Is there a definitive recommendation for the best way to lay out TSM DB files
on ESS disk? The ESS is raid 5; I have no control over placement. Is it best
to do one or two large db "volumes" or many smaller db "volumes". My db is
current 59 gig and I will be increasing it to about 100 gig.
D
The install doc that I use is found here:
http://publib.boulder.ibm.com/tividd/td/SMAIXN/GC32-0770-02/en_US/PDF/anraqs522.pdf
>>> [EMAIL PROTECTED] 05/04/05 6:42 AM >>>
Hi experts,
Does any one have nice document relating to installation of TMS on Aix and
Linux having all details related to T
At TSM 5.2.3.5 running on AIX 5.1 with a 3494 library with 3590 drives, my
devconfig.dsm file was serial numbers and WWNs in the DEFINE DRIVE statements.
How will I need to edit devconfig.dsm at a DR Test with different physical
drives with different serials and WWNs?
David
>I have a volume that appears to empty, but I can't delete it, or reclaim it.
>Any ideas?
>
Do a
audit vol XX fix=yes
I got to thinking about 3494 category codes in the middle of the night when I
should have been sleeping (can you tell my DR test is soon?).
If I load the DR 3494 with my copypool tapes and let the 3494 do an audit
operation while I am restoring my tsm machine from a sysback tape, will all the
>>You can empty your log with a db backup (incremental or full).
Unless, as others have noted, you have pinned data.
>>Why is TSM attempting to use this tape for reclamation when it has been
labeled as "vault" and is "offsite"?<<
Its not. TSM is trying to using the primary pool version of the data to
recreated the copy pool version of the data and is unable to do so because it
is not for some reason able to r
You need to cd to /opt/tivoli/tsm/server/bin before starting dsmserv. It
expects to find dsmserv.dsk in the current directory.
>>> [EMAIL PROTECTED] 05/18/05 12:40 PM >>>
Please find the details asked for.
Definition of $PATH env variable is
bash:
/usr/kerberos/sbin:/usr/kerberos/bin:/usr/loc
>There is no useful data on the tape - it is gone from the remote TSM
>instance. I could
>check it in as private, but then what? It ultimately needs to be scratch.
>
Check it in as private, then 'del vol xx discarddata=yes' to get rid of the
data obsolete data references.
David Ehresman
In TSM for Databases, each backup file has a unique name and rman is
responsible for "deleting" them, thus triggering the TSM expiration cycle.
Does Backtrack work the same way?
>>> [EMAIL PROTECTED] 05/19/05 2:34 PM >>>
Hi:
We've just discovered q peculiar problem with expiration involing the
Have you tried 'dsmc -virtualnode= . . .'?
>>> [EMAIL PROTECTED] 05/20/05 11:02 AM >>>
We have two clients on NT boxes with oracle databses. (One TSM install).
One is the flat file backup, the 2nd with its own dsm.opt file is for
SQl-BACKTRACK.
We cannot query the backups on SQL-BACTRACK since the
Then, since TSM see them all as still active, I'd guess that Backtrack is not
expiring the files.
David
>>> [EMAIL PROTECTED] 05/20/05 11:05 AM >>>
Yes, SQL-BACKTRACK is responsible for updating the status of the
uniquely name file.
>>> [EMAIL PROTECTED] 05/20/2005 10:43:53 AM >>>
In TSM for Da
Any clues in the dsmerror.log?
>>> [EMAIL PROTECTED] 05/20/05 11:41 AM >>>
Hello *SM'ers,
Has anyone observed the following;
We are experienceing performance problems with LanFree backups. The
issue goes away with a retsart of the storage agent, but at some point
in the future comes back, requir
Issue the restore from the system you want to restore to. Add the option
"-virtualnode=OTHERSYSTEM"
to your dsmc command where OTHERSYSTEM is the one on which the data was backed
up. Backup and restore systems do need to be the same OS type.
David
>>> [EMAIL PROTECTED] 05/24/05 5:54 PM >>>
I
>>We walk this way.
