TSM Support told me , that it is not recommended to abort an audit, since it
could happen during
repairs, that something is dropped and rebuild from a temporary table.
If you stop the audit by killing the process, the result could be quite
uggly.
Regards Stefan Holzwarth
I'm not sure, if this
hi Denis
if you ask for my opinion i do not recomend lanfree backups, why thats
spimle when you do a restore it dosn?t matter if your using san or not youre
restore data always goes trough the lan. and that?s bad, then you have lost
75% of your reason to have lanfree backups, because what is
whats your e-mail then
Kvedja/Regards
Petur Eythorsson
Taeknimadur/Technician
IBM Certified Specialist - AIX
Tivoli Storage Manager Certified Professional
Microsoft Certified System Engineer
[EMAIL PROTECTED]
Nyherji Hf Simi TEL: +354-569-7700
Borgartun 37105 Iceland
Petur,
you say that restores with SAN agent go through LAN regardles of
configuration. Is it true???
According to doc, restores should also go through SAN. Or is it something
special with Connectrix?
Or am I missing something?
Ing. Jozef Zatko
Login a.s.
Dlha 2, Stupava
tel.: (421) (2) 60252618
Zatko
I was recently visited by a man from Tivoli who said this. I have not
confirmed this information.
He told me that this was the problem in current Tivoli Configuration, and it
would be fixed in version 5.
Tivoli has been known to say that this or that isn?t posible but when you
often find
Hi,
I'm a complete novice to TSM. We currently manage a large network of UNIX
servers (mixture of AIX, Solaris, HP-UX and Tru64) and have recently taken
over support of a small group of AIX servers (5 of them) that consist of an
TSM Backup server and 4 clients. The machines run Oracle databases
I think a little clarification is needed.
LAN-free with the B/A client does Backup and Restore through the SAN. Also
the API client can do backup/restore through the SAN.
I understand there is one TDP product for EMC using the timefinder function
that backs up LAN-free, but must restore over
I use this script to gather tapes for DR testing. I think this is what you
want:
Name Line Command
Number
-- --
BRSLIST1 select volume_name,status from volumes where
John,
I was recently asked to run AUDITDB by the Tivoli support center.
They warned that it could run a long time. Everything I saw in the manuals
for my version indicated that AUDITDB was a standalone process. Some
contributions to the list suggested that there was an on-line version,
I have been seeing some postings about the unload/load database
commands. I am a learn as you go tsmer, so
this has peeked my interest. I am in the process of cleaning up
some of the nodes that had resided in TSM. I was
wondering if the unload/load procedure is a
Thanks to everyone that helped!
The issue was that we formatted a new recovery log file. Before the file
format was complete, we tried to extend the recovery log. This corrupted
the log08 file space. When the server attempted to restart, it could not
read the recovery log. Because we are in
Also, as an fyi...
Tivoli also informed us that TDP for oracle on EMC is not supported when
TSM is running on
an AIX RS/6000.. Kinda funny huh? It's actually only supported for Oracle
databases running on a
Solaris system and TSM ALSO must be running on Solaris. Don't vendors
usually make
I asked this question of BMC at least 6-12 months ago and officially got the
answer of this is unsupported . We are getting around it by creating a
separate profile in a different location and move it back and forth between
the temporary area, and the real area before starting the backup.
Hi everybody,
My envorinronment is:
AIX 433 TSM Server 421 STK 9310 ( 4 SCSI Drives
and 4 FC Drives)
The 4 SCSI drives are conected to San Data Gateway
2108 From here these are conected to switch Brocade
The 4 FC drives are conected to Brocade directly
From Brocade to RS6000 are two fiber
We were hit by the klez virus over the weekend In trying to restore some files from
pre infected state I found what appears to be a fatal flaw in the way I have
versioning set up (or a feature in tsm) We are running win/nt 40 sp6a and tsm
421 client, the server is on aix and also 421 These are
Yes I believe when Tivoli looked at the development of the Timefinder
function they looked at the market first. It was concluded that an EMC/Sun
solution represented the larger market share of what customer have
installed.
ie there are more EMC/Sun installed sites than there are EMC/Aix
Jason,
I completely agree with you.
The majority of our databases are running on Solaris attached to EMC.
But, I think that the percentage of TSM Servers running on Solaris versus
running on AIX is a lot more for AIX. I would venture to say that AIX is
the
preferred operating system for TSM.
Cory,
I presume you are using your GUI client for the restore: you need to find
the option to see inactive versions of your file and try again!
