Quantcast
Channel: ADSM.ORG
Viewing all 2470 articles
Browse latest View live
↧

Client Backing up all files again after Ubuntu upgrade

September 15, 2014, 12:48 pm
$
0
0
Hi Folks,

We upgraded our Ubuntu servers to release 12.04

We then noticed that it seemed to be backing all of our files all over again. This is a simple file server and there were minimal changes during the upgrade period.

As I understand it TSM does not use the archive bit for tracking backups.

I thought of cloning the server OS partition and sending it to our data center, then restoring all of the data and then backing it up again to see if that would then stop the remote server from doing a full backup gain and only continuing with the incrementals.

Any advice would be much appreciated.
↧
Search

Database moved, backups won't run

September 16, 2014, 5:52 am
$
0
0
Our TSM Server had over 40 DB volumes defined. On IBM's suggested we moved to a smaller number of volumes. This... did not go as planned.

We're on TSM 6.3.5 running on AIX 7.1. Sunday night we disabled all access to the server with a 'disable ses' command, then ran a full backup. We followed the instructions from IBM (which have been mirrored here in several threads) to delete the old instance of the DB, run the DB restore from the commandline. The Restore took about 5.5 hours. Since then, things have been horribly wrong and IBM doesn't seem to have an answer.

Backups that started during our scheduled window... are still sitting there as if they just started. The most any session has transferred is 2.1MB. We usually move 15TB in a night. Expiration will not run. Any attempt to start an expiration process spawns nothing. However, you can't re-run the command as 'an expiration thread is already in progress'. Cancel expiration doesn't stop this either.

Licence audits have also been frozen since 10am yesterday.

Our Arch Log size has remained static and the active log is empty. I've got 4000 sessions standing open in 'run' status.

The only thing that works right now is tape operations. Tape to tape copies, direct to tape backups of the DB, and Lanfree seem to work.

Anyone seen this? Any ideas?
↧
↧

Impact on NDMP data when upgrading from 6.2 to 7.1

September 17, 2014, 6:10 am
$
0
0
I have a significant amount of NDMP data (+2 PB) stored in a TS3500 on LTO5 though TSM 6.2 server
Majority was backed up NDMP over IP but a small amount was NDMP direct to TS3500 from netapp.

I am soon to embark on upgrade to 7.1 and want to ensure I do not loose the ability to restore existing data.
The majority of the restores I perform are from a TSM CAD running on a Windows server to allow me to trawl the TOC (this BA client is also 6.2).

My question is what order of upgrade.... ba client with CAD first, TSM server first, if I do the ba client do I loose access to the NDMP within the TSM 6.2 server until the server moves to 7.1?

Also my 7 mode N Series 6xxx are soon to be replaced with cluster mode Netapp FAS... I'm assuming NDMP restore from N Series to FAS is supported????

Any other tips welcome (slightly off topic I hear 7.1 has issues with offsite reclaim... anyone encountered this and seen it fixed at 7.1.0.100 or 7.1.1)

Thanks
↧

Tapes in vault not being reported to go back to Library

September 17, 2014, 7:32 am
$
0
0
Hi Everyone,

We have been using TSM for about 4 years and I consider myself still a novice. One of the issues we have is our consultant generated scripts that email us reports of what tapes are going to the vault each day and which tapes should come from the vault back to the library. The email we get that has tapes going back to the library seems to miss tapes though and we end up with a bunch of tapes that are in the vault that have been removed from the pools. (i think they might be the db backup tapes, but i'm not sure).

