March 12, 2008 11:55 AM
To: ADSM-L@VM.MARIST.EDU
Subject: Re: 3494 and 3592 WORM media
On Mar 12, 2008, at 10:24 AM, William Boyer wrote:
> ... What just isn't connecting is how does TSM know a regular scratch
> tape from a WORM scratch tape? ...
Good question. The TSM manuals leave
Sent: Wednesday, March 12, 2008 11:55 AM
To: ADSM-L@VM.MARIST.EDU
Subject: Re: 3494 and 3592 WORM media
On Mar 12, 2008, at 10:24 AM, William Boyer wrote:
> ... What just isn't connecting is how does TSM know a regular scratch
> tape from a WORM scratch tape? ...
Good question. Th
I have a client that wants to start using WORM tapes for specific nodes/data.
They are TSM 5.4.2.0 on Winders2003 SP2, 3494 library
and 3592 (TS1120) drives. Reading the manuals/redbooks it seems to activate
this:
- Change WORMSCRATCHCATEGORY= on the library definition
- Create a DEVCLASS with W
Nothing that I'm aware of, but you could always run a couple select statements
before and after the expiration process to list the
total occupancy of the stgpool(s) defined in your VTL:
Select sum(physical_mb) from occupancy where stgpool_name in
('stgpool1','stgpool2',...)
The list of stgpool(
On the library manager instance..
Del volh t=remote tod=+0 force=yes volume=091007
Bill Boyer
"A life? Cool! Where can I download one of those?" - ??
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Zoltan
Forray/AC/VCU
Sent: Wednesday, March 05,
e.
>
> -Original Message-
> From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf
> Of William Boyer
> Sent: Wednesday, February 27, 2008 11:45 AM
> To: ADSM-L@VM.MARIST.EDU
> Subject: TOR Limits?
>
> Is there a limit to the number of TSM servers and/or
Is there a limit to the number of TSM servers and/or reports that can be
configured for TOR? I'm running 5.4.2.0 on a Windows2003
server, 7 TSM servers are defined with multiple reports/monitors for each.
Nothing runs! The service is running, the
reports/monitors are activated, but I don't get an
Might run for a while
Select distinct volume_name from contents where type='Arch'
Bill Boyer
Tell me again how lucky I am to be working here...I keep forgetting!
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Lepre,
James
Sent: Monday, F
IBM Tivoli Storage Manager Version 5.4 & 5.5 Technical Guide
Revised: February, 15, 2008
More details are available at
http://www.redbooks.ibm.com/redpieces/abstracts/sg247447.html?Open
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Remco Post
Sent:
Active Data Pools? Or maybe IMAGE backups? IMAGE backup would probably be the
faster restore/recovery and is only a single object to
TSM.
Bill Boyer
"An optomist is just a pessimist with no job experience" - Scott Adams
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PRO
Yes, but then you need to specify the OBJECTS='' to tell the client WHAT to
SELectively backup. Only Incremental backups work with
the DOMAIN, plus SEL doesn't do the SYSTEMobject/state/services.
Put this node in a domain where the mgmtclass is set for MODE=ABSOLUTE instead
of the default of MOD
I know this won't help you now, but if you are running with LOGCOPY volumes
I've been able to recover from a corrupted log file in
the past by renaming the primary log volumes in the filesystem and bringing up
TSM. It will recognize that the primary log files
aren't there and use the copies. I've
So with the 5.5 client, I would to a "Typical" install and specify
"SNAPSHOTPROVIDERFS VSS" in the DSM.OPT file to globally enable
OFS processing of ALL filespaces on that machine. If I wanted to NOT do OFS on
a filesystem I would specify an include.fs with
SNAPSHOTPROVIDERFS=NONE. Or the other
Trying to figure out OFS with the new 5.5. client. With previous releases you
had to install the OFS support which installed the
LVSA agent. Then the FILELEVELTYPE=SNAPSHOT|DYMANIC was used to determine if
you did OFS for locked files. That option isn't
available in 5.5. Now there's the SNAPSHOTP
Does anyone know of a way to see if the OFS support was installed with the TSM
client? I have a client that wants to inventory their
TSM nodes to see which have the OFS/LVSA support installed and which don't.