Steffan
- Original Message -
From: Cory Heikel [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Sent: Tuesday, March 05, 2002 7:04 AM
Subject: Urgent
Gavin,
There are some things to consider here. Both Oracle and Tivoli (TDP for
Oracle) no longer supports Oracle 7. I highly recommend that you upgrade
that Database. In fact, Oracle is pushing all customers to Oracle 8i or
higher. Having said that, TDP for Oracle 2.2.0 in theory will backup
It seems to me that this was a very rare situation! As you stated, TSM would
not have changed the extension on the file from .exe to .avb, that was the
anti-virus software or virus itself. Since this file did not change in 33
days this was the only file, which was **ACTIVE** since Nov. 29th,
I am using TSM Mirroring for the Database, having 2 copies of each DB
volume.
TSM shows these as Copy1 and Copy2.
From some performance monitoring, it is always the Copy1 DB volumes that
are busy. I am trying
to have a particular DB volume to be Copy2 but not having much luck.
Starting out:
That's the problem, it does not show any inactive versions. The only version, inactive
or otherwise, are the renamed versions. This is why I believe that tivoli is not
recognizing a difference between the renamed file (.avb) and the original file(.exe).
If it saw the .exe version of the file
Hi Just shell script question
I want to manipulate dates and get diff of two variables
But result I want in hrs:min:sec
Start_Time=`date +%H%M%S`
sleep 1 (Just for testing diff is 1)
End_Time=`date +%H%M%S`
TOTAL_Time=`echo $End_Time - $Start_Time|bc`
echo
If you are still having the problem, it may be TCP/IP.
When you restart the ssi, you probably had to give it the host name; if it
can't access that host name via TCP/IP, then you are stuck.
ACSLS is server software, TSM is server software, the ssi sits in between
them (even if they are all
The original file (.exe) is GONE after 33 days since it did not change and
was finally changed (.avb) on 3/2/02! 'Retain Only Versions' specifies how
many days TSM retains the ONLY backup version it has of a file when the
original file has been **DELETED** not changed from the workstation!
What would be the best way to produce a report that shows the number of
client backup successes vs. failures, for a given day?
Peter Glass
Distributed Storage Management (DSM)
Wells Fargo Services Company
* [EMAIL PROTECTED]
I have a problem, I have a shark and I must do a flashcopy of an ms sql and
a lotus domino databases; after that I have to backup this flashcopy using
tsm. I need some information about the requeriments needed to make this
task.
Can you recommend me some sources or information.
Thanks for your
Are you running it as a TYPE=SCSI library or a TYPE=ACS library?
Are you seeing any errors in the TSM log?
If not (and probably not, since this is a TYPE TEMP error),
it may harmless.
Did someone open the library door, or is the CAP full?
If it continues and you can't identify the source, mail
I use this SQL script to get the volumes used by a specific node in a specific
storagepool.
select distinct node_name,volume_name from volumeusage where node_name=upper('$1') and
stgpool_name=upper('$2')
You can either run this from a command line and replace the $1 with the node name and
$2
Thanks to everyone for the responses...
The problem was that the correct SSI was not started. We are reviewing our
startup process.
-Original Message-
From: Prather, Wanda [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, March 05, 2002 11:19 AM
To: [EMAIL PROTECTED]
Subject: Re: Question on
Agreed, my concern is why the retain only did not kick in at the time the files was
renamed to .avb. Is tsm really not recognizing a change in the file even though its'
type changed? I have displayed active and inactive versions for these files and it is
as if the .exe version never existed. I
I have noticed this peculiarity of TSM to have it's own way of making
which is copy1 and copy2. Other than keep playing the add and delete
game, I have a thought that I've been meanin to try out.
The DB and Log vols are in the dsmserv.dsk file. Maybe the simplest
way to put them in the order
Don't know about your adapter/zone questions.
But you can get the ACSLS manuals from the STK web site.
Go to: www.stortek.com
(If you get the run time error pop up, just click NO)
Click CUSTOMERS
Click CUSTOMER SUPPORT/already a customer
That will take you to the Cusomer Resource Center.
I need all info that I cane get , because we are going into a test between
Bright Store and TSM v 4.2 .
So cane anyone help
Hugo Badenhorst
[EMAIL PROTECTED]
011 285 5587
HI folks,
I have an IBM 3584 LTO Library.
I noticed one of the grippers is offline. I was able to download the logs, via
StorWatch and my web browser, but how to I view them? I don't know what format the
file is.
Miles
I have a few answers.
