Disaster recovery Strategy for Unix servers

[gallows]

We currently have Sun V100s, 280R, and 440s running Solaris 9, one Media server with Netbackup 5.1. Most of the servers have root, var, and tmp filesystems. We do Daily Inc and full backup of all servers. On almost all servers the root disk is mirrored with Sun SDS. In the event that both disks go bad, I am up the creek. According to Veritas, I would need to reinstall the Solaris OS and then I could restore the Applications from netbackup. There has to be a better way. This would take a long time. Can anyone please offer a better solution?

Thanks,
gallows

 

[lenski]

Do you have jumpstarts set up for these servers?

The Veritas point of view is the same as every backup supplier. The surest way is to build the server up to the point when you can load and use NBU. But talk to Veritas again because they do have a process for flash rebuilding server. We don't use it so I can't remember the details.

 

[comtec17]

Also ask Veritas about there "Bare Metal" product.

 

[gallows]

Thanks lenski and comtec. I have jumpstart set up but it takes between 2 1/2 & 3hours to Jump. This includes the Recommended patches. Then we could use Netbackup to restore the apps. I haven't done a restore so I don't know how long that would take.

I have a copy of the Bare Metal. It is basically attaching another disk to the system that has a bare bones OS and the Netbackup client. Then do a full restore. The additional disk isn't an option for me.