Globally Clustered Replicated Master Server

[krish06]

Hi All,

In the following scenario, Netbackup Master 7.1 is on SFW HA 5.1 SP1, single node in each DC, both being gblobally clustered. All the clients in both the DCs are being managed by only one master server which is the part of GCO.

My question here is, if we failover the master instance to SDC DR Site, can that node able to continue the backups as well as restores??

[krish06]

[Mark_Solutions · Accepted solution]

NetBackup is NetBackup however it is configured ...

so it you fail it over and bring it up on the DR site it will still have its polcies active so as long as it can communicate with its media servers and they can communicate with clients it will all work

But you may find data backing up across the WAN?

Hope this helps

[Omar_Villa]

If the remote site is too far and you have many clients using multiple streams, snapshots or special things that will need to comunicate with the master server if your bandwidth is to small you can expirience some socket errors or 25's.

About your media servers is the same deal when they communicate with EMM, so only be sure you can handle this communication and I will recommend to test for a week the remote node so you will know all the caveats you will face and I say a week so you cover all the posible fulls you may have.

Regards.

[krish06]

Thanks for the comments,

We do have a reliable network links as well as fibre links between the two sites.  Primary DC clients will be backedup by primary DC media servers to the Secondary DC Tape library over fibre links, and vice versa. Its a traditional cross-site backup approach.

but the Master server, we would like to have a HA for cross-site, which actually provides HA as well as DR as well.

Thanks

Krish

[Mark_Solutions]

Sounds like you are fine then - nice setup!

Dont forget it a particular naswer helped you to use the "mark as solution" to close the thread and help others searching for a similar question in the future