cancel
Showing results for 
Search instead for 
Did you mean: 

Restore SAP HANA DB Netbackup 8.2 and Studio

new2nbu
Level 4

I'm testing an SAP HANA backup policy within Netbackup and have been reading the SAP HANA Admin guide but I'm hoping to get some clarity on some of the points in the guide.  I'm cold to SAP HANA and especially configuring this within Netbackup so it seems that there are a few different possibilities to configure SAP HANA NBU policy.

NBU Environment (Master/Media):  Netbackup Version 8.2 with Windows Server 2016 (non clustered)
Primary Disk Storage Unit: EMC Data Domain with OST
Client:  RHEL 7.6 with SAP HANA Version 2.0 w/Netbackup client 8.2

The NetBackup policy is configured with the Default-Application-Backup, Hana client, but left the Backup Selections blank.   Leaving the Default-Application-Backup schedule as is I thought would allow the backup to perform the complete DB backup including differentials, incremental, and fulls).

Leaving Backup selections blank I understood this would allow less configuration changes to the multi-tenant db script. (initially we had tested with configuring a multi-tenant db script) 

However, after successful backups (and this seems by design) that I am not able to 'browse'  HANA backup contents using BAR and can not initiate any restores from NBU BAR.  According to the documentation, the Restores have been done through STUDIO.  

My question is in order to perform restore from Studio (through backint) how does Studio reference the backup files that have been backed up and stored on the Data Domain?  Do they need to share a mount point or is there a "link" between Studio and Client?  It seems I'm missing something obvious so thanks in advance.

 

2 ACCEPTED SOLUTIONS

Accepted Solutions

Mouse
Moderator
Moderator
Partner    VIP    Accredited Certified

Yes, the missing link is NBU client on the HANA nodes and its configuration. The client knows which master server to browse (and the server knows if the client is authorised to browse its backups). Then in the NBU Client, there is a module for integration with the backint interface and the initSAP.utl which you customise for your instance during the installation.

So when you see backups in HANA studio, they're coming from the NBU layer which knows about them because you have configured the lookup parameters during the initial configuration. The information as to when the backups are performed and where they are stored in the NBU catalog on the master server. Because NBU handles data placement as a backend process, a client does not need to know where exactly the data is stored, as long as the storage unit is accessible to a media server and the client can talk to the media server, the data will be restored to the client.

View solution in original post

Awesome!  It's all starting to click - thank you so much.  Yes, I saw this article as well and glad to know that I'm understanding it lol.. Thanks again. 

View solution in original post

4 REPLIES 4

Mouse
Moderator
Moderator
Partner    VIP    Accredited Certified

Yes, the missing link is NBU client on the HANA nodes and its configuration. The client knows which master server to browse (and the server knows if the client is authorised to browse its backups). Then in the NBU Client, there is a module for integration with the backint interface and the initSAP.utl which you customise for your instance during the installation.

So when you see backups in HANA studio, they're coming from the NBU layer which knows about them because you have configured the lookup parameters during the initial configuration. The information as to when the backups are performed and where they are stored in the NBU catalog on the master server. Because NBU handles data placement as a backend process, a client does not need to know where exactly the data is stored, as long as the storage unit is accessible to a media server and the client can talk to the media server, the data will be restored to the client.

Thanks so much.  So the initSAP.utl file is the file with the parameters set that allows Studio to manipulate the backups/restores from the HANA client and pass the data through NBU.  So the benefit allows the data to be easily stored off the HANA client onto the NBU primary backup targets.  I also understood through the documentation that Studio is needed for the restores of the HANA client?  Is that correct?  I'm not able to initiate a restore from Netbackup to the HANA Client or another client destination?  Thanks again

Mouse
Moderator
Moderator
Partner    VIP    Accredited Certified

Correct. While backups can be triggered through a manual process, by NBU or from Studio, all restorest must be started from the client, in this case, you'll need to use Studio. This is consistent with the complex restores of other enterprise databases managed by NBU, such as Oracle, for example, or SAP ECC.

Here is a reference for the restore procedure https://www.veritas.com/content/support/en_US/doc/16226696-127422304-0/v110895327-127422304

Awesome!  It's all starting to click - thank you so much.  Yes, I saw this article as well and glad to know that I'm understanding it lol.. Thanks again.