IBM LTO 3581 L38 are not ready for TSM ?

2007-05-29 Thread Wu Tony

Hello Advancer:

My TSM Server upgrade to V5.3.5.1 on Windows 2003 Server,
but cannot support IBM TotalStorage 3581 2U Autoloader Model L38 , In TSM
Supported Devices Page cannot find any message
for 3581-L38 (*
http://www-306.ibm.com/software/sysmgmt/products/support/IBM_TSM_Supported_Devices_for_AIXHPSUNWIN.html
)*

i don't know what version can be upgrage and support L38 ?

Thanks~


Re: IBM LTO 3581 L38 are not ready for TSM ?

2007-05-29 Thread Efim
IBM 3581 Ultrium Tape Autoloader  mean that all 3581 Tape Autoloader 
supported from the base levels 5.2.0.00
You must use Atape driver from 
ftp://ftp.software.ibm.com/storage/devdrvr/Windows/Win2003/Latest

You don't need upgrade server from current level.


* Wu Tony [EMAIL PROTECTED] [Tue, 29 May 2007 18:41:02 +0800]:

Hello Advancer:

My TSM Server upgrade to V5.3.5.1 on Windows 2003 Server,
but cannot support IBM TotalStorage 3581 2U Autoloader Model L38 , In
TSM
Supported Devices Page cannot find any message
for 3581-L38 (*


http://www-306.ibm.com/software/sysmgmt/products/support/IBM_TSM_Supported_Devices_for_AIXHPSUNWIN.html

)*

i don't know what version can be upgrage and support L38 ?

Thanks~


TSM Incremental Database backups never expire

2007-05-29 Thread Schneider, John
Greetings,
I have just started doing incremental TSM database backups to a
TSM Filesystem devclass.  This has worked very well on some of our more
active TSM servers, where the log would grow so fast that a single full
TSM database backup performed daily was not sufficient.  We don't always
have enough tape drives or scratch tapes available to do extra full TSM
database backups in the middle of the night just because the log is
grown too big.
So we set up a filesystem devclass, and pointed the TSM database
backup trigger to do it's incremental backups to the filesystem.  This
has solved the problem with the log filling up.  But now the filesytems
are filling up!  We have DBBACKUPEXPIREDAYS set to 5, so the full TSM
dbbackups expire in 5 days, but apparently nothing comes along and
deletes the old incremental DB files that get created by the incremental
backups.  I guess I just assumed they would expire when the full
database backups expired.  From the Admin Reference apparently they
aren't expiring because they aren't in the VAULT status?  
If I issued 'set drmfileprocess yes', that could cause our move
drmedia commands to include the file device class files, but that
doesn't sound like a good idea, since the files can't actually be
checked out.  The manual isn't very clear on what happens in this case.
Am I missing something in how to set this up?  

Best Regards,

John D. Schneider
Sr. System Administrator - Storage
Sisters of Mercy Health System
3637 South Geyer Road
St. Louis, MO.  63127
Email:  [EMAIL PROTECTED]
Office: 314-364-3150, Cell:  314-486-2359


Re: TSM Incremental Database backups never expire

2007-05-29 Thread Richard Sims

The TSM DRM regimen must be rigorously followed for it to properly
and fully work.  Deviations make for problems.  Refer to IBM Technote
1115957 and the Disaster Recovery Strategies with Tivoli Storage
Manager redbook for the necessary steps and guidance.  Consider also
that your File approach to DB incrementals was set up as a tape
utilization expedient; but be clear what role disk-locked database
backups can play in the overall disaster recovery strategy, where the
site is presumed lost, and thus disks with it.

   Richard Sims


Re: TSM client backup in a DiskXtender environment

2007-05-29 Thread Yang, Fred
You need to work with your storage admin to determine how you want to
backup and how the retention policy will be on Centera or DX side, below
are a few questions you need to have an answer:

1. Is the file being purged once it's migrated to Centera?
2. How frequent the purge rule was setup to run?
3. Is the files in Centera pool was configured to be never deleted?
4. Are there multiple Centera was configured to replicate each other?
5. How you want to recovery the system when the whole Windows server
with DiskXtender need to be recovered?

