This error is caused by duplicate keys in the TSM Studio configuration file. In our own testing labs we have re-created this issue by starting more than 1 copy of TSM Studio at the same time. There is no issue running multiple issues of TSM Studio simultaneously but we suggest waiting until TSM Studio is fully activated before opening another copy.

This will be resolved in TSM Studio 2.9.3.4. In the meantime to remove the duplicate keys now download the following program –> DuplicateKeysFix

Permalink.

What are the Activity Log, Events and Summary Table.  

The Activity Log is the message log for TSM. Everything that is logged in TSM goes into the activity log. The Events Table stores information about anything that is running from the internal TSM Scheduler including admin tasks and client backups.   The Summary log is a summary of all activity that happens on the server such as admin tasks ( expiration, reclamation etc ), backups and restores.   These TSM object are stored in the TSM Database so changing any retention values will affect the size of TSM database so keep an eye out of the database utilization when increasing these values.   To change the retention values for these objects in TSM Studio, right click on the TSM Server in the TSM Server navigation windows and select Server Properties

Goto the logging tab and simply choose the new retention values for the Activity Log, Event  and Summary Tables. The values are measured in units of days

Permalink.

When TSM Studio 10 is installed, it is installed with a prepackaged version of IBM’s TSM Admin CLI   (dsmadmc.exe).  The version in 2.9.4.6 is not supported on Windows 10.

To run on Windows 10 you will need to install and use the version that is in the latest TSM Backup Client you can use the instruction in the TSM Studio Installation documentation see link below.

https://www.spiritsoftware.biz/knowledgebase/installing-the-administrative-cli-windows/

Then go to setting –> options

And setup the new location of the dsmadmc executable

This issue will be resolved in version 3 which will be a free upgrade if your support and maintenance is up-to-date

Permalink.

All new Tapes being put into a Physical Tape Libraries will require a barcode on them. In most cases the vendor that supplied the tapes should also supple the barcode label.

TSM requires that the first time you put a new tape into the library you run the LABEL command.The LABEL command instruct’s TSM to write an Internal Label at the start of the Tape and this internal label will match the external barcode label.

The Assumption with these commands are that you are using a Fibre Channel or SCSI attached Library The tapes can be either placed into the slots in the library or they can be placed in the CAP slots ( TSM called these external IO slots – BULK IO )

When the tapes are ready you can run the following command

LABEL LIBV libraryname SEARCH=BULK LABELSOURCE=BARCODE CHECKIN=SCRATCH
Where libraryname is the name of your library
SEARCH=BULK tells TSM to search the external Input Output slots for the new tapes If you are not using the CAP ( External IO Slots ) and the Tapes to be labelled are already in the library then change the SEARCH=BULK to SEARCH=YES
LABELSOURCE=BARCODE tells TSM to make the internal tape label the same as the barcode
CHECKIN=SCRATCH tells TSM to mark this tape as a scratch tape When using TSM Studio open the Libraries Dataview in the Libraries and Volumes Category

Right Click on the Library that has the tapes in that you wish to Label and select the Label Command

Enter the options for the Label Command as above and press Ok

NOTE: When using the SEARCH=BULK TSM will prompt you to enter the Tapes and wait for a reply. Click Here to see the Article on replying to requests

Permalink.

The RelabelScratch parameter is aimed for use with Virtual Tape Libraries (VTL) and instructs TSM to relabel the tape when it becomes a scratch tape

NOTE: Do you not use the parameter below TSM 5.5.3 as there is a major bug with this option which can cause your tapes to be relabeled incorrectly

Why is this parameter needed?  ( This is based on my experience with the Falconstor VTL Product)

When using VTL’s with De-Duplication, the Virtual Tape Library Engine and DeDupe Engine are separate subsystems ( either within the same System or in different Systems ). In these cases the VTL Engine is just a transient area for your data before it is deduplicated and is moved to the storage managed by the DeDupe Engine.

Over time TSM will expire old data, reclaim tapes and tapes will move back to the scratch pool. The problem is that the VTL and DeDupe engine do not know that these tapes have become scratch tapes and so will not remove the expired data. The only way that the VTL and DeDupe systems know that a virtual tape has become a scratch tape is when TSM starts writing on that virtual tape from the beginning again.

To overcome this problem the RELABELSCRATCH parameter was introduced. This will tell TSM that it should rewrite the internal TSM label on the virtual tape when it becomes a scratch tape. This will let the VTL and DeDupe engine know that the data on that tape can be removed from the VTL and DeDupe Engines.