Looking for a quick and dirty way just to see if it's
been installed.
Bill Boyer
Tell
/2008 10:28 AM >>>
That sounds great! So to confirm, there are no technical gotchas that are going
to prevent data movement from one media type
(LTO1/2) ton another (LTO3/4)? Anyone else done this?
Thanks, Bill!
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECT
Just define up your new tape drives/devc/stgpools and then set the NEXTPOOL= on
your LTO1/LTO2 stgpool(s) to the new LTO3 or LTO4
stgpools. Now just MIGRATE STGPOOL the LTO1/2 over to the LTO4 pool. No need to
do individual MOVE DATA/NODEDATA. Just let migration
handle it all for you. Cancel it a
Had a client of mine have some backup "hangs" when he started loading servers
on a new bladecenter. He opened a PMR and this is the
response from Level2Don't know if this is your problem...
I've reviewed it and the files that you sent in and this issue appears to be
related to Known Issues
Redhat ES 4, ISC6.0.1.1 install. Just loaded the Redhat and nothing else is on
it. The install goes through the 6.0.1 and then
proceeds to the 6.0.1.1 patch, gets to 99% and then hangs. Never goes any
farther. This is the 2nd time I've gotten this condition.
Which is why I went back to a fresh in
Looking for help...TSM 5.4.1.1 on WIndows2003 running the latest IBM tape
driver. The library and drives are at the latest firmware
as of about 2-weeks ago. The drives have application encryption enabled. This
was done through the CE interface on the back of the
drives. A query of the drive VPD d
I would make it DURATION=59 DURUNITS=MINUTES. That way if the previous hourly
backup ran longer, it would have an extra 59-minutes
before registered as a miss. Instead of only an extra 15-minutes.
Bill Boyer
"My problem was caused my a loose screw at the keyboard" - ??
-Original Message
Again, searching the actlog or the summary table for the restore information is
fine as long as what you're looking for is within
the retention of the ACTLOGRETENTION and SUMMARYRETENTION values you've set. Or
if you're saving the actlog on a regular basis you
can search it still. But if it's pas
As long as the last restore was within the retention you've set for the summary
table. QUERY STATUS to find out the current valut.
SET SUMMARYRETENTION to set/change.
Same as for a previous suggestion for querying the actlog. The restore has to
be within the ACTLOG retention value.
Only long te
Also sounds like you are using POLLING. At least for that client. Then you get
a start time randomized based on the DURATION of the
schedule and the SET RANDOMIZE value. If you seitch to PROMPTED for that
client, you should see it pretty close to the starttime of
the schedule.
You can do it via
The barcode labels came with the library and the clips for the tape slot
columns. I don't remember how many different partitions you
can create from them as they are pre-labeled. I think 4. I'm not where I can go
in a look at our 3584. The "clip" for the drives is
on the front of the drive. You j
i Moyer
Highmark
Storage Systems, Storage Mngt Analyst III Phone Number: (717)302-9966
Fax: (717) 302-9826
[EMAIL PROTECTED]
"William Boyer" <[EMAIL PROTECTED]>
Sent by: "ADSM: Dist Stor Manager"
10/22/2007 01:57 PM
Please respond to
"ADS
I do not believe that the DOMAIN option is valid for the ARCHIVE function. The
description for DOMAIN is the drives to be included
for incremental backup processing. For ARCHIVE you have to specify what you
want processed in the OBJECTS= parameter for the
schedule. ARCHIVE also does not process y
Maybe the easiest is to either from the original server or from another Windows
box with the TSM client, start the client CLI and
run a QUERY BACKUP for *.PST files with -SUBDIR=YES. If you want all, the
include the -INA flag to get the inactive version(s), too.