Backup Storage Pools (Criticals)
Backup TSM Db
Backup Storage Pools (non-Criticals)
Expiration
Move drmedia
Run Drm Prepare
DR preparation and mksysb on TSM server to tape
Take tapes to Data Vault.<<
I would highly recommend moving your DRM PREPARE to immediately after your DB
It sounds to me that you've run into some unusual situation which will be best
addressed by TSM for Sysback support.
David
>>> [EMAIL PROTECTED] 05/24/05 1:48 AM >>>
Goran,
I've first tested it a couple of months ago and it all went fine. Now I'm
setting it up for production and the tsmdev is m
My daily process flow is centered around the need to be able to restore in a DR
situation.
So
1) clients backup data to TSM disk and tape primary storage pools
2) Backup stg for disk storage pools to copy pool tape
3) backu stg for tape storage pools to copy pool tape
(we backup from disk first
g to wait for disk
and then do tape, I try to get the critical (DR) done and the SLA for the
noncritical (in a DR situation) would be 1 day old vs the up to the morning
(after 13:30 pick up) on the critical DR.
Thank You,
Bill Rosette
Data Center/IS/Papa Johns International
WWJD
/25/05 1:21 PM >>>
How many tape drives??
Thank You,
Bill Rosette
Data Center/IS/Papa Johns International
WWJD
David E Ehresman
<[EMAIL PROTECTED]To: ADSM-L@VM.MARIST.EDU
LLE.EDU> cc:
RUn the q_columns tsm script, "RUN Q_COLUMNS volumes" to see all the column
names for the volumes table. run q_tables gives you all the table names.
David
>>> [EMAIL PROTECTED] 05/25/05 3:07 PM >>>
Howdy folks,
Does anyone know what are column names and table name(s) for WWN and Serial
Number
Try:
select * from backups where node_name='XXX' and ll_name='file.name'
>>> [EMAIL PROTECTED] 10/09/02 11:17AM >>>
I need to find a specific file on tape, and the user does not know
where
the file was, nor when it disappeared - so basically I have to look at
all
the backups on a specific node t
Is there any way to tell when a backup session that would normally go to
disk is instead forced to tape because there is not enough room left in
the disk storage pool? Are there any server side (or client side)
messages that identify this condition?
David
I just had Tivoli level 1 support tell me that comm wait time and media
wait time were overlapping. That media wait time was also counted in
comm wait time. Can anyone confirm that is indeed correct? Or that it
is incorrect?
>If you look through your activity log you will see when the disk
storage
>pool fills up it will automatically get migrated to tape based upon
the
>low and high water migration values.
>
I'm not asking about disk storage migration but the case when a node
backup goes directly to tape because it i
On the server, you have to define the path between the library client
and the drives.
What is the show command to see who has the log pinned? Is it in AIX
server 5.1.1.6 level?
David
>Does anyone know of a way to track the number of active sessions
>overnight.
>
Servergraph/TSM graphs this number. See http://www.servergraph.com .
David Ehresman
A satisfied customer
I have backups successfully going lan free but on archives in the same
domain I get:
Total number of bytes transferred: 517.13 MB
LanFree data bytes: 0 B
Is there some trick to getting archvies, as opposed to backups, to go
lan free?
David
>I would bet you backup copygroup(s) points to SAN-shared tape while
>archive copygroup(s) points to diskpool (which is not shareable).
>
Well when I first starting looking into this you would have been right.
But I fixed that last week. Now my select shows:
tsm: ULTSM>select class_name,set_nam
Anyone know what happens when you have processes/sessions running and
using tape drives and vary the drives offline with the command:
update drive LIB DRIVE online=no
Do the tapes get dismounted? Processes/sessions terminate? Data
corrupted?
3590 drives in a 3494 library running TSM 5.1.1.6 o
So what are the urls of the week for downloading patches and fixes?
>I can't seem to force the client to change its password. I
>issue the command "upd node client forcepwreset=yes" expecting it to
>force
>the client to change its password the next time it logs into TSM.
I think this only forces a password change by the client AFTER an admin
resets the password.
>What I am trying to do is force certain clients to change its password
every
>60 days or so
update node x passexp=60
David Ehresman
I'm getting "ANR1229W Volume 700270 cannot be backed up - volume is
offline or access mode is "unavailable" or "destroyed" " messages for a
volume. A "q libvol 3494 700270" reports the volume is in the library.