anyways, here is an output from the actlog that shows a particular tape 000006L4 that was used, but theres no indication that it was ever expired from inventory:

tsm: GDC-UTIL-2>q actlog search=000006L4 begindate=09/01/2014

Date/Time Message
-------------------- ----------------------------------------------------------
09/03/2014 12:11:25 ANR1423W Scratch volume 000006L4 is empty but will not be deleted - volume access mode is "offsite".
09/04/2014 07:00:03 ANR1341I Scratch volume 000006L4 has been deleted from storage pool COPYPOOL. (SESSION: 46346, PROCESS: 647)
09/04/2014 07:00:03 ANR6684I MOVE DRMEDIA: Volume 000006L4 was deleted. (SESSION: 46346, PROCESS: 647)
09/04/2014 07:00:03 ANR2753I (ADMIN_RETURN_TAPES):000006L4 was deleted. (SESSION: 46346)
09/04/2014 19:33:08 ANR8810I Volume 000006L4 has been labeled in library LIBRARY. (SESSION: 46601, PROCESS: 654)
09/05/2014 09:31:25 ANR8337I LTO volume 000006L4 mounted in drive DRIVE2 (mt6.0.0.5). (SESSION: 47495, PROCESS: 659)
09/05/2014 09:31:25 ANR0513I Process 659 opened output volume 000006L4. (SESSION: 47495, PROCESS: 659)
09/05/2014 09:31:31 ANR1360I Output volume 000006L4 opened (sequence number 1). (SESSION: 47495, PROCESS: 659)
09/05/2014 09:36:02 ANR1361I Output volume 000006L4 closed. (SESSION: 47495, PROCESS: 659)
09/05/2014 09:36:02 ANR0515I Process 659 closed volume 000006L4. (SESSION: 47495, PROCESS: 659)
09/05/2014 09:36:10 ANR8336I Verifying label of LTO volume 000006L4 in drive DRIVE2 (mt6.0.0.5). (SESSION: 47495, PROCESS: 659)
09/05/2014 09:37:19 ANR8468I LTO volume 000006L4 dismounted from drive DRIVE2(mt6.0.0.5) in library LIBRARY. (SESSION: 47495, PROCESS: 659)
09/05/2014 09:50:44 ANR2017I Administrator ADMIN issued command: MOVE DRMEDIA 000006L4 wherest=mountable tost=vault wait=no (SESSION: 47509, PROCESS: 661)
09/05/2014 09:50:44 ANR6696I MOVE DRMEDIA: CHECKOUT LIBVOLUME for volume 000006L4 in library LIBRARY starting. (SESSION: 47509, PROCESS: 663)
09/05/2014 09:51:12 ANR8829I Remove volume 000006L4 from slot 17 of library LIBRARY at your convenience. (SESSION: 47509, PROCESS: 663)
09/05/2014 09:51:12 ANR6697I MOVE DRMEDIA: CHECKOUT LIBVOLUME for volume 000006L4 in library LIBRARY completed successfully. (SESSION: 47509, PROCESS: 663)
09/05/2014 09:51:12 ANR6683I MOVE DRMEDIA: Volume 000006L4 was moved from MOUNTABLE state to VAULT. (SESSION: 47509, PROCESS: 663)
09/17/2014 09:18:49 ANR2017I Administrator ADMIN issued command: QUERY ACTLOG search=000006L4 begindate=09/01/2014 (SESSION: 60863)
09/17/2014 09:20:10 ANR2017I Administrator ADMIN issued command: QUERY ACTLOG search=000006L4 begindate=09/01/2014 (SESSION: 60863)


tsm: GDC-UTIL-2>q vol 000006L4
ANR2034E QUERY VOLUME: No match found using this criteria.
ANS8001I Return code 11.


tsm: GDC-UTIL-2>q drmedia 000006L4
ANR2034E QUERY DRMEDIA: No match found using this criteria.
ANS8001I Return code 11.



how can we capture that that tape is no longer needed and can be returned to the library w/o having to manually inventory the difference between q drmedia and what's actually in the vault?

Thanks in advance.

Matt.
↧

vCenter is down, now what??