It's quick and easy. Pipe the outp
>From the readme in the latest release (5.4.1) of the TDP Exchange x64 for
>Exchange 2007:
This Tivoli Storage Manager for Mail: Data Protection for Exchange package
(Version 5.4.1) addresses the following requirement:
* Microsoft Exchange Server 2007 Legacy Support - Support for Microsoft
If you have a stand-alone SCSI tape drive, attach that to the TSM server,
define it and then generate the backupset to that
devclass.
Or what about creating a FILE devclass with the with a MAXCAPACITY of either
the CD size or DVD if you're using that. Then FTP them
over to a Windows box to burn
Search the activity log for the reason the tape is being put back to Private
status. Things such as write-protected tape, tape not
labelled and you don't have AUTOLABEL=YES defined for the library, I/O error,...
Bill Boyer
"Backup my harddrive? How do I put it in reverse?" - ??
-Original Me
Windows or Unix?
Windows is easy...just use the TSM Management Console application, Right click
on your computername under the Tivoli Storage Manager
and select Add a new TSM Server... Very wizard based to create another instance.
For the Unix's, there's a section in the Admin Guide for "Running
You would rebuild the ServerA instance with a blank database of the same size
or larger as what ServerA had originally. Then you
would define the server-to-server communication to ServerB. Once that's done,
shutdown ServerA. All the configuration information is
now in the DEVCONFIG file(s). Now y
I have a client with a TS3310 running the latest firmware (just did it a couple
weeks ago) and SCSI LTO3 drives. This has now
happened 4 times in the past year were the I/O ports go inaccessible. During a
checkin we get:
ANR8386W Slot 16 of library TS3310 is inaccessible.
ANR8386W Slot 17 of lib
We have a client with a couple old Netware 4.11 servers running TSM client
4.2.3.0. These nodes will be "retired" and the hardware
will go away. But there is a requirement to keep the data for possible future
restores. We were thinking of installing a newer
version of Netware as a VMware virtual
Windows2003, TSM client 5.4.1.0
TSM Server 5.3.4.0 on WIndows2003.
This one client continually gets combinations of:
08/12/2007 03:00:38 ANS1228E Sending of object
'\\vmtntansa\c$\WINDOWS\system32\CatRoot\{127D0A1D-4EF2-11D1-8608-00C04FC295EE}'
failed
08/12/2007 03:00:38 ANS4005E Error processin
ion Code Script.
I am going to implement this for tonights backup hopefully it will work. If
anyone cares to add anything or see any possible errors
feel free to let me know.
Again Thank you all very much.
Tim
William Boyer wrote:
>Maybe being an ex-programmer, but you have no condition
Maybe being an ex-programmer, but you have no condition code checking to make
sure that the NET STOP/START commands completed
successfully. If you NET STOP's don't work, do you still want to run the
backup? Now if the last NET STOP command fails, then the
preschedulecmd will return with a non-zer
Only thing with doing a script to stop the services, dsmc incremental, and
start the services is, you loose the successful with
skipped files for the event. An ACTION=COMMAND schedule either works (rc=0) or
failes (rc>0). And in your script you have to make
allowances for DSMC INCREMENTAL returni
Just do a DEL VOL DISCARD=YES
And the next time BA STG runs the files that were on that damanged tape will be
re-copied to new copypool media.
Bill Boyer
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Davis,
Adrian
Sent: Monday, August 20, 200
Yeah, but unless the Access application opens the database with exclusive use,
you can get a backup of an open Access database. And
even Microsoft says that's a no-no. I use to have a link to a KB article at
Microsoft that said even doing a COPY of an open MDB
file would leave the copy corrupt an
The volhist TYPE=REMOTE entries are created when a library client requests a
scratch tape and the library manager changes it to
status=private and creates this REMOTE entry with LOCATION of the owning
instance name. It doesn't have anything to do with MOVE
DRM. There is a TYPE=REMOTE for every vo
I have a client that required certain directories on each file server to be
encrypted. We made the changes to the DSM.OPT to enable
AES128 and the include.encrypt statemsents and did a selective always backup of
those directories so that the active version was an
encrypted version and all the oth
August 06, 2007 11:22 AM
To: ADSM-L@VM.MARIST.EDU
Subject: Re: ANR1076E Path not found on UNC name in the domain
Try running dsmc from a domain admin account and see if it's a
rights/permissions issue.
