCLASS type=full" commands.
Here is a sample volhis
Thanks for the info. I will have to keep an eye out for that situation
the next time I upgrade. The Backup Series numbers (405, 409, 410) are
lower than the lowest number of the current backups (574 to 593). So
that doesn't match your problem.
I guess I should also mention this is an AIX server
You da man :-)
Exactly 3 entries deleted that I didn't want. Seems so obvious that I
should have tried that. But that's easy to say in hindsight.
Thanks,
Ed
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Mark D. Rodriguez
Sent: Thursday, August
I can't begin to tell you the amount of times a network support person
has told me there were no problems on their side (even when evidence
points to an external problem) only to come back later and prove it was
on their side.
As far as auto negotiate, I have always had better reliability setting
Are you looking for the "Cache migrated files" option in the disk pool
settings ?
This will allow you to restore files that are still in the disk pool
even after they have been migrated to tape (as long as there is room in
the disk pool to store the data).
Also, if you set migration delay to 1
Out of curiosity, what is the reason for this requirement ?
TSM and an appropriate configured library can be pretty fast at
restoring files even if the last active version was backed up a long
time ago.
What if you sent the data you wanted to keep in the diskpool to a
separate diskpool from other
As a precaution I run a daily report that does "backup stg tapepool
copypool preview=volume wait=yes" for each of my storage pools. This
way I will know if anything snuck out of the diskpool to the tapepool
prior to backing it up to a copypool.
Just wanted to pass that on in case someone didn't
How about:
backup stg tapepool copypool preview=volume wait=yes
This will give you a list of tapes that are needed to get the tapepool
and copypool in sync.
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
William Boyer
Sent: Friday, Septe
You could just leave the old node defined exactly the way it was. The
archives will last however long the definition states. When the new
node needs to access data from the old node you can do:
newnode# dsmc retrieve -virtualnode=oldnodename
To transfer the information I wil
Do a vmstat while the TSM server is starting and look for PI/PO. If
present you are paging and that is probably the cause of the disk
activity on rootvg. You can also run a filemon during the startup to
more accurately identify the active LV and most active files.
-Original Message-
F
We have a similar set-up with FAStT storage in two locations that are
not mirrored. Although we use the IBM 3583 tape library. This is how
we approached the problem of not extending the SAN to the DR location:
TSM in location A is created with library client and library manager
instance.
DR site
"backup stg primary_pool copy_pool preview=vol wait=yes"
This will tell you if any primary_pool volumes are needed to get the
copy_pool in sync. Just run without the "preview=vol" to actually
create the additional copy_pool volumes.
-Original Message-
From: ADSM: Dist Stor Manager [m
I have never tried it for the include/exclude file, but other TSM client
files like dsm.sys use the asterisk for a comment.
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Thomas Denier
Sent: Monday, January 30, 2006 11:14 AM
To: ADSM-L@VM.MARIST.ED
Have you eliminated network throughput by testing an FTP of a large file
and see if that gives you significantly better throughput than TSM
client throughput ?
Do you have too many clients connecting at the same time ? Have you
tried a single client backup with no other activity ?
Just a few t
Have you considered sending the data to a virtual volume copypool at you
local TSM server and then sending those tapes offsite ? That way you
could send data daily and have daily offsite copies.
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Raje
You can still create DRM tapes. Just do the "backup stg diskpool copypool"
prior to the migrate of diskpool to tapepool so the data is still in the
diskpool. This way you are always copying from disk to tape. No need for tape
to tape copy. This is applicable to all environments, not just
If you can fund a backup system, you can configure DRM to automate
things. You could even use the DR location for offsite copypool storage
using server-to-server communications and virtual volumes.
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Ma
"backup stg primary_pool copy_pool preview=vol"
This will show you what tapes are needed if they are not in sync.
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Tim Brown
Sent: Thursday, August 03, 2006 1:19 PM
To: ADSM-L@VM.MARIST.EDU
Subject: [
26 matches
Mail list logo