In our environment, since we have Centera configured to be replicated
between both site, so we do not need to backup original files which has
been migrated to Centera. In DiskXtender, you can configured whether you
want to trigger the file retrieval from Centera when your bakcup program
try to access the file, I have our DiskXtender configured to use
Special application filtering, and edit the list to set DSM.EXE,
DSMC.EXE, DSMCSVC.EXE, and DSMAGENT.EXE so when TSM need to backup
the file, it will only backup the stub file. 
However, if TSM backup always occur before DiskXtender move the files to
Centera, TSM will still backup the files which deemed to be moved to
Centera, but those files can be configured with shorter retention since
finally it will go to Centera. 
Also make sure your DX admin setup scheduled meta data export, so you
can configured TSM to backup the meta data (along with DiskXtender
registry), which is critical when you need to recover the whole
environment.

Regards,
Fred Yang
Sr. System Engineer
Northshore LIJ Health System

-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Schneider, John
Sent: Friday, May 25, 2007 5:29 PM
To: ADSM-L@VM.MARIST.EDU
Subject: TSM client backup in a DiskXtender environment

Greetings,
We are making plans to implement EMC DiskXtender on a Windows
Filesystem.  We are unclear how the TSM client backs up a filesystem
like this.  In a DiskXtender environment, as files get to a certain age,
say 90 days, they are sent to a DiskXtender backend store, which in our
case will be an EMC Centera.  The files are then replaced with a stub
file that looks like the original name, but is a very small file.  If an
application tries to open a stub file for reading, then DiskXtender
retrieves the file from its backend store, and returns it to the
filesystem This way the filesytem can seem to contain TBs of data, but
only be a couple hundred GBs in actual size.
The functionality is similar to the TSM HSM software, I guess.
So when TSM makes a pass through the filesystem the first time,
it is backing up original files.  Then eventually they get turned by
DiskXtender into stub files. But I am told the stub file has the same
date and name and size as the original file, so will TSM back up the
stub, or will it skip over it because it doesn't look changed?  Or, if
it tries to back it up, won't DiskXtender recall the original file, and
so TSM will back up the original?
My storage guy is saying that TSM needs to back up the stubs, so
that if some sort of corruption happens in the filesystem, we can
restore all the stub files.   He is telling me that if TSM was smart
enough to undertand the Windows Extended Attibutes, it would know how to
back up a stub file. Or does DiskXtender have a way to recreate the stub
files?

If anyone has this implemented, please describe how it works.

Best Regards,

John D. Schneider
Sr. System Administrator - Storage
Sisters of Mercy Health System
3637 South Geyer Road
St. Louis, MO.  63127
Email:  [EMAIL PROTECTED]
Office: 314-364-3150, Cell:  314-486-2359

  -Original Message-
 From: De la Fuente, Jason  
 Sent: Friday, May 25, 2007 12:13 PM
 To:   [EMAIL PROTECTED]; [EMAIL PROTECTED]; Schneider, John;
 [EMAIL PROTECTED]; '[EMAIL PROTECTED]'
 Subject:  Centera Stub Files/DX/TSM
 
 Anyone,
 
 At EMC World this week, some DX folks were telling me that TSM does
 not recognize the extended file attributes that are associated with
 stub files.  Consequently, a backup of a stubbed file would cause the
 entire file to be pulled back from the Centera.  In a later session,
 another customer stated that IBM had addressed this issue and that TSM
 does now support extended file attributes.
 
 Can someone shed some light on this as it is relevant to our current
 direction of leveraging the Centera for the Enterprise Imaging and
 Medical Imaging (PACS) archive.
 
 Thanks!
 
 Jason de la Fuente
 MISD Enterprise Storage Lead
 314-364-3148
 [EMAIL PROTECTED]
 



-
The information contained in this electronic e-mail transmission
and any attachments are intended only for the use of the individual
or entity to whom or to which it is addressed, and may contain
information that is privileged, confidential and exempt from
disclosure under applicable law. 