1. Looking at expiration and seeing that it expired 1 backup object does
not necessarily mean it expired 1 file on your client node. The way TSM
aggregates files etc. (I won't go into details, mainly as I don't know the
fine details), makes it an abstraction as to the
Any thoughts on making NEXT STG POOL a virtual volume. i.e. going to device type
server over the network?
Regards,
Joe Wholey
TGA Distributed Data Services
Merrill Lynch
Phone: 212-647-3018
Page: 888-637-7450
E-mail: [EMAIL PROTECTED]
Scott,
How many times a day you archive the Transaction Logs?
Developer Server - once a day @ noon
Production Replication/Mail Hub - every 6 hours
Has anyone experimented with the Buffers and Buffersize settings in the
Domdsm.cfg file?
Not Really
Hope that helps,
Shawn Tullier
ITS
That is correct! In this case, TSM is not concerned about the contents of
the file, but the extension of the file since that was the only thing that
changed! It will appear that the .exe file did not exist because it
immediately expired on March 2/3 when expiration ran. Are you looking in the
If file a file named 'a' is renamed to 'b', there is no way for TSM or any
other application (except perhaps for the application that performed the
rename, but then that depends on the application) to know that 'b' is the
same file as 'a'. It is not a bug in TSM that it doesn't recognize that
We have been using ADSM for a few years. Management has asked if it is
possible to start over, that is make it look like this is the first day we
are using ADSM and run all full backups.
Any ideas?
Thanks,
Debbie
Can't she just run a selective backup instead of an incremental?
-Original Message-
From: Miller, Ryan [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, March 05, 2002 12:26 PM
To: [EMAIL PROTECTED]
Subject: Re: Full backups
How many nodes and filespaces do you have? Do you need to keep the
How many nodes and filespaces do you have? Do you need to keep the existing data you
have? and if so, how long do you need to keep it for after this 'start over'?
-Original Message-
From: Lane, Debbie [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, March 05, 2002 11:17 AM
To: [EMAIL
A couple of ideas:
1) Rename the node's existing file spaces on the TSM server to something
else (i.e. rename /usr to /usr_old). Then the next incremental backup will
create a new file space on the TSM server, with brand new backups. After
you are comfortable that you don't need the /usr_old
Thanks Andy, I think option 2 is what they are looking for.
-Original Message-
From: Andrew Raibeck [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, March 05, 2002 9:29 AM
To: [EMAIL PROTECTED]
Subject: Re: Full backups
A couple of ideas:
1) Rename the node's existing file spaces on the TSM
OK. Perhaps it doesn't need to be said, but I'll mention it anyway: don't
forget to reset MODE to MODIFIED after the full backups complete! :-)
Regards,
Andy
Andy Raibeck
IBM Software Group
Tivoli Storage Manager Client Development
Internal Notes e-mail: Andrew Raibeck/Tucson/IBM@IBMUS
There are several options, that is one, Andy outlined a couple more. It really
depends on what you want to do with the data you have, if you need it or not. If they
are really wanting to start over fresh then it sounds like you don't need to keep the
old data and can reclaim tape space, then
My experience has been to reset the MODE an hour after the backup starts.
It appears that once it actually starts doing the backup that it no longer
looks at the MODE.It makes it easier to schedule administrative tasks
before and after the backups start for the monthly FULLS. Does this seem
I have a problem, I have a shark and I must do a flashcopy of an ms sql
and
a lotus domino databases; after that I have to backup this flashcopy
using
tsm. I need some information about the requeriments needed to make this
task.
Can you recommend me some sources or information.
Josi,
Good points. Sounds like I need to ask more questions.
Thanks!
-Original Message-
From: Miller, Ryan [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, March 05, 2002 9:50 AM
To: [EMAIL PROTECTED]
Subject: Re: Full backups
There are several options, that is one, Andy outlined a couple more.
What do you guys think about this article?
Source: Gartner Research
As the amount of storage being backed up continues to grow exponentially and
tight budgets increase the importance of improving staff productivity and
utilization rates, the following trends have become obvious: the move to
Depends on what O/S your TSM server is running on but for an AIX server, a
script running out of cron which queries events for the past 24 hours seems
to work for me (plus the script checks lots of other things)...
George Lesho
AFC Entprises
Glass, Peter
Hi TSMer,
I have a question for you with HA-TSM Solutions.
Yesterday I tried to make a 3-Mirrow of a TSM DB on a NFS/Network Sharing.
At the beginning the process was going. After 1 Minute I got the return code
4 from TSM.
My question what are the TSM response times [ms] for makeing db/log
And rember if you change the mode to absolute, it's going to affect your
existing backup versions. You will have an extra version of everything,
which may bump some versions into expiring. Possibly some unintended
consequences...