A "q vol 700270 f=d" reports access=read/write and status=filling. The
tape is not a
>Are any of your drives offline? Do you have any drives available?
What
>happens when you do an audit vol 700270 fix=yes?
>
No dirves offline. Lots of drives available. The audit vol fails with
a "ANR2321W Audit volume process terminated for volume 700270 - storage
media inaccessible "
David
>1. check the tape out of the library with remove=no option.
>2. Open up the library and put the tape in the recovery slot.
>3. Check the tape back into the library.
>
I had already tried a checkout remove=yes followed by a checkin and
that had not helped. I tried your way but I am still getting
>I have run into a problem with SAN clients where this can happen.
Looks
>like it takes a bounce of the server to recover.
Paul,
You have to bounce the StorageAgent or the TMS server? Is there an
open apar on this?
David
TSM 4.2 has a Trace Facility Guide manual. Is there a version of this
manual for TSM 5? If so, where might it be found?
select stgpool_name,count(*) from volumes group by stgpool_name
Are these NT/W2K clients? What level of TSM did you upgrade from? We
are seeing a similar problem at tsm 5.1.1.0 on AIX 5.1.0.0 with both
server and aix in 64 bit mode. I have an open issue with Tivoli (PMR
68765,082) and I'm waiting on a client to get back from vacation so we
can run a trace d
Did you go to 64 bit mode when you went to 5.1.1.6? We have some
clients that start out backing up at a good rate and then shift into a
mode where they are running very very slow. We now think this might be
a 64 bit mode server problem.
David
>>> [EMAIL PROTECTED] 11/15/02 09:20AM >>>
We were r
>Or, what other ideas have been used to verify backups in TSM?
We are a Servergraph/TSM customer. Their latest version breaks out
backups by completed, completed with missed files, completed with missed
file spaces, and not completed. It is a web based tool, so you can
drill down to see the file
One of the very nice features of Servergraph/TSM is that it queries the
filespaces daily and sends email notifications when a filespace is not
successfully backed up. It does not rely on a schedule event successful
message to determine whether the backup did in fact complete.
David Ehresman
A sat
We migrated our TSM server off of OS390 about 6 months ago. But when we
were on OS390 we had that problem and had to go to maintenance level
4.2.1.9 to fix it. A work around seem to be to cycle the TSM server
every day.
David
>>> [EMAIL PROTECTED] 12/04/02 09:41PM >>>
Hello All
Our TSM server
http://www.tivoli.com/support/public/Prodman/public_manuals/td/TSMC/SH26-4121-01/en_US/PDF/SH26-4121-01.pdf
documents the trace options.
>>> [EMAIL PROTECTED] 12/06/02 10:58AM >>>
Hi all,
I am having difficulty finding the various trace options listed
anywhere for
W2K backup clients. I especial
>Is there a reason not to set maxscratch high say for all storage
pools,
One reason not to set maxscratch high is that if you set it
realistically (based on the number of tapes you have minus a cushion),
you can use the storage pool percent utilized figures to let you know
when you are going
There is an AIX 5.1 64 bit mode tcpip bug you will want the fix for. It
causes slow TSM backups. It is currently available as an efix but is
suppose to be officially released next week. The apar number is
IY36925. It goes on top of Aix 5.1 ML03. The work around is to run AIX
is 32 bit mode.
The tape drives have to be available via SAN to the Storage Agent
machine. The data to be backed up up does not.
David
>>> [EMAIL PROTECTED] 12/13/02 10:56AM >>>
Hi,
I am confused about Storage Agent use. To use it with SAN, do we have
to keep the data to be backuped on SAN storage or can it be
>Backupsets are the optimal solution
In TSM v5, I've replaced my backupsets with MOVE NODE DATA as my
preferred way of consolidating a node's backup data onto a minimum
number of tapes.
David
select drive_name,drive_state,allocated_to,online from
drives
>>> [EMAIL PROTECTED] 12/19/02 07:34AM >>>
Hi list,
Does anybody know of a SELECT statement that will show if any drives
are
available for use (not in use by another process or session)?
I need this for scripting purposes as I have
My allocated-to field contains either the storage agent name or the tsm
server name when there is a tape mounted in the drive.
