High Availability options for the CASO server in BE estate

[DunFraggin]

Hi,

Does anyone know if there is a resilient solution for the CASO server in Backup Exec 12.5? 
We mirror all catalogues to our CASO server currently but in the event of a full site DR would need this server to be available at our DR location to enable fast recovery from tape of our backup sets etc without having to re-catalogue each tape (our production solution uses some LTO tape libraries but currently DR has only stand-alone tape drives attached to a BE server and so could potentially take some time to catalogue all tapes in a given media set before we start the recovery process).

Is it possible to set up another replication of the catalogues to our DR site or alternatively a redundant CASO server implementation of some kind?

All ideas greatfully recieved :)

Many Thanks.

[CraigV]

Hi there,

I don't think something like this exists natively on the software itself.
What you can look at doing is to create a separate backup job that backs up the Data folder and any other required folders on the CASO server onto a folder on your DR CASO server.
Failing that, you can look at creating a script that would automatically copy the files across your LAN/WAN to the DR server every night. Once there, you could start up your services and redetect the DB and all associated catalogs, and make sure they're current.

Laters!

[DunFraggin]

HI Craig,

Thanks for your response.  I was afraid there would be no easy approach out of the box to do this.
I think the idea of replicating the CASO server's data to our DR site is going to be the best approach so I will look into this further and see what options can be found.  If I find a good solution I will post it back here.

Thanks!

[CraigV]

Cool!

[teiva-boy]

 What about BackupExec System Recovery installed on the CASO box?  

[DunFraggin · Accepted solution]

Ok so here is what we have put in place this weekend...

Our CAS server is a VM running on our production virtual infrastructure and connecting to a remote SQL cluster for its DB using a SQL alias.
To protect the DB we have set up log shipping to one of our SQL servers at the DR site.  The SQL alias used will mean a quicker DB restore in DR as we won't have to worry about the SQL servers name.
As the CAS server itself is a VM we have set up a weekly clone replication of the CAS server to our DR virtual environment (If it was physical then I guess that teiva-boy's suggestion above would cover this just fine).  Lastly we set up a replication of the Backup Exec data folder from the CAS server to a file store at DR so we always have the latest catalogues available to us at DR.

To envoke DR for the CAS we now simply need to do the following:

- Restore the BackupExec DB on the DR SQL server and set up the SQL Alias as per production.
- Bring the last replicated instance of the CAS online in DR and correct IP etc.
- Restore the Backup Exec data folder from the file store to the DR CAS server (ensuring all services are offline at this time).

We have tested this offline and it all seems to work fine.  We are able to browse the catalogues and select media to restore etc.  Its not a simple click of a button to envoke the DR but for our needs its sufficient.  I guess ultimately it would be nice to see a feature in future releases of BE that can provide this level of HA even if its a paid for added extra as I'm sure we are not alone in having this kind of setup.

Hope this is of use to anyone else and thanks guys for your thoughts!
 

[CraigV]

Hi there,

Cool...PS: this would make a really good article if you flesh it out a bit with more detail...consider creating it.

Laters!