-Original Message-
From: Lane, Debbie [mailto:[EMAIL
Instead of using CRON to schedule the script, why not have a TSM client
schedule with ACTION=COMMAND pointing to the script. Then the 'script' could
be for any OS. We do this on AIX and NT/2K also also use Perl to pretty-up
the output. This way if you need to change the time the script runs, you
The article sounds quite reasonably.
What is the real question behind your question?
Juraj Salak
-Original Message-
From: Pétur Eyþórsson [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, March 05, 2002 6:56 PM
To: [EMAIL PROTECTED]
Subject: Gartner Research article, Puts TSM in Second
Your management confronts you with a required solution (full backup)
and wants you to implement it,
but the problem / business case
the solution shall solve is unknown.
So neither you nor we are able to check
what implementation will fix the problem.
regards
Juraj Salak
-Original
The reason they didn't develop this for AIX first is because TSM is
developed for the broadest user community first. Everybody knows EMC
and Solaris is practically inseparable.
--
Joshua S. Bassi
Sr. Solutions Architect @ rs-unix.com
IBM Certified - AIX/HACMP, SAN, Shark
Tivoli Certified
I am having some problems with the LAN Free backups. Server is 4.2.1.12 (Windows
2000), client is 4.2.1.20, SAN client is installed and I guess setup. I see the node
in question connecting to the TSM Server (under q ses I can see the node and its
session type is server). When I launch the
Well, you can change the copy mode to absolute, which we did for one mangement class.
This effectively ( or should ) make this the equivalent of a selective backup,
excluding nothingit is the mode of archives. Then you could change it back to copy
mode modified the next day.
However, when
Dear all
Our system:
TSM SERVER 412 on IBM AIX 433
TSM CLIENT B/A 412 LINUX on FREEBSD 45-RC
We have problems with backup/restore It doesn't work for some file
systems
We used emulator linux for freebsd to install tsm client
Sequence used to instalation:
1 pkg_add emulator
We have an NT client that's associated with a Solaris TSM Server's daily
backup schedule On the 3rd of every month, an application process runs
on the NT client that clashes badly with TSM's scheduled backup, so we've
been asked to disable the TSM backup of this client on the 3rd of each
month
What would be the best way to produce a report that shows the number of
client backup successes vs. failures, for a given day?
Tough or impossible to answer, IMHO. What constitutes a failure? If
you can understand the answers to that question, you may be able to
answer your own question.
Is
Personally, I find that filespaces are a good indicator. Generally, the
backup_end should be more recent than backup_start. Backup_end should also
generally be within the past day or it missed.
select count(distinct(node_name)) from filespaces where backup_end
(current_timestamp - (1 day))
Hi. One solution is to do the scheduling elsewhere. On that NT client, use
the AT command (or WinAT) to, at Noon, or whenever, on the 3rd day of each
month, execute
net stop TSM Central Scheduler Service
At Noon on the 4th, it executes
net start TSM Central Scheduler Service
Alex Paschal
Put that NT client in a schedule all by itself... (call it domain
nt_special )
then set an administrative command schedule to run at 00:01 on the 3rd of
each month that does an update of the nt_special schedule to set its
startd=+1
should work, I've used admin schedules to alter client schedules
Hi MichaelYes, the load/unload procedure is a kind of house cleaning feature.This is probably something that would result in about 100 messages in the list, but i'd suggest to do a unload/load at least once a year. Preferable, every quarter. However, as this is a time consuming procedure, normal
The most effective way to not run a schedule on a certain recurring day is
to build an admin schedule which runs a TSM script - The name of your
scheduled backup will be different than mine which in this case is
powkarpr2:
tsm: AFCOPS2q sched updatemontly type=a f=d
Schedule Name:
I'd suggest just stopping the TSM scheduler at the start of this job, then
starting it again after the job is done. The NT command NET STOP TSM
SCHEDULERNAME is used to stop the scheduler service, NET START TSM
SCHEDULERNAME will bring it back up.
NT trick, not TSM trick. Just make sure you
As we were preparing to move Netware clients from TSM 37 to 42119, we
did a test on one machine In the process we had to modify the DOMAIN in
dsmopt file Reading the Tivoli manual on Netware client it states that NDS
is NOT backed up with DOMAIN All-Local We then specified NDS: in the
Kent,
We have a weird situation also..
We have a NT client, who does a full backup every week and daily(s) the
other 6.
We wrote a batch job to kill the daily schedule and enable the weekly
schedule (Sat)..