David
I get server names in the allocated-to field with tsm server 5.1.1.6.
David
Can anyone tell me whether a incremental backup of a virtualmountpoint
sets the last backup date/time stamp?
David
>Is there a chance to reconstruct the data ??
It won't be easy. But you can backup your current data base, restore
the data base to an earlier backup that has the archive files. Export
the archive files. Restore your current data base from backup. Import
the exported archive files.
> Since then I've been
>able to put an SQL query together to get that info but it takes quite
a
>while to execute.
Can you share that code with us?
David
Is a license for Managed Library required for a 3494 library? Our q lic
reports:
Number of Managed Libraries in use: 0
Number of Managed Libraries licensed: 0
even tho we are actively using a 3494. Where is this documented?
Also, what does a Shared Library license cover? We are ap
We are running 5 lan free storage agents. In the TSM v5 Admin Ref, I
find "mgsyssan.lic
To license Managed System for SAN. Tivoli Storage Manager requires
this license for each storage agent. The Tape Library Sharing feature
is
required on the Tivoli Storage Manager server." I read this to mean I
What is the best way to research whether a given tsm problem has been
reported to IBM and whether a fix has been released or is being worked
on? Is there a web page to do this kind of search on or is the best
route to call the support center?
David
We have a Netware cluster running Netware 6, SP2 and Cluster Services
v1.6. They are backing up to TSM using the 5.1.0.0 client in prompted
mode. When the server contacts the client at the scheduled time, we
sometimes get a server abend which Novell says is caused by the TSM
client trying to acc
Now that the Tivoli pages are gone, does anyone know where I can find
the TSM Messages manual (TSM 5.1). It is not at
http://publib.boulder.ibm.com/tividd/td/StorageManagerforAIX5.1.html
where I would expect to find it.
David
Can I migrate directly from TSM server AIX 5.1.1.6 to 5.1.6.0 without
loading the 5.1.5.0 code?
>10 goto google.com
Unfortunately google points to the tivoli page for the Performance
Tuning Guide.
David
>So, I was wondering:
>
>How do you folks deal with the desire to automatically answer the question
>
>"Are my copy pools up to date"
>
>?
>
>
I email the output from the following query run at the end of my backup stgpool
cycle:
select node_name as "Node Name ",stgpool_name as "STGPool ", -
cas
Debbie,
We use Servergraph/TSM ro record info like this as well as send out alerts to
contact persons for failed back ups and a host of other reporting and alerting
purposes. Details at http://www.servergraph.com. I highly recommend you take
a look at it.
David Ehresman
University of Louisvi
When my desktop hard drive died a couple of weeks ago, I had the chance to
"practice" a XP BareMetalRestore. I also had the problem that the XP install
CD did not have a driver for my network card. So I installed XP from cd,
installed the drivers for the network card, then ran the ASR scripts
> Do you know what type of database this is?
TSM uses a proprietary database with a SQL subset interface.
David
I use a script that starts enough backup stg to use all my drives. Then it
checks periodically to see how many backups are still running and when enough
backups have completed, start the next one, then wait again, until all are
done. If I were rewriting it today, I would count and wait on free
Karin,
Did you set up your tsm install cd EXACTLY as described in the documenation?
>>> [EMAIL PROTECTED] 06/28/05 5:36 AM >>>
Hi all,
after solving the problem with the network drivers, thanks a lot!, we have
now the problem, that automatic installation of the tsm client is failing.
tsmasr.cmd
This fix appears to be already included in AIX 5.2 ML6?
>>> [EMAIL PROTECTED] 06/27/05 3:12 PM >>>
A big FYI for AIX 5.x sites...
I spent most of the weekend pursuing a problem in our TSM system
where HSM operations could not proceed. The whole AIX 5.2 system got
so screwed up in the kernel that
I don't know about your library, but on an IBM 3584 library, a tsm 'audit
library' command will just reconcile its libvol data with the data the library
maintains. It does not move any parts in the library because the library
already has a record of which tapes are where.
David
>>> [EMAIL PRO
Check the dsm.sys file. Maybe its the errorlogname value that is wrong?
David
>>> [EMAIL PROTECTED] 07/08/05 2:45 PM >>>
We have been trying to get the TDP for Domino on Solaris, running. We
finally got past most hurdles except for this one. We get this every time
we try to use DOMDSMC.