Re: Audit library problem.

2007-05-29 Thread Roger Deschner
We've seen something like this. It was a tape library microcode bug, not
a TSM problem. It's still unresolved, though it only bites us about once
a year. Still in a finger-pointing dispute with Quantum over their
library microcode.

Looking at your time stamps, it appears that your library is giving back
its stored inventory to TSM, and that it has become corrupted. So you've
got to force the library to actually re-read all the barcodes, instead
of regurgitating them from its own (corrupted) memory. A first step
would be to power-cycle the library and also restart the TSM server.

Roger Deschner  University of Illinois at Chicago [EMAIL PROTECTED]
=== Copy protection: a headache only for the law-abiding. 
=== --New York Times ===


On Mon, 28 May 2007, Erik Björndell wrote:

Hi All!
Have anyone experienced any problems like this one below?


05/28/07   14:08:02  ANR8457I AUDIT LIBRARY: Operation for library 
TLS-412180
  started as process 15. (SESSION: 15, PROCESS: 15)
05/28/07   14:08:03  ANR8788W Unable to read the barcode of cartridge in
  slot-id 16 in library TLS-412180;  loading in drive 
 to
  read label. (SESSION: 15, PROCESS: 15)
05/28/07   14:08:03  ANR8300E I/O error on library TLS-412180 (OP=C0106C03,
  CC=207, KEY=05, ASC=21, ASCQ=01,
  
 SENSE=70.00.05.00.00.00.00.0A.00.00.00.00.21.01.00.C0.00-
  .02., Description=Device is not in a state capable of
  performing request).  Refer to Appendix D in the
  'Messages' manual for recommended action. (SESSION: 
 15,
  PROCESS: 15)
05/28/07   14:08:03  ANR8942E Could not move volume NOT KNOWN from 
slot-element
  16 to slot-element 63000. (SESSION: 15, PROCESS: 15)
05/28/07   14:08:03  ANR8460E AUDIT LIBRARY process for library TLS-412180
  failed. (SESSION: 15, PROCESS: 15)


Erik Björndell
SYSTEMS SPECIALIST

Zetup by Semcon
Mellanvägen 7, SE-461 38 Trollhättan, Sweden
Mobile +46(0)736 84 04 06
E-mail [EMAIL PROTECTED]  mailto:[EMAIL PROTECTED]
www.zetup.se http://www.zetup.se/





Re: Audit library problem.

2007-05-29 Thread Kelly Lipp
Another thing we have seen with Qualstar libraries and later versions of TSM is 
a problem like this if you do not have the Drive Last parameter in the library 
set correctly.  For instance, if you only have eight drives in your twelve 
drive library you need to ensure that drive last says 8 and not 12.  


Kelly J. Lipp
VP Manufacturing  CTO
STORServer, Inc.
485-B Elkton Drive
Colorado Springs, CO 80907
719-266-8777
[EMAIL PROTECTED]

-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Roger 
Deschner
Sent: Tuesday, May 29, 2007 12:20 PM
To: ADSM-L@VM.MARIST.EDU
Subject: Re: [ADSM-L] Audit library problem.

We've seen something like this. It was a tape library microcode bug, not a TSM 
problem. It's still unresolved, though it only bites us about once a year. 
Still in a finger-pointing dispute with Quantum over their library microcode.

Looking at your time stamps, it appears that your library is giving back its 
stored inventory to TSM, and that it has become corrupted. So you've got to 
force the library to actually re-read all the barcodes, instead of 
regurgitating them from its own (corrupted) memory. A first step would be to 
power-cycle the library and also restart the TSM server.

Roger Deschner  University of Illinois at Chicago [EMAIL PROTECTED]
=== Copy protection: a headache only for the law-abiding.  
=== --New York Times ===


On Mon, 28 May 2007, Erik Björndell wrote:

Hi All!
Have anyone experienced any problems like this one below?


