DR for Netbackup

[Sean_Krantz]

All,

My company is asking the I come up with a plan in the event of a total burn of our datacenter.

If we had:

We had an identical hardware and software setup in a state.

We only want the mail and the oracle servers to be up and running in 24 hours.

HPUX Master
HPUX Oracle Servers
NT Mail

Thoughts are welcome

[Stumpr2]

alot depends on $$money

Many upper management are gungho for DR until they see how much it can cost and then they are afraid of losing some profit sharing :)

1. you could replicate and be up in a matter of hours
2. you could have a hot dr site that is always ready and all you would need is the
offsite tapes (including catalog) to be delivered and put into the library. You
could even ftp the catalog on a daily basis, kind of a poor man's replication
3. I suggest using inline copies for Oracle. It can literally save you a day on the
restore side becuase of the way vault places all the Oracle threads on the
same tape.

[Hywel_Mallett]

A 24 hour Recovery Time Objective is quite feasible with backups, provided you're organised and plan ahead.
If you already have the duplicate hardware, I would create a test environment, take the Veritas/Symantec documentation and the tapes you need, and pretty much follow the instructions for disaster recovery in the documentation.
From experience with Backup Exec, if there's nothing too unusual in your setup, and your databases are protected with agents, the documentation pretty much works. We had to make a few adjustments to accommodate minor differences between the production and standby environments, but nothing major.
Don't forget to document everything, and when you have the DR procedures drawn up, keep testing them regularly to make sure any changes in your environment don't break them.

[Sean_Krantz]

Bob what do you do at your site for your master?

[Stumpr2]

# 2

[zippy]

Sean,


You once said in prior posts that you are an HP shop. So...

Set up an Ignite on your master backup server and Ignite all your servers to this server.

Keep a daily log or email of the backup DB tape numbers, these are the tapes that backup your Master Backup server.

Ignite your master server onto tape (DDS)

To do a total restore of you enviroment, you have a few options.

Restore the master using the DDS tape, then restore the Netbackup database using the DB tapes.

Restore the Ignite images to the master and start restoring the rest of the production servers using ignite images, once you done building the systems you then can use the newly restored Master to restore the data on the other production servers.

You have to keep track of (physical location of) the DB tapes and the Ignite tape or tapes and your good to go.

Cool......