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] Behalf O
Windows 2003 server, TSM client 5.3.2.0, TSM Server 5.3.2.0 on Windows.
The share in question is on a NAS type device called a SNAP! server. We have
changed the TSM Scheduler service to use a domain logon
that has access to that share. For testing we logged on as that user and were
able to map a
If you server has enough resources (CPU, memory, storage...) then put the new
instance on the same server. You have to update the
clients DSM.OPT file for the ones that backup to the new "instance" anyway, but
instead of chaning TCPSERVERADDRESS you change
TCPPORT.
You currently have your TSM in
This is from the README.SRV from the 5.3.2.1 patch level. If you check out the
readme for the 5.3.2.0 base level I'm sure it'll
probably be the same:
Migration and LAN-free Considerations
*
Only storage agents Versions 5.2.3, 5.2.4, and 5.3 are supported by
Ti
Yes, Include.encrypt does not INCLUDE files for backup processing or assign
mgmtclass to files. It only specified files to have
encryption applied to them. I just ran in to a similar problem where a user
wanted to exclude everything on a drive except for a
single directory and have that encrypted
What kind of throughput did you see on the unload/load processes?
Bill Boyer
"Growing old is mandatory, growing up is optional" - ??
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Robben Leaf
Sent: Thursday, June 28, 2007 5:44 PM
To: ADSM-L@VM.MAR
Instead of having to restore each *.LOG file every day, why not just code a
POSTSCHEDULECMD that FTP's all the *.LOG files to a
central server as .dsm*.log
Bill Boyer
"Life isn't about how fast you run, or how high you climb but how well you
bounce" - ??
-Original Message-
From: ADSM:
Has anyone just gone through sizing a VTL solution for a library replacement?
Is it as simple as taking your current
occupancy/retention, applying for some compression and using that figure for
the amount of storage behind the VTL? Or maybe I'm just
trying to make something harder than it is. Wou
II Phone Number: (717)302-9966
Fax: (717) 302-9826
[EMAIL PROTECTED]
********
"William Boyer" <[EMAIL PROTECTED]>
Sent by: "ADSM: Dist Stor Manager"
06/19/2007 10:25 AM
Please respond to
"ADSM: Dist Stor Manager"
To
ADSM-L@VM.MARIST
I think what you need to do is have the library device in AIX re-initialize. I
believe that TSM is only reporting on what the Atape
device driver knows about the last time it was initialized at the last boot or
cfgmgr.
Try stopping TSM and 'rmdev -l /dev/smc(x)' to put it to s defined state, or
Did you try putting the filespace name between the {} brackets?
q ba {/adsmorv}/*
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of goc
Sent: Monday, June 18, 2007 10:25 AM
To: ADSM-L@VM.MARIST.EDU
Subject: non visible RMAN backups ...
hi all,
is th
Doing a HELP 8355 from the TSM admin command line gives a suggestion that the
PATH statements do not correctly point to the right
drive(s). Possibly the DEVICE= specified in the PATH isn't the correct DRIVE of
the ELEMENT= you specified. Most libraries and TSM
today support the ELEMENT=AUTODETECT
Try looking at these 2 articles on IBM: 1114638 and1194688.
Bill Boyer
"My problem was caused my a loose screw at the keyboard" - ??
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Zoltan
Forray/AC/VCU
Sent: Thursday, June 07, 2007 2:18 PM
To
If you are completely...totally sure that TSM-SERVER-3 has no record of this
volume, you can do a couple things:
1. AUDIT LIBRARY CHECKL=B on the library client. This will re-sync
the OWNER of the tapes checked in and the TYPE=REMOTE
entries in the volhist on the library manager.