September 17, 2014, 9:57 am
$
0
0
Since TSM4VE requires vCenter to be up to restore VM's. What if vcenter is down? Without having to purchase more licensing from VMWare (we all know how that goes) to build a secondary vCenter. How are you all protecting vCenter? With our entire datacenter running on VMWare, in a restore scenario vCenter would have to be online first along with its database instance before you can start restoring VM's. What are your backup methods for making sure you are able to restore vCenter with TSM? In our case vCenter is a VM, so backing it up with VE would not be helpful.
↧
↧

dsmcad won't load web gui - Redhat Linux

September 17, 2014, 10:21 am
$
0
0
I have a new RHEL 6.5 server. I installed the 7.1 client and everything seems to work except the web gui. I can run dsmc just fine and the scheduler runs automatic backups. But even though dsmcad is running, when i point my browser at http://myserver:1581 it says "This page can't be displayed".

Also, shouldn't i have a dsmwebcl.log and dsmj.log file in the /opt/tivoli/tsm/client/ba/bin directory?
↧

"Connected servers that have no policy"

September 17, 2014, 11:24 pm
$
0
0
After successfully deploying FastBack, creating a FastBack BMR boot image and completing a BMR test, my FastBack server heartbeat reports the following:

Code:
=========================
Unprotected Servers
Connected servers that have no policy,
or servers that all their policies/schedulers are paused:
      'fastback-bmr' (No Policy)
I'm assuming this is because during BMR restore the client connected to the FastBack server to perform some sort of "syncronization" whereby the client was somehow registered with the FastBack server.

The only reference I can find to this warning message is a an IBM article noting that a similar issue was fixed in level 6.1.7:
http://www-01.ibm.com/support/docvie...id=swg1IC85902

I'm running 6.1.9 so... how do I get rid of this (non-existing) client? Am I supposed to manually clean out this "state.txt" file? I can't find any references to a client "fastback-bmr" in the FastBack Server Manager.
↧

Win 2008 R2 Server - SQL large file backups

September 18, 2014, 7:53 am
$
0
0
Hi Guys,

We have couple of MS SQL servers running on Windows 2008. We are not using TDP, but taking up backup of .BAK flat files created via SQL backup.

These files are quite large, around 50GB in size each and there are about 12 of these. The incremental backup is quite slow it takes around 12-15 hours to finish.

We also run archive backup of these files on monthly basis which takes around 24-26 hours to finish. The backup is done on a dedicated backup network.

On the TSM server side the backup is being written to file storage pool configured on a Data Domain, the TSM Server is v6.2.3 and the BA client is 6.2.4.

I would appreciate any suggestions you have that can help to improve the performance. Or any pointers what bottlenecks should I look for?

Many Thanks
SG
↧
Search

Query to see how much data archived during last 24 hours?

September 18, 2014, 5:15 pm
$
0
0
Guys

I am stuck in a situation where i need to know how much data is being backed up or archived for a particular node in last 24 hours ... my TSM server version is 7.1..

Is there any SQL query which can be used to find out this information ?

Please advice!!
↧
↧

What happens when a new EXCLUDE is applied?

September 19, 2014, 2:07 am
$
0
0
I was wondering what happens to the affected data when a new EXCLUDE is added. Does it get flagged as "To be discarded" and the space it occupies get flagged for reclamation (thus, very quickly (compared to data that is processed according to criteria set out in the governing MC), making the space it occupied available again)?

Perhaps an example would help to explain what I'm getting at ...

Let's say that I decide to exclude the C:\temp directory on all my Windows servers. I decide to do this by adding the following to the relevant CLOPTSET:

EXCLUDE.DIR "C:\temp"

The next time the backups run, the above gets applied to all the Windows nodes and the C:\temp and its contents are flagged as ... what? "To be deleted"? "To be discarded immediately"? Presumably, data that becomes "redundant" as the result of a new EXCLUDE doesn't go through the same process as a file that has been deleted or a file that has dropped off the end of retention due to the number of file versions esceeding the versions threshold.