To see how this parameter is set on your library

Run the following QUERY LIBRARY FORMAT=DETAILED and you get output like the following. Checkout the Relabel Scratch Option

tsm: SPIRITTSM55_SERVER1>query library format=detailed

Library Name: LB0.0.0.2
Library Type: SCSI
ACS Id:
Private Category:
Scratch Category:
WORM Scratch Category:
External Manager:
RSM Media Type:
Shared: No
LanFree:
ObeyMountRetention:
Primary Library Manager:
WWN:
Serial Number: 9EZ7Z00M08
AutoLabel: No
Reset Drives: No

Relabel Scratch: Yes

Last Update by (administrator): ADMIN

Last Update Date/Time: 02/04/2010 04:58:33

When using TSM Studio go to the Libraries Dataview in the Libraries and Volumes Category and check out the Relabelscratch column

To change the relabelscratch parameter the Library cannot have any activity to it

UPDATE LIBRARY libraryname RELABELSCRATCH=YES

where libraryname is the name of your Library

To change using TSM Studio

Right Click on the Library to bring up the Commands Menu and choose Edit

Check the Relabel Scratch Option and Press Ok

Permalink.

If SQL Server is running on the same Server as the TSM Studio Server service then we recommend 12GB of memory as a minimum to ensure the healthy running on TSM Studio Server.

By default SQL Server will be configured to use all available memory this will need to be changed so that there is enough free memory for  TSM Studio Server to function. It is recommended  that SQL Server be given 4GB of memory and the other 8GB is left available for TSM Studio Server.

To change the default memory settings in SQL Server. Start up SQL Server Management Server right click on the server and select properties

Select Memory and then enter 4096 into the Maximum server memory field

Permalink.

In TSM Studio Server 2.9.011 and above there are a number of added enhancements to help tune TSM Studio for you environment.

The first area to check is on the Concurrency Tab of the TSM Studio Server properties dialog which can be accessed from Configuration–>Properties as shown here



 

The main setting to check here are Concurrent Reports and Concurrent Collector Threads. Report creation can consume large amount of memory so it is recommended to keep this number low.



Also some of the collections can consume a large amount of transaction log space especially the activity log collector.  Each collector  can be individually tuned.  Goto Configuration->Collection schedules and edit the Activity Log collector and change the Concurrent Collectors to a small number (less than 5 )



Also check out the below link for SQL Server memory settings


https://www.spiritsoftware.biz/knowledgebase/microsoft-sql-…-studio-server

Permalink.

As of TSM Studio Server version 2.9.1.0 security has been tightened up and no longer stores any security information in the registry.

The default logon to TSM Studio Server is now admin with a default password of password, (Domain field should be blank otherwise TSM Studio Server will assume it is a windows userid).

We strongly advise changing the admin password using the following procedure.

To change the password for the inbuilt admin userid use the Security–>Change Admin Password

Security in TSM Studio Server is now defined by assigning role(s) and Windows groups to a TSM Studio Server security group.

The TSM Studio Server security system will automatically create a group named Admins that has access to all functions in TSM Studio Server, the first time it is loaded it will look for the old group definitions in the registry and load them into this group otherwise it will populate with following default groups Administrators, Domains Admins and TSMStudioAdmins.

Also two roles will be created, one called Administrators which is assigned to the Admins Server group and one called Operational Report Viewer for allowing access to the Operational Report Viewer.

To access TSM Studio Server Groups select Security–>Groups

To modify the Admins Group Double Click or select edit from the Toolbar

Add Windows Group that should have admin access to TSM Studio Server, One Group per line

Permalink.

Alternate Row BackColor Example

Permalink.

Permalink.

This will display Line Numbers as shown below

Permalink.

Permalink.

Permalink.

If you find that it takes upwards of 30 seconds of  establish a session to the TSM Server from the Admin CLI (dsmadmc.exe). It maybe because your TSM server is unable to resolve the clients name or the TSM Server is unable to communicate with any DNS Servers.

Permalink.

TSM has the ability to encrypt data at the client node before sending the data to the TSM server. There are two methods that are available

1. Transparent Encryption

• This is where the encryption key is managed by and stored on the TSM server
• If the client node needs to be rebuilt data can be easily restored
• Data can be restored back to any node that is allowed to impersonate the original node

2. Client Side Encryption

• This is where the encryption key is manually managed and stored on the client using an encryption password
• More secure as data can only be restored if the encryption password is known
• If the password is lost then the data cannot be restored

To enable encryption at the client there are two parameters for setting up and a couple of include and exclude statements for selecting or excluding which files are to be encrypted.

ENCRYPTKEY

The ENCRYPTKEY option is used to choose either transparent encryption or client-side encryption. For client-side encryption there are two options to choose from

ENCRYPTKEY=SAVE  ( Client-Side )

This option will prompt for an encryption password on the initial backup and then store this password in the password file. The password will be retrieved from this file for each subsequent backup.

ENCRYPTKEY=PROMPT ( Client-Side)

This option will prompt for an encryption password for each backup and restore. To be able to restore the data the same password that was using when backing the data up will be required

ENCRYPTION=GENERATE (Transparent)

This option will have TSM generate an encryption key password which is stored on the TSM server and managed by the TSM server.