2. On the lib
I have a 3583 with 3 LTO1 drives and 1 LTO3. I have 2 devclass's defined with
FORMAT=ULTRIUMC and FORMAT=ULTRIUM3C and stgpools
defined to use the specific type. I notice that if I had specified
FORMAT=DRIVES, then it was possible that when asking for a
scratch that I would get an LTO1 when I rea
Instead of doing a SELECTIVE backup on a periodic basis, which won't update the
last backup date/time of the filespace, use the
MODE=ABSOLUTE of the backup copygroup. In your domain, make a copy of the
active policy set and change all the management class
backup copygroups to MODE=ABSOLUTE instea
I believe that with LTO3 tapes you don't have the choice of 6 or 8-character
tape numbers. You can only use 8. And the TS3310 only
supports LTO3 drives. I guess that's why there's no option for 6/8-char volsers.
If you're going to have to use the TS3310 for the older tapes labeled with the
6-cha
Until you get some admin that installs TSM and copies the DSM.OPT from a client
that has the NODENAME specified. Once the
CAD/Scheduler startsthat nodename now points to this new box and the real
NODENAME isn't getting backups anymore. I always pay
attention to these messages as they could be
Why not just take out ALL the DOMAIN lines from your DSM.OPT and let it default
to ALL-LOCAL. Which will get you all local drives
(now and in the future!) plus the systemstate,systemservices. If there's a
specific drive you DON'T want as part of your domain,
it's easier to list what you don't wan
How many tapes are in the TAPEPOOL1 storage pool? You have a MAXSCRATCH of 250.
If you have 250...then that's your issue. Try
updating the stgpool to a higher MAXSCRATCH= value.
SELECT COUNT(*) FROM VOLUMES WHERE STGPOOL_NAME='TAPEPOOL1'
Bill Boyer
"Life isn't about how fast you run, or how hig
Appendix D of the TSM B/A client manual describes the steps necessary to
configure TSM to backup cluster resources. I think you're
missing the step for the registry key replication.
Bill Boyer
"Law of Cybernetic Entomology - There is always another bug." - ??
-Original Message-
From: AD
If you do that, then the next backup will EXPIRE the backup copies on the TSM
server. They will then be managed by the Versions
Deleted, Retain Extra and Retain Only parameters. They will not be deleted
immediately! There is a technique that has been discussed
on this list and I've used on severa
Just saw this announcement:
HP AND IBM PASS MECHANISM COMPLIANCE TESTING FOR LTO ULTRIUM FORMAT
GENERATION 4 TAPE DRIVES
Brings New Generation of Tape Hardware with Drive-Level Encryption,
WORM, Capacity and Performance Enhancements
SILICON VALLEY, CALIF. - (April 11, 2007) - HP, IBM and Quantum
Using one of the undocumented SHOW commands:
SHOW SLOTS
Will list out how many usable slots there are in the library. Then
SELECT COUNT(*) FROM LIBVOLUMES
Will tell you how many you're using. Simple math from there. You can also make
a server script
Def scr q_slots 'SELECT <#slots>-count(*)
I always use
Select volume_name from libvolumes where status='Private' and last_use is null
To get volumes that should be (maybe) scratch but aren't for some reason. Now
if you have a real private volume checked in that
hasn't been used yet, it will still show last_use as NULL. You need to check
I have a client with 2 versions of Oracle on the same Solaris machine. They
want to be able to do TSM backups using RMAN and TDP for
both instances.
Solaris 5.8
TSM Server 5.2.9 on AIX 5.2ML9
TSM Client 5.2.2.0
TDPO 32-bit 5.2.0
Oracle 32 bit version = 9.2.0.6
ORACLE_HOME Path = /oraapps/ipocdb/
You don't need to generate this list and do the CHECKIN for each library
clients' tapes with OWNER...Just checkin all the tapes in
the library as PRIVATE. At this point they will show with no owner. Then on the
library client run an AUDIT LIBRARY. It seems that
the entire volume inventory from th
Q VOL ACC=UNAVAIL
To get the volume name. Then you should query the actlog to find out what
happened to make TSM put the volume as unavailable.
Usually because it couldn't be mounted. Could be that the tape isn't really in
the library, or is in the library but couldn't be
mounted because of a ha
Question is more WHY did it go unavailable? Query the ACTLOG around the time
indicated to see what happened that made TSM take the
drive offline. I would also Q PATH to make sure that path is online and check
the OS logs (errpt for Unix and event log for Windows.
