Video Screencast Help
Protect Your POS Environment Against Retail Data Breaches. Learn More.

Question: Backup at one site, restore at another.

Created: 09 Jan 2013 • Updated: 16 Jan 2013 | 8 comments
This issue has been solved. See solution.

Hi,

 

I got a Netbackup environment that I inherited from a previous admin.

We have two sites (A and B); each has a media server and 2 tape libraries. Master server is at site A.

I have a request to run a backup to tape at site A (a once off full backup) and then eject the tape from the library at site A, ship it to site B and run a restore at site B using the library there.

This is what I have done:

- Created a new policy for the client machine that will use the volume pool where the tape that I will be using exists (this pool is dedicated to the client I'm doing this task for).

- This client has 5 dedicated tapes in their volume pool, 2 of them are configured with a specific retention period, the other 3 have "0 Kilobytes" and "0 Valid images" and a retention period of "0" (i.e retention level).

- As I do not want to break any tape rotation, I will be running this backup on one of the 3 tapes with no images/data.

- The policy that I have created uses a retention level of 24 which translates to "infinity" on this system.

My understanding is that all I now need to do is:

- Run a manual full backup for that policy, it will use one of the empty 3 tapes that has a retention level of "0" and create an image that has an infinite expiry (level 24).

- Once the backup is completed, right click the tape, "eject from robot".

- Go to the physical library, pull the tape out, take it to site "B", and load it in one of the libraries there.

- Run a restore job.

- Once the restore is completed, go to the catalog, expire the image.

- eject the tape, unload it from the library at site B, return it to site A, Put it back in the same slot it was in. run a quick erase. This tape should now be back to the status before the backup ran (i.e "0 Kilobytes" and "0 Valid images" and a retention period of "0")

Is there anything else that I need to do? Or anything I'm missing?

Comments 8 CommentsJump to latest comment

sazz.'s picture

We have two sites (A and B); each has a media server and 2 tape libraries. Master server is at site A.

Are you sure they have got the same master server?

If yes then at Site A check if it has used a different media server to backup the data and you have to change the media host override to do the restore from different media server or you can change the tape ownership to new media server.

If you have got the complete different setup(with different master server) at site B which I think you should then you have to do phase  1 and 2 import to restore the data.

SOLUTION
Manjunath Rajanna's picture

It looks like you are following the right process.

Manjunath Rajanna's picture

From Site A:

Start Manual backup to use 3 empty tapes with infinity retention . Once the backup is done will eject those tapes from  libary and send to site B

In Site  B:

Then you load the medias in site B library then you will start the restoration. Before that you have to change the ownership for that media because your backed from different media server (Site A) while restoring will use different media server (Site B). Once the restoration is done. Will expire the media from the catalog then eject the tapes from library and send to site A.

In Site A, load the medias in libary then run quick erase on those media and reuse for backup.

 

 

 

Marianne's picture

 

- Go to the physical library, pull the tape out, take it to site "B", and load it in one of the libraries there.

-----> Configure Media Host Override in Master server Host Properties -> General Server

- Run a restore job.

- Once the restore is completed, go to the catalog, expire the image.

- eject the tape, unload it from the library at site B, return it to site A, Put it back in the same slot it was in. run a quick erase. .......

No need for quick erase. If all images on media is expired, NBU will automatically overwrite from the beginning.

Supporting Storage Foundation and VCS on Unix and Windows as well as NetBackup on Unix and Windows
Handy NBU Links

maxsven's picture

HI All,

I did all the recommended steps, but I'm getting the following error when I run the restore from the site B media server to the site B client (we are restoring to a different client that the one we backed up from). Please note I have the same ports from media server 1 that does the backup to the client open. Mainly 1556 and 13724. Name resolution forward and reverse works OK. any tips?

 

16/01/2013 7:50:59 PM - granted resource *****
16/01/2013 7:50:59 PM - granted resource "tape drive name"
16/01/2013 7:51:01 PM - Info bptm(pid=7504) INF - Waiting for mount of media id ***** on server "site b media server" for reading.
16/01/2013 7:51:04 PM - Error bpbrm(pid=6464) bpcd on "client name"exited with status 59: access to the client was not allowed
16/01/2013 7:51:05 PM - connecting
16/01/2013 7:51:07 PM - Error bpbrm(pid=6464) bpcd on "client name"e exited with status 59: access to the client was not allowed
16/01/2013 7:51:07 PM - Error bpbrm(pid=6464) cannot put rename file on "client name"     
16/01/2013 7:51:07 PM - Info bpbrm(pid=5224) child done, status 59         
16/01/2013 7:51:07 PM - Info bpbrm(pid=5224) sending message to media manager: STOP RESTORE "client name"e_1358307134   

Marianne's picture

Status 59 - One of two possibilities:

  1. Destination client does not have SERVER entry for DR media server
  2. Destination client does not resolve IP address for DR media server to the exact same name as SERVER entry.
    So, if SERVER entry contains DR-media, but client is resolving IP address to  DR-media.FQDN, the comparison will fail.

Create bpcd log folder on destination client and test connectivity from DR-media using bptestbpcd or another restore attempt.

Please rename the resulting log in bpcd folder to bpcd.txt and post as File attachment

Supporting Storage Foundation and VCS on Unix and Windows as well as NetBackup on Unix and Windows
Handy NBU Links

maxsven's picture

Thanks all...
The last error was due to the guy managing the node making a wrong entery in the hosts file (i.e DNS). all sirted.

Marianne, appreciate your help!