Good solution for offsite backups using TSM on a slow link.

Posted on 2008-10-28
Last Modified: 2013-11-14

We are currently looking into removing tapes in our offsite offices and use WAN backups instead. These backups will be done using TSM to a central office. Problem is, the initial backup is way too big for our current WAN links. Here is the solution we came up with:

1- Go physically onsite and copy over the data to a laptop (ROBOCOPY) and keep the same folder structure
2- Go back to the central office and install the TSM client on the laptop using the offsite server nodename
3- Backup the data on the laptop to the TSM server on LAN
4- Install the client on the offsite server
5- Configure the TSM client to use the same nodename as we did on the laptop in dsm.opt.
6- Configure the scheduler so it would only take incremental backups.

But that didn't work. We ended up with a full backup been taken on the first scheduled backup (first site had a good link so that's ok). So my question is how do i make this procedure work???

Question by:BFUSA
  • 6
  • 4
LVL 68

Expert Comment

ID: 22821698

first thing is, the two machines must use the same FS type (no mix between NTFS, FATnn). Without that, nothing will work.
You can verify this by looking at 'q filespace [nodename]' on the server.
Next, there is a so called guid (globally unique identifier) every TSM client machine carries with it.
On TSM console, you will se that the server does notice the change of the guid when you change between machines (although using the same nodename).
I don't know whether this guid is really an obstacle for your procedure, but it might be so.
I'll do some research ...

LVL 68

Expert Comment

ID: 22821763

Hi again,
found the following>

The GUIDs help to uniquely identify a particular machine (for reporting purposes), regardless of the following:
How many node names are used on the machine
Which network adapter is used to connect to the Tivoli Storage Manager server
Which nodes are connected to the Tivoli Storage Manager servers
For example, if you use nodes GORDON, DONNA, and DAGORDON to connect to a Tivoli Storage Manager server from your desktop machine, all three nodes will have the same GUID. Similarly, if nodes GORDON, DONNA, and DAGORDON connect to multiple Tivoli Storage Manager servers, each server will show the same GUID for these nodes.
When you install the Tivoli software, the tivguid.exe program is run to generate a GUID which is stored in the Windows Registry on all Windows platforms, except 64-bit. The GUID for a client node on the server can change if the host system machine is corrupted, if the Windows Registry entry is lost, or if a user uses the same node name from different host systems. You can perform the following functions from the command line:
Create a new GUID
View the current GUID
Write a specific value
Create another GUID even if one exists.
Table 6
describes the GUID functions and the associated commands.
Table 6. GUID commands Function Enter on the command line:
Create and store a new GUID on the host if one does not exist.
If a GUID already exists, the current value is displayed.
tivguid -Create
Display help for the tivguid commands.
tivguid -Help
Return the value of the current GUID.
tivguid -Show
Write the GUID that is specified in the -GUID option to the Windows Registry.
For example,
-Write GUID = 'string'
uses the value in 'string' rather than creating a new GUID. The string must be a valid Tivoli GUID (32 hexadecimal values).
This function is useful in the following cases:
If the Tivoli GUID is corrupted you can use the administrative client to query the server for the value using the q node nodename f=d command, and set that value on the current machine.
If you want to set up multiple physical machines with the same guid (for example on cluster).
 tivguid -Write
Create a new GUID even if one exists.
tivguid -Write -New  

Author Comment

ID: 22822016
Thanks for the quick reply.

Like i mentionned, i used robocopy to keep the integrity of the data and ACLs (NTFS). If i understand correctly, there is a value in the registry that i could change in order to fool TSM in thinking my laptop is the same entity as the server?

LVL 68

Expert Comment

ID: 22822127
Yes, that's it.
Go to the TSM administrative commandline client, do a 'query node [nodename] f=d' for one of your nodes, take the nn... value beneath 'Globally Unique ID:' nn.nn.nn....  and use it with the command
 tivguid -Write -guid=nn.nn.nn........
to set the same guid on all the machines that will participate in your procedure.
Let's see if this will help!

Author Comment

ID: 22822156

If i change the TivGUID value directly in the registry without using the command line would it work? I exported the key for backup purposes and imported the value of my server on my laptop's registry.

How your wiki can always stay up-to-date

Quip doubles as a “living” wiki and a project management tool that evolves with your organization. As you finish projects in Quip, the work remains, easily accessible to all team members, new and old.
- Increase transparency
- Onboard new hires faster
- Access from mobile/offline

LVL 68

Expert Comment

ID: 22822231
I never did that, but I guess it would work.

Author Comment

ID: 22822703
That did not still does the full backup. I see both computers in file level in the restore window of the GUI. Maybe that's the whole problem? The computer name needs to be the same?
LVL 68

Accepted Solution

woolmilkporc earned 250 total points
ID: 22823056
This could be the clue, as the computername (only under windows, I think) is contained in the high-level part of the TSM resource structure for a node (filespace name).
Do you have access to an administrative command line?
If so, issue a 'query filespaces [nodename]'
If you see filespaces with different beginnings (\\name1\xx and \\name2\xx for example)
under the same node name, then you are right -
computernames need to be identical, and this could be a real show-stopper then.
LVL 68

Expert Comment

ID: 22823075
read 'query filespace' not 'query filespaces'

Author Comment

ID: 22823757
Thanks a bunch. The computers are not on the same domain so maybe i could just rename my laptop to that server's name and prey it works. if this does not work we'll figure something else.

Featured Post

Why You Should Analyze Threat Actor TTPs

After years of analyzing threat actor behavior, it’s become clear that at any given time there are specific tactics, techniques, and procedures (TTPs) that are particularly prevalent. By analyzing and understanding these TTPs, you can dramatically enhance your security program.

Join & Write a Comment

How to fix error ""Failed to validate the vCentre certificate. Either install or verify the certificate by using the vSphere Data Protection Configuration utility" when you are trying to connect to VDP instance from Vcenter.
By default, Carbonite Server Backup manages your encryption key for you using Advanced Encryption Standard (AES) 128-bit encryption. If you choose to manage your private encryption key, your backups will be encrypted using AES 256-bit encryption.
This tutorial will walk an individual through the steps necessary to enable the VMware\Hyper-V licensed feature of Backup Exec 2012. In addition, how to add a VMware server and configure a backup job. The first step is to acquire the necessary licen…
This tutorial will walk an individual through the process of installing of Data Protection Manager on a server running Windows Server 2012 R2, including the prerequisites. Microsoft .Net 3.5 is required. To install this feature, go to Server Manager…

758 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question

Need Help in Real-Time?

Connect with top rated Experts

20 Experts available now in Live!

Get 1:1 Help Now