Thank you for your attention.
↧

Dealing with removed library

September 19, 2014, 4:27 am
$
0
0
Due to operational mix-up, the old LTO4 tape library was physically removed from TSM server and new LTO6 library attached, without first checking out the volumes from the old library. Now TSM thinks that the old library is still there and has tapes in it.

I tried running CHECKOUT LIBVOLUME on old library and volumes with REMOVE=NO or REMOVE=BULK, CHECKLABEL=NO, FORCE=YES, but the command fails.

Code:
09/19/2014 14:06:50      ANR8840E Unable to open device /dev/tsmscsi/lb0 with file
                          handle 2 and PVRRC 153. (SESSION: 878, PROCESS: 12)
09/19/2014 14:06:50      ANR8441E Initialization failed for SCSI library MSL2024.
                          (SESSION: 878, PROCESS: 12)
09/19/2014 14:06:50      ANR8437E CHECKOUT LIBVOLUME for volume EID620L4 in library
                          MSL2024 failed. (SESSION: 878, PROCESS: 12)
How do I make TSM believe that the old library is really gone? Some of the old volumes contain data that I might want to restore with the new library, so I can't just DELETE VOLUMEs.
↧

Cancel a tape mount on TSM 7.1

September 19, 2014, 10:19 am
$
0
0
we just converted to TSM 7.1 under zlinux and are using Media Manager as the go between our 3592 ATL and TSM. We've several instances where TSM hangs waiting for a tape mount; sometimes there is a mount msg on media manager and sometimes there is nothing.

The hung process will not cancel and all tape activity will come to a screeching hault; the only remedy is to recycle the media manager.

So, can a tape mount be cancelled within the confines of the TSM OC? I've looked for this command in the manual and can't seem to find it. Perhaps there is a better way to terminate these pending processes to keep from recycling the media manager.

Any thoughts?
↧

Baclient can't see backups on server

September 20, 2014, 6:05 am
$
0
0
Hi

I am trying to restore files to a Linux server. The files were backed up from an NFS partition but I wanted to restore them to a ext4 partition. The occupancy report on the node statest that there is 1TB of data but when I go in through the webclient I see the folder structer but no files. The situation does not change even when I press to view inactive files. Someone mentioned that filters of some kind might be blocking the view of the files but I have never heard of that. Is this because the files are from an NFS partition. The data is still within retention policy as well.

Thank you
↧
↧

Use backups in a long time

September 20, 2014, 9:29 pm
$
0
0
I got a file pool configuration for backups, each client is assigned to one file pool (on one HDD), when they fill, move to another file pool (on another HDD), and after that send it to a tape. My goal is save 90 versions for a file (the versions will jump from one file pool to another until 90 days had been completed), after three months fill a tape, and after fill it, take it off from the library and then fill another tape for the next three months. Not sure if it is a correct method but is needed for company standard. If this method is correct, how can I use these backups from here to long time, for example, if I want to take one backup from three years ago, how TSM can take it?
↧

Moving Management Console to another server

September 21, 2014, 7:34 pm
$
0
0
Hi Everyone,

Hoping for some advice, maybe best practice, on how to move my TSM Management console over to a new another server.

We are running 5.5.

Basically, the current server where we are hosting this, needs to go.
Just to be clear, this is Windows MMC plug-in, which allows me to generate reports etc etc.
Its basic, but I need to try and take everything which has been setup, over to the new server.

Advice appreciated, as always.

Thanks
↧
Search

Management class question

September 22, 2014, 6:59 am
$
0
0
I created a new stg pool, mgmt class & copy pool w/ destination=<new pool>, and a new node & sched with -archmc=<new mc> in the objects field. The schedule ran fine and copied data to the new stg pool, however, for some reason, directories were backed up using a pre-existing mgmt class. Why is this the case? I want everything (including dir info) to be stored in the new pool.

Thanks!
↧

export server from 5.5 to 7.1 ...