05/28/07   14:08:02  ANR8457I AUDIT LIBRARY: Operation for library 
TLS-412180
  started as process 15. (SESSION: 15, PROCESS: 15)
05/28/07   14:08:03  ANR8788W Unable to read the barcode of cartridge in
  slot-id 16 in library TLS-412180;  loading in drive 
 to
  read label. (SESSION: 15, PROCESS: 15)
05/28/07   14:08:03  ANR8300E I/O error on library TLS-412180 (OP=C0106C03,
  CC=207, KEY=05, ASC=21, ASCQ=01,
  
 SENSE=70.00.05.00.00.00.00.0A.00.00.00.00.21.01.00.C0.00-
  .02., Description=Device is not in a state capable of
  performing request).  Refer to Appendix D in the
  'Messages' manual for recommended action. (SESSION: 
 15,
  PROCESS: 15)
05/28/07   14:08:03  ANR8942E Could not move volume NOT KNOWN from 
slot-element
  16 to slot-element 63000. (SESSION: 15, PROCESS: 15)
05/28/07   14:08:03  ANR8460E AUDIT LIBRARY process for library TLS-412180
  failed. (SESSION: 15, PROCESS: 15)


Erik Björndell
SYSTEMS SPECIALIST

Zetup by Semcon
Mellanvägen 7, SE-461 38 Trollhättan, Sweden Mobile +46(0)736 84 04 06 
E-mail [EMAIL PROTECTED]  mailto:[EMAIL PROTECTED] 
www.zetup.se http://www.zetup.se/





Windows 2003 R2 File Restore statistics

2007-05-29 Thread Kelly Lipp
Folks,
 
Last week I weighed into a discussion about restore performance on
Windows platforms and stated that it is only possible to restore
something like 50-60K files per hour.  Someone else in the list provided
information that contradicted my assertion.  This caused me to think and
since I had the resources in the lab, I decided to run some new tests to
see what I might see.
 
Something has apparently changed.   I will stick by my 18 month old
results, but I have now observed much higher restore performance.  Not
sure what the story is.  My original testing was done on Windows 2003
Server.  The new testing I am doing is on Windows 2003 R2 and the latest
service pack.  I conjecture that something changed in the file create
code or that the hardware I am running now is significantly faster than
18 months ago.  I like the former but willing to consider the latter.
My testing all those months ago was on the latest hardware at the time,
including I/O subsystems and it didn't really matter what the storage
was: locally attached SCSI or FC attached SAN storage.  Results were
very similar.
 
Today's numbers look something like 200K files/per hour restoring from
file device class pools on Compellent SAN over GigE to another FC
attached storage array.  Large read and write caches on both sides.
Hardly any CPU utilized on the client and upwards of 80% network
utilization.  I believe my bottleneck is on the Compellent SAN but will
investigate further by teaming my NICs and getting a bit more network
involved.
 
I intend to publish a new STORServer/TSM performance report in the next
couple of weeks detailing what I've learned.  Expect this to include
LTO-4 performance as well as the backup/restore stuff already mentioned.
Thought, though, that I should get what I had together out to all of you
sooner rather later since I was very wrong in my original post.
 
Thanks,
 
Kelly J. Lipp
VP Manufacturing  CTO
STORServer, Inc.
485-B Elkton Drive
Colorado Springs, CO 80907
719-266-8777
[EMAIL PROTECTED]
 


Re: TSM Incremental Database backups never expire

2007-05-29 Thread Schneider, John
Richard,
I could't agree more.  The incremental database backups going to
disk will not help us in a disaster recovery scenario; we still get a
full TSM database backup to tape every 24 hours for DR.  The incremental
database backup was proposed to us by an IBM Global Services consultant
because the TSM log can't grow past 13GB, and some of our TSM servers
have too much activity and their logs were filling up.  We used to allow
the TSM server to trigger full backups in that situation; but the
problem was that occasionally there weren't enough tape drives
available, and the log would fill up while waiting for one.  Then I was
up in the middle of the night recovering a TSM server by extending it's
log temporarily.  Not pleasant to do 2 or 3 times a week, and always
risky.
I would still like to ask that someone address my question about
how do I expire the old incremental database files?  If I turn on
DRFILEPROCESS so move drmedia picks them up, will they go to VAULT
status, then get deleted by expire inv? 