You don't specify your OS) for an
Take a look at some of the existing report SQL scripts. Here's the related code
from the hourly monitor CSM value:
scheduled_start between '%s' and '%s'
Lots of good information can be found by looking at the scripts provided by TOR.
Bill Boyer
"A life? Cool! Where can I download one of those?
Almost sounds like the client upgrade to 5.3.4 didn't work. I've had problems
in the past where an "upgrade" over existing client
code doesn't replace all the files. I usually like to uninstall and then
install the new version. I saw this a lot when going from
the 4.x clients up to the 5.1 code.
EXCLUDE.DIR stataments are prrocessed first and nothing else for a match. You
need to change them to:
EXCLUDE "*:\...\Profiles\...\*"
INCLUDE "*:\...\Profiles\My Documents"
INCLUDE "*:\...\Profiles\Desktop"
This will mostly do what you want. Only the files/directories in My Documents
and Destop
You can run in to this same situation if you do not have the correct TYPE of
tapes checked in to the library. If the media doesn't
match the drives. I have a client that uses SDLT drives and some of their older
labels check in as an incompatible type for the
drive, BUT it still checks them in as
I did just what you're asking for a client last month. You can backup-restore
the DB between 32-bit and 64-bit with no problems. I
was fortunate with this conversion that the DB and LOG volumes were on SAN
attached disk. So we just mapped the volumes over to the
new 64-bit server with the same dr
Yes, but only for today...since midnight last night until midnight today. Most
backup windows span midnight..starting early evening
and running through the morning. Use the BEGIND=,BEGINT=, ENDD= and ENDT= to
get the range you need.
Q EV * * BEGINT=-24 ENDD=+0 ENDT=NOW EX=YES
Will show the miss
Maybe a restartable restore? QUERY RESTORE
Bill Boyer
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Gill,
Geoffrey L.
Sent: Monday, January 08, 2007 1:39 PM
To: ADSM-L@VM.MARIST.EDU
Subject: move data
I am trying to move data from an offsite ta
But I believe that the database must be configured for transaction archive log
mode before you can do online backups.
Bill Boyer
"Backup my harddrive? How do I put it in reverse?" - ??
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Thomas Rupp
S
And also the supported level of the TSM server for your device type. Make sure
that your library/tape drive(s) is/are supported on
TSM 5.3.2.0. You may have to upgrade and based on the critical flaw
notification IBM sent out a week ago, you should probably think
about going to 5.3.4.0.
Bill Boye
I have a client that has their 3590 drives connected to an IBM2109
switch...only problem is that the person who knew the admin
password to the switch is no longer there and they dont' have any up to date
contact information for him. They are upgrading to the
new TS1120 drives.
Does anyone know/r
Couple suggestions...
1. there is the -OUTFILE= parameter to specify the output of the dsmadmc
command.
2. use the SQL concatenation operator
select 'move data ' || volume_name from volumes where
stgpool_name='CPSATABG'
That way you only end up with 1 column in the result instead of 2
I am documenting recovery scenarios for MSX. It looks to me like I can do a
recovery from a FULL legacy backup to a database in the
Recovery Storage Group. However, it looks like if I want to apply incrementals,
I will have to restore all of the databases from
that storage group into the Recovery
>From DSMADMC:
Set sqldisplaymode wide
Select 'del vol ' || volume_name || ' discard=yes wait=yes' from volumes where
stgpool_name='COPYPOOLSTORAGEPOOLNAMEINCAPS'
Then cut this and save it to a file. Then from the admin run the file with the
MACRO command.
Bill Boyer
"A life? Cool! Where can I
Use the SQL contatenation operator:
Select 'del vol ' || volume_name from volumes where
stgpool_name='WHATEVERINCAPS'...
Do it right from the dsmadmc command prompt and cut 'n paste right back to
execute them. You may have to issue SET SQLDISPLAYMODE
WIDE so it all fits on a single line.