All of t
Do a "smitty sysback" -> TSM Configuration -> TSM Backup Management -> Remove a
TSM Backup Image -> select your TSM virtual device . You'll get a list of
current backups. Check to see if any are still open. If so, select it ((Esc
7) and then press Enter to remove it.
Then retry your backup.
The Linux 5.3 client README says:
"- The TSM client does not support the default locale for Linux, en_US.UTF-8.
There may be some files that cannot be backed up with the default locale,
causing error messages in the dsmerror log and the backup operation
to stop. To avoid this problem, set the
No, TSM for Sysback is at 5.6.
David
>>> [EMAIL PROTECTED] 07/26/05 2:46 PM >>>
I belive you mean 5.3 since that is the most recent level of TSM.
ftp://service.boulder.ibm.com/storage/tivoli-storage-management/maintena
nce/
ftp://service.boulder.ibm.com/storage/tivoli-storage-management/patches
You are licensed for Sysback, right? It is a different and separate license
from TSM.
Once I log into Passport Advantage, there is a download page that shows me all
the products I am entitled to. Sysback shows up there. There is both a
Download and a Order Media button on that page.
David
You do need the base code, 5.6.0.0. It is download loadable from Passport
Advantage if they think you are properly licensed.
>>> [EMAIL PROTECTED] 07/26/05 2:57 PM >>>
I have downloaded the 5.6.5 and 5.6.6 patches. But I thought you had to have
the base 5.6.0.0 to get the license file.
- O
download and media access site.
I guess I'll have to get in touch with the Sales guy to get a copy.
- Original Message -
From: "David E Ehresman" <[EMAIL PROTECTED]>
To: ADSM-L@VM.MARIST.EDU
Subject: Re: IBM Tivoli Storage Manager for System Backup
Date: Tue, 2
Update the vol to status=readonly, run the audit vol ... fix=yes, the set it
back to status=offsite. The tape should now show EMPTY
david
>>> [EMAIL PROTECTED] 07/28/05 3:41 PM >>>
Can't - it's offsite.
--- "Ford, Phillip" <[EMAIL PROTECTED]> wrote:
> Try an audit volume on it and see what ha
I have no insight into your problem but if I thought I was not backing up
everything, I would be on the phone to TSM suport immediately!
David
Do you have many tapes that age offsite for two years? I just checked and we
have 1800 offsite tapes (3590s) and only 4 that are older than two years. We
do reclaims on copypools at 60% reclaimable.
David
>>> [EMAIL PROTECTED] 08/12/05 12:46 PM >>>
We use 3590's and LTO, with copies onsite an
I second the recommendation to look at Servergraph, http://www.servergraph.com.
Not only will its trending components help you forecast the need for more
tapes, its various other charts will help you determine how the tapes are being
used and tons of other info about your TSM installation.
Dav
>We back up around 220 Gb of data to the disk pool overnight, then migrate
>(with caching on) to cartridge in the morning.
>
I assume you mean TSM storage pool caching rather than 3390 caching. If so,
I'd turn off caching for a day and see if that makes a difference. It is my
understanding th
The preschedulecmd can be an os script or program. So if you can identify your
conditions from an OS program, you can certainly program it to do different
things.
david
>>> [EMAIL PROTECTED] 08/17/05 3:16 PM >>>
Hi,
Is there a way to run different commands before and after different schedule
>Regarding COPYPOOL, I don't see any reason to collocate that at all.
Uhh, how about for restore performance when it matters most, during a disaster!
David
Servergraph/TSM, http://www.servergraph.com, can send alert emails based on any
message that appears in the TSM log. That's how we get notified that a drive
is having a problem.
David Ehresman
University of Louisville
>>> [EMAIL PROTECTED] 08/29/05 12:26 PM >>>
The Operational Reporter shows a
I found the following in the 5.3 client README in the
What's new in the IBM Tivoli Storage Manager Version 5.3.0 Clients:
section.
- Multiple nodes can backup, archive, restore and retrieve objects
under a single node
*---
How would this new functionality be used?