 ENCRYPTIONTYPE

The ENCRYPTIONTYPE parameter selects what type of encryption is used either DES56 or AES128 with the AES128 algorithm being the stronger of the two

Next is to select which file or directories to include in the backup

use the include.encrypt statement to include files and directories to be encrypted and takes the same format as any other include statement
use the exclude.encrypt statement to exclude files and directories to be encrypted and takes the same format as any other exclude statement

example

ENCRYPTKEY=GENERATE
ENCRYPTIONTYPE=AES256
INCLUDE.ENCRYPT /home/…/
EXCLUDE.ENCRYPT /home/…/test.fil
INCLUDE.ENCRYPT  C:…*
EXCLUDE.ENCRYPT  C:windows…*

When using the client-side encryption the encryption passwords are stored in the TSM.PWD files in unix or in the registry for windows

I would recommended using transparent encryption unless you have a specific requirement not to.

I am option asked how to prove that the data is encrypted. There is no way to do this with TSM and they only way to do this is use a network packet tracing tool such as wireshark. If you are interested on how to do this just send me an email  gelliott@spiritsoftware.biz

For more information see Chapter 5 of IBM Tivoli Storage Manager: Building a Secure Environment

Permalink.

The following command will show all unique management classes used by a node

SELECT DISTINCT(CLASS_NAME) FROM BACKUPS WHERE NODE_NAME = ‘XXXXXX’

to see only management classes used on directories use

SELECT DISTINCT(CLASS_NAME) FROM BACKUPS WHERE NODE_NAME = ‘XXXXXX’ AND TYPE = ‘DIR’

to see only management classes used on files use

SELECT DISTINCT(CLASS_NAME) FROM BACKUPS WHERE NODE_NAME = ‘XXXXXX’ AND TYPE = ‘FILE’

Permalink.

In Windows you can use the Preview Include-Exclude option in the backup client.

Start the backup client and go to the utilities menu and select Preview Include-Exclude

In the Preview Include-Exclude Dialog, choose the type either Backup or Archive ( Your normal nightly jobs will usually be of type backup )

Choose either to just show included files, excluded files or both. This information is ascertained from settings in your dsm.opt file

Choose  the directories and files to be reported on. If you have a large file system and you choose a top level directory this command may take a long time to run and produce a large output file

Lastly choose the output file and select ok

When the task finishes open the output file with notepad and it will look as follows with the files on the left and the management class on the left hand side

The same preview include-exclude can be achieved with the backup-archive command line in both unix and windows as follows.  This example will produce the same out as above

the -traverse=y tells the backup-archive client to include subdirectories

Permalink.

This article describes installation and configuration of the TSM Scheduler Service on windows. The TSM Scheduler service is used by the TSM Scheduling Services to automate client node backups.

I am using the version 6 client for these examples, even though the screens look different it is still the same process for the version 5 clients

After installing the Windows Backup Archive client start the Backup-Archive GUI go to the Utilities–>Setup Wizard

Select Next

Select Next on TSM Schedule Wizard Screen

Choose Install a new or additional scheduler

Choose a Name for the Service. This the name you will in the Windows Services Manager

This is what you will see in Windows Service Manager

Choose an options file to be associated with this TSM Scheduler Service. It is possible to run multiple TSM Scheduler Service’s using different option files

Choose the node name for this TSM Service. Again it is possible to use a different node name for each TSM Scheduler Service and input the password. This node name will need to be registered on the TSM Server or the service will not start.

Choose the userid that the service will start as.  This userid will need to have access to all the files it must backup or be a member of the backup operators group in windows

Choose the location for the scheduler service log files

NOTE:  Some of these options are stored with the TSM Scheduler service and not in the options file and will override the option file settings. These values are stored in the registry at the following location

HKEY_LOCAL_MACHINESYSTEMCurrentControlSetServicesTSM Client SchedulerParameters     (replace TSM Client Scheduler with the name you used when installing the service )

Permalink.

For TSM version 5

SELECT * FROM NODES WHERE NODE_NAME NOT IN ( SELECT NODE_NAME FROM COLLOCGROUP WHERE NODE_NAME IS NOT NULL )

For TSM version 6 and above

SELECT * FROM NODESVIEW WHERE NODE_NAME NOT IN ( SELECT NODE_NAME FROM COLLOCGROUP WHERE NODE_NAME IS NOT NULL )

Permalink.

TSM v6 and Above

SELECT NODE_NAME FROM NODESVIEW WHERE NODE_NAME NOT IN (SELECT NODE_NAME FROM OCCUPANCY)

Permalink.

For TSM v 5

SELECT COUNT(*) FROM NODES

For TSM v6 and above

SELECT COUNT(*) FROM NODESVIEW

Permalink.

SELECT DOMAIN_NAME,SCHEDULE_NAME,COUNT(NODE_NAME) AS NUMBER_OF_NODES FROM ASSOCIATIONS GROUP BY DOMAIN_NAME,SCHEDULE_NAME

Permalink.

SELECT DISTINCT NODE_NAME, CLASS_NAME FROM BACKUPS WHERE NODE_NAME=’XXXXX’

Replace XXXXX with your node name

Permalink.