Actua
Just went through this on a TDP Informix installcheck the DSMERROR.LOG
file. The new 5.3 client REQUIRES a write-able error log
location. If you have ERRORLOGNAME specified in the DSM.SYS, remove it and add
a DSM_LOG environment varilable for the user.
Bill Boyer
"Law of Cybernetic Entomolog
I've had this problem before upgrading from 5.1 versions on Windows. As part of
the "upgrade" you have to uninstall the previous
version. The uninstall of the licenses didn't delete the file and then when you
went to do the install, the new module wasn't copied
over. I would just uninstall the TS
You'll have to upgrade to at least 5.2.7.0 to get support for LTO3 drives.
With other libraries, adding expansion is just as simple as shutting everything
down, adding the capacity and bringing it back up
again. I've done this several times on 3583, 3584 and the new TS33xx libraries.
After Winde
I'm pretty sure that you cannot update/change the devclass of a stgpool. I
would just:
- Define the the new 3584 hardware
- Take all the tapes out of the 3583 and put them in the 3584.
- Update the devclass(es) to point to the 3584 library
- Update the devclass(es)if needed...MOUNTLIMIT,FORMAT,.
'Cause the TSM device driver is for non-IBM devices. You have an IBM Ultrium3
drive and should load the IBMtape driver. The latest
can be found here:
ftp://ftp.software.ibm.com/storage/devdrvr/Windows/Win2003/Latest/
Bill Boyer
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[E
The problem is that after "cloning" the existing database to the 2nd instance,
they BOTH have the same volume information. When you
do the AUDIT LIBR on the library client, the library manager updates the
ownership of all the tapes known by that client. So if you
were to then run an AUDIT LIBR fr
Instead of using a SELECTIVE backup which you need to specify what to backup,
just change the management class(es) used for the
weekly/monthly/yearly backups to MODE=ABSOLUTE. Then you backup schedules are
just INCREMENTAL using the existing DOMAIN, but
actually FULL backups each time.
You also
This is on your library manager instance. If you delete the volhist TYPE=REMOTE
entries, the next time you do an AUDIT LIBRARY from
the library client they will re-sync all the volume information from the client
and build new TYPE=REMOTE entries for only the
volumes that the client has assigned.
For autostarting on Suse:
http://www-1.ibm.com/support/docview.wss?rs=663&context=SSGSG7&q1=linux+start+dsmcad&uid=swg21240599&loc=en_US&cs=utf-8&lang=en
And RedHat:
http://www-1.ibm.com/support/docview.wss?rs=663&context=SSGSG7&q1=linux+start+dsmcad&uid=swg21159406&loc=en_US&cs=utf-8&lang=en
Or just Q DRI F=D on the library manager. Gives you the "Allocated to:" as well
as the "Volume Name:"
Bill Boyer
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Pitt,
Stuart
Sent: Wednesday, September 06, 2006 10:39 AM
To: ADSM-L@VM.MARIST.EDU
S
I have a client using library sharing of a 3584 with LTO3 drives. Right now
their one main instance is so large it needs to be
split. The client does daily incremental as well as monthly backups. The daily
backups are domain STANDARD using stgpools 3584POOL
and 3584CPPOOL. The monthly backups use
Only thing I would disagree with is the order of your checkins. Maybe for an
actual disaster, but for a D/R test where you have your
copypool tapes shipped to the recovery center, depending on the timing of the
dbbackup you'll be restoring and those tapes, there
could actually be "live" tapes tha
Get the documentation. It describes several different methods of backing up a
VM. Shutting down the VM, pausing the VM and using a
functioin "ADDREDO" which leaves the VM running and changed blocks are
written/saved to the redo file. At the end of the backup you
issue a COMMIT of the redo log int
This is part of the new ESX 3 server. There are different processes for backing
up ESX 2.x servers. If you go to the TSM support
page and search on VMWARE you will find this one in particular which has a lot
of links to resources on both IBM and Vmware sites.
http://www-1.ibm.com/support/docview
1 - 100 of 314 matches
Mail list logo