Best Regards,

John D. Schneider
Sr. System Administrator - Storage
Sisters of Mercy Health System
3637 South Geyer Road
St. Louis, MO.  63127
Email:  [EMAIL PROTECTED]
Office: 314-364-3150, Cell:  314-486-2359


-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Richard Sims
Sent: Tuesday, May 29, 2007 10:44 AM
To: ADSM-L@VM.MARIST.EDU
Subject: Re: [ADSM-L] TSM Incremental Database backups never expire


The TSM DRM regimen must be rigorously followed for it to properly and
fully work.  Deviations make for problems.  Refer to IBM Technote
1115957 and the Disaster Recovery Strategies with Tivoli Storage Manager
redbook for the necessary steps and guidance.  Consider also that your
File approach to DB incrementals was set up as a tape utilization
expedient; but be clear what role disk-locked database backups can play
in the overall disaster recovery strategy, where the site is presumed
lost, and thus disks with it.

Richard Sims


ext2 extended file attributes?

2007-05-29 Thread Dave Mussulman
One of my users pointed out that ext2 has an extended attribute to tell
dump not to back it up.  He was wondering if TSM honored that setting.
I gave it a try (chattr +d test_file) and then did an incremental backup
of the directory.  TSM seemed to ignore the no_dump bit and backed up
the file, and upon restore, didn't restore the ext2 attributes (they
were all blank.)  Is this expected?  I can understand, perhaps, not
honoring the no_dump (since it's not dump,) but I'm a little concerned
extended file attributes for ext3 aren't backed up/restored.  Is there a
special flag or setting I need to define on the client for that
behavior?  A bug?

This is on a 5.3.3 client on a RedHat AS4 system.

Dave


Re: TSM Upgrade to V5.3.4.2

2007-05-29 Thread Prather, Wanda
Um.  It's not a patch in the sense that it has an APAR number, it's a patch in 
the sense that it isn't in a formal install package.
 
Go to the FTP site and look in the TOOLS directory: 
 
ftp://ftp.software.ibm.com/storage/tivoli-storage-management/tools/
 
Select either WINDOWS or UNIX depending on where your TSM server runs.
 
Download the files and read the README very CAREFULLY, there are several steps.
 
After you copy the files, you shut down your server and enter a dsmserv runfile 
command to load the files, then bring your TSM server back up.  It's in the 
instructions.
 
I've installed the patch many times on both Windows and AIX, never had a 
problem.
 
Happy TSM-ing!
 
 
 



From: ADSM: Dist Stor Manager on behalf of Lamb, Charles P.
Sent: Fri 5/25/2007 3:25 PM
To: ADSM-L@VM.MARIST.EDU
Subject: Re: TSM Upgrade to V5.3.4.2



Our consultant recommended that we update to V5.4.3.2 for now and in the
spring of '08, update to V5.4.x.x which will allow V5.4.x.x to mature
more.  We have loaded the TSM command line interface V5.4.0.4 on some
PCs which has help our TSM operational personnel since they perform a
lot of command line functions without having to use the ISC V6.0.1.1
media package software.

I did not know there was an APAR to allow the old web GUI to work with
V5.3/V5.4.  What is the APAR number??

-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Prather, Wanda
Sent: Friday, May 25, 2007 2:06 PM
To: ADSM-L@VM.MARIST.EDU
Subject: Re: [ADSM-L] TSM Upgrade to V5.3.4.2

My biggest problem with the 5.3.4.2 ISC was the java-sometimes-pop-up
window for the command line function; it didn't process output from
scripts and command line the same way as the old admin GUI.