David
I run in roll forward mode, so with a data base backup and a good recovery log,
I can recover to the point of crash. So I do not do TSM mirroring of my db
which is on ESS but I do mirror my (much smaller) recovery log to maximize the
chances I'll be able to recover in the unlikely case that I h
If you need to be able to do month end restores for seven years why not let TSM
work as designed and just save your incrementals for seven years. Depending on
your rate of data change, that could well be less data than FULL monthlies.
David
>>> [EMAIL PROTECTED] 09/02/05 4:26 AM >>>
Yes, but t
Servergraph does have an admin command interface as part of their
Servergraph/TSM product.
David
>>> [EMAIL PROTECTED] 09/02/05 6:05 AM >>>
Maurice,
> And for the people who's interested: I will keep it open source
freeware
This is a REAL good idea, and I bet that lots of people would be
i
Add a volrange=A0,A9 to your checkin command and it will only checkin
the Ax tapes and leave anything else in INSERT status.
>>> [EMAIL PROTECTED] 09/07/05 3:25 PM >>>
We have 2 libraries that use 3590 tapes, one mainframe and one UNIX. The
problem arises when an operator puts a non-U
What kind of library? If it uses atldd, did you reinstall that software?
>>> [EMAIL PROTECTED] 09/07/05 3:54 PM >>>
Over the weekend part of the computer room lost power, thanks for it being
the TSM server. I have lots of disk issues but at the moment I just want to
get the library connection wor
If you're still having problems, you might call tivoli. There was a min level
of sysback that I needed for aix 5.2 and I think it might have been 5.6.6 but
I'm not sure.
>>> [EMAIL PROTECTED] 09/07/05 5:53 PM >>>
Server
TSM 5.2.2.5 running on AIX 5.2
NIM master
TSM4SBR 5.6.5.0
Client
AIX 5.2
When we had a similar problem Servergraph/TSM told us the bottleneck was the
network I/O. Upgrading the TSM server to a gig ethernet card and switch
dramatically improved the backup times of our Novel clients.
>>> [EMAIL PROTECTED] 09/12/05 1:43 PM >>>
Hello,
We have a Novell client who's back
All of our san attached disk comes thru a SVC. It's great!
David Ehresman
University of Louisville
>>> [EMAIL PROTECTED] 09/12/05 4:40 PM >>>
We are considering IBM's SVC (SAN Volume Controller) vitalization layer
in our new TSM server configuration.
Anyone currently using IBM's product?
The
If there is outstanding tape activity or requests, that inventory will never
finsih. If it is going to work, it will complete in just a couple of minutes.
You may need to cancel the inventory and then do the checkin. I always use the
volrange=xx,yy parm on my checkins to make sure I d
Bill,
Did you ever get this resoved? I'm guessing that the tape you're trying to do
move data on has a file split across tapes and you don't have one of those
tapes onsite. So it goes to the primary pool instead to find the data.
David Ehresman
>>> [EMAIL PROTECTED] 09/16/05 9:26 PM >>>
TSM
Do you have a TSM DRM Plan? The devconfig backup is contained within it if you
have one.
David
>>> [EMAIL PROTECTED] 09/26/05 3:46 PM >>>
I am needing to restore 3 TSM databases (actually 4, but the primary
library manager is already done). I must not have run backup devconfig
for a while, bec
If size is not a problem, turn off client side compression!
>>> [EMAIL PROTECTED] 09/28/05 10:51 AM >>>
hi, i have a question :
any ideas how to speed up AIX client backup of 8,259,711 files ?
it's not about size , these 8 mil files have cca. 3GB even less ...
client aix 5.2 via TCP network LAN
Have you tuned your aix memory settings, e.g. maxperm/minperm?
>>> [EMAIL PROTECTED] 09/27/05 2:27 PM >>>
No we are not but the expire is starting to take so long it is starting
to overlap. We have 4 GB memory and are using 2 GB of it and starting to
page at about 40% we have a 2 GB paging pool
-
>>Traversal of the file system can also have a profound impact. Even if you
are not actually backing up a lot of the data, it can take a long time to
traverse 8+ million files.
RESOURCEUTILIZATION of 5 might help... see reference info for this option
in the client manual.<<
Andy,
Are these two pa
201 - 300 of 766 matches
Mail list logo