And then on Sunday we wrote a batch job to disable the weekly and enable
the daily.
There maybe better way out there, but how about this solution
If Scheduled Task is on NT, then this should work.
Simply created two batch files, called stop_adsm.bat, and start_adsm.bat
with client's schedule name.
C:\type stop_adsm.bat
net stop TSM Schedule Name
C:\type start_adsm.bat
net
This is a really good doc which comes with the install package of TSM for
netware.
It's located in the install directory of the client.
This is an old document which hasn't been updated since V3 but is still
very informative on
TSM and NetWare.
Hope this helps...
(See attached file:
David,
Are you asking how to best backup your NDS structure?
You are correct, you only need to back up NDS from selected servers We have a primary
NDS box and a couple that hold replicas of the domain Luckily we have never had to
restore that structure, our testing has been successfull but
Is there an easier way of rebouncing the dsmc sched scheduler on Netware
without having to reboot the machine? Seems like every time I go to check
a hung scheduler, I try to unload and then load. When unloading, it never
comes down and thus hangs the server prompt. Only way out is to reboot
Hello dear TSM's gurus,
I am faced now with the writing of an Operational run guide for TSM.
Let say, what we would like to document is all tasks that a TSM
administrator has to do every day and periodically. Nothing to do with the
installation of the TSM server code, but with all daily tasks
If the session is active on the TSM server, cancel it there first, then you should be
able to unload it on the Netware box. If there is no session on the TSM server, the
only way we have been able to correct this is with a reboot. You shouldn't have
numerous instances of this, if you do, you
I looked at our new 4.2 install and sure enough there was ntwback.htm.
It seems to have some good info, our Netware Admin is looking at it.
I never think of looking at these docs other than the READEME's. The
PDF doc for Netware didn't seem to have any detail info.
Thanks,
David Longo
[EMAIL
We have a 3584 with 6 3580 (LTO) drives and 3494 with 14 3590E1A drives
shared between 3 AIX servers containing 4 TSM servers. The 3590 drives are
SCSI attached to 2 san data gateways which are attached to 2 brocade
switches and the AIX servers are also attached to the brocade switches. I
have
Hi Taketoshi,
On Wed, 2002-03-06 at 08:11, taketoshi ide wrote:
Our system:
TSM SERVER 412 on IBM AIX 433
TSM CLIENT B/A 412 LINUX on FREEBSD 45-RC
We have problems with backup/restore It doesn't work for some file
systems
We used emulator linux for freebsd to install
Protect agent works on our Oracle 8.0.5 standard edition.
When we try it with 8.1.7 we get:
RMAN-03022: compiling command: allocate
RMAN-03026: error recovery releasing channel resources
RMAN-00571: ===
RMAN-00569: === ERROR
Is anyone using this product?
Can you share any information with me as to how effective it is?
How well it integrates with TSM.
Easy/Difficult to keep working.
How much administrative time to keep working (outside of TSM).
TIA!
Steffan
It might make sense to put efforts towards an ndmp client, and abandon
a TSM specifc effort entirely.
(Assuming TSM will be ndmp compatible at some point.)
Robert Clark
The Regence Group
Storage Administrator
503-220-4743
Carl Makin
Hi guys
I had installed TSM 4.2.1.14 on AIX 4.3.3 . By nightly incremental backups
were fine till they stopped doing anything and reports sucess
Error is self explanatory . Please advise
06-03-2002 06:02:18 TCP/IP accepted connection from server.
06-03-2002 06:02:18 Querying server for next
I thought the restriction was that both the Production database server and
Backup server (where the BCVs get mounted) had to be Solaris, but that the
TSM server could be on a third server of any type.
_
William Mansfield
Senior Consultant
Solution Technology, Inc
The other thing to consider is that you have to use EBU with the TDPO for
7.x Oracle. Unlike RMAN, EBU requires fairly intense scripting and
detailed knowledge of the DBs.
_
William Mansfield
Senior Consultant
Solution Technology, Inc
Neil Rasmussen [EMAIL
HiEDT integrates very well with TSM.The administrative work to keep the platform running isn't much, however there are a lot of configuration involved setting up the inital platform. But when you have your inital platform running, the administrative work is minimal.For what purpose are you
Where do you download just versions like 4.2.1.14? Although we have soft. subs.
with IBM on whole tivoli products I do not know where to find the latest
version other than 4.2.1.11
[EMAIL PROTECTED]
Sent by: [EMAIL PROTECTED]
06.03.2002 05:24
88 matches
Mail list logo