I'm curious as to why you stopped with 5.3.4.2 instead of 5.4; the ISC
5.4 command-line window seems much better; queries and scripts come back
pretty consistently the same as the old Admin GUI window.

In the meantime, install the patch that lets you use the old web GUI
with 5.3. 
(Doesn't work with IE7, but works fine with older IE and Firefox).

Naturally it doesn't have any of the new function in it, but the old
stuff works fine. 
And it works with 5.4.

W




From: ADSM: Dist Stor Manager on behalf of Lamb, Charles P.
Sent: Wed 5/23/2007 4:07 PM
To: ADSM-L@VM.MARIST.EDU
Subject: TSM Upgrade to V5.3.4.2



We just upgraded from TSM V5.2.7.3 to V5.3.4.2.  The TSM ISC software
needs some work ( our e-mail will not work with the words I wanted to
use).  We installed the ISC version from the TSM V5.4 media pack which
IBM sent me.

Any word on updates to ISC software??  I am using command line to
complete our daily TSM work since the ISC does not have the q ac
function which the old java web software had (search on both txt and
message).  A lot of re-training for us since everything is moved around
in the ISC software.


Re: ext2 extended file attributes?

2007-05-29 Thread Mueller, Ken
I ran into a similar situation where the TSM client (v5.2.3 at the time)
wouldn't backup/restore the ext3 extended acls on RHEL3.  The problem
turned out to be that the TSM client is looking for libacl.so but
couldn't find it.  It was available as libacl.so.1 (which symlinked to
libacl.so.1.1.0).  By adding a symlink for /lib/libacl.so pointing to
libacl.so.1 I was able to backup and restore files w/extended acls. 

-Ken Mueller


-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Dave Mussulman
Sent: Tuesday, May 29, 2007 3:15 PM
To: ADSM-L@VM.MARIST.EDU
Subject: ext2 extended file attributes?


One of my users pointed out that ext2 has an extended attribute to tell
dump not to back it up.  He was wondering if TSM honored that setting. I
gave it a try (chattr +d test_file) and then did an incremental backup
of the directory.  TSM seemed to ignore the no_dump bit and backed up
the file, and upon restore, didn't restore the ext2 attributes (they
were all blank.)  Is this expected?  I can understand, perhaps, not
honoring the no_dump (since it's not dump,) but I'm a little concerned
extended file attributes for ext3 aren't backed up/restored.  Is there a
special flag or setting I need to define on the client for that
behavior?  A bug?

This is on a 5.3.3 client on a RedHat AS4 system.

Dave


tdpsql client 5.3 is it compatible on 32bit windows OS ?

2007-05-29 Thread Gary Osullivan
Anyone
 
I am in the process of installing tdpsql client on windows 32bit machine.
 
The install was easy for 5.2.1.6
 
It is a completely different story when I try to install tdpsql client 5.3.3 on 
windows 32bit.
 
Is it compatible on 32bit or does (tdpsql client 5.3) only run on 64bit windows 
?
 
You will never quess how many times I tried to reinstall.
 
 
Gary
 
 
 
 
 
 
 
 
 
 
 
 


Re: tdpsql client 5.3 is it compatible on 32bit windows OS ?

2007-05-29 Thread Del Hoobler
Hi Gary,

Data Protection for SQL 5.3.3 was only released for Windows x64-bit.
You should not try to install it on Windows 32-bit.
The latest version for Windows 32-bit is 5.2.1.6.

Data Protection for SQL 5.3.3 for x64 bit is the same functional
level as Data Protection for 5.2.1.6 but compiled and tested on Windows
x64.
I hope this helps.

Thanks,

Del




ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU wrote on 05/29/2007
04:31:53 PM:

 Anyone

 I am in the process of installing tdpsql client on windows 32bit
machine.

 The install was easy for 5.2.1.6

 It is a completely different story when I try to install tdpsql
 client 5.3.3 on windows 32bit.

 Is it compatible on 32bit or does (tdpsql client 5.3) only run on
 64bit windows ?

 You will never quess how many times I tried to reinstall.


 Gary