September 22, 2014, 10:08 am
$
0
0
... is this scenario compatible? or must go in two steps via a 6.x server?

for answers thanks in advance
↧
↧

Restore database sql server

September 23, 2014, 2:35 am
$
0
0
Hello all,
I WOULD LIKE TO RESTORE DATABASE SQL SEVER USING TDP , I USED RELOCATE DIRECTORY USING GUI BUT UNFORTUNATLY I FOUND A PROBLEM , Logical file 'DIAM_Data' is not part of database 'DIAM'.
Use RESTORE FILELISTONLY to list the logical file names.
here's the screen shot

Thanks for help
Attached Images
↧

TSM4VE same Node Configuration on multiple Proxy Servers

September 23, 2014, 9:20 am
$
0
0
Hi all

My question is, is't possible to have multiple TSMVE Proxys with the same Node configuration? TSMVE and TSM Server are running on V7.1.1.

It means:
I have ProxyServer Proxy1, Proxy2, Proxy3 and Proxy4.....

On all of this, i want the same configuration for VCNode, VCNode CLI, DataCenter node and Datamover Nodes exist.

VCNode = VCENTER
VCNode CLI = VCENTER_CLI
Datacenter = VCENTER_DC
Datamover = VCENTER_DM1, VCENTER_DM2....
all this are registered Nodes on the TSM Server and

I want to hold all the VM backups under one Nodename, in my example VCENTER_DC.

Is this a possible solution for that? Or must any Proxy Server has his own "TSM Node Names" configured?

Regards
Mario
↧

Advice on tape checkin script

September 23, 2014, 8:32 pm
$
0
0
I've attempted to automate the checkin process for off-site media coming back into a tape library (for a customer with no TSM administrator and little TSM knowledge). I'm using a VBscript (see attachment).

The process is:
1. customer places returning tape cartridges in the I/O station of library

The script logic is then as follows:
2. check tapes into the library as scratch libvols (this will fail for most tapes)
3. check tapes into the library as private libvols
4. query the activity log for all recent checkin activity to get the volume names of the tapes
5. find out the DR status for each tape and process appropriately eg if COURIERRETRIEVE then move to ONSITERETRIEVE

This process normally works fine for a single tape library but this customer has 2 physical tape libraries for copy storage pools. What I'm seeing is that a copypool tape that was originally created in library A when it is placed in the I/O station of library B and processed by the script will *not* automatically become scratch. For example,

ANR2017I Administrator ADMIN issued command: MOVE 000999 tostate=COURIERRETRIEVE
ANR6683I MOVE DRMEDIA: Volume 000999 was moved from VAULTRETRIEVE state to COURIERRETRIEVE
ANR8443E CHECKIN LIBVOL: Volume 000999 in library B cannot be assigned a status of SCRATCH
ANR8430I Volume 000999 has been checked into library B
ANR2017I Administrator ADMIN issued command: MOVE DRMEDIA 000999 wherestate=COURIERRETRIEVE tostate=ONSITERETRIEVE
ANR1341I Scratch volume 000999 has been deleted from storage pool OFFSITE-A
ANR6684I MOVE DRMEDIA: Volume 000999 was deleted

all looks normal, but if you QUERY LIBVOL B 000999 it still displays as private.

Library Name Volume Name Status
------------ ----------- ----------------
B 000999 Private

If the customer happens to guess correctly and puts a returning tape originally created in library B back into library B then the volume moves to a status of scratch. Since the customer cannot easily determine which returning tape should go into which tape library my workaround in the script is to try to update *all* libvols to scratch. This will change 000999 to scratch but is not very elegant.

Library Name Volume Name Status
------------ ----------- ----------------
B 000999 Scratch

Can anyone suggest a better way of scripting/automating this process ?

The reason I don't just move all DRmedia from COURIERRETRIEVE to ONSITERETRIEVE is because there is a limit on the number of tapes being returned every day and often it may take 2 days or more for a COURIERRETRIEVE tape to come back onsite.
Attached Files
↧
Viewing all 2470 articles
Browse latest View live
Search