Backup/restore of large mailstore (over 100gb)

Discuss the Scalix Server software

Moderators: ScalixSupport, admin

seany
Posts: 24
Joined: Thu Oct 25, 2007 4:53 am
Location: Cheltenham, UK

Postby seany » Thu Jan 17, 2008 7:47 am

We have 2xmailscanners, 1 as failover. The scalix servers are IBM x3650s, in a failover configuration and hosting their own OS locally, we only plan to store mail data on the Eonstor.

I've got to go to Frankfurt to configure/install it all in 2 weeks time. That's assuming it all arrives by then :)

adhodgson
Posts: 176
Joined: Thu Mar 02, 2006 8:09 am

Postby adhodgson » Sat Jan 19, 2008 8:11 pm

Hi,

Further to these posts - I found out that BackupExec was the faulting application which was stopping the snapshot from being released - it seemed that if the backup was still occuring whilst the snapshot was still open, and the script was closing it, it would go wrong, even though BackupExec had no business reading from that snapshot mount anyway!

The upshot of this now is that we have consistant backups of Scalix which are running and taking just over 7 hours to complete. I can cope with this.

We initially were going to put this on a Plasmon FC RAID controller, but we had serious issues with this model. Furthermore, there was no real tools that they provided for doing snapshots or live replication between the arrays, like I have seen on other systems, which is why I went down the local storage route.

We are using SATA with just under 200 users now on RAID-10 in a Dell PE2950 server, and we don't really have any IO issues.

Andrew.

adhodgson
Posts: 176
Joined: Thu Mar 02, 2006 8:09 am

Postby adhodgson » Sat Jan 19, 2008 8:12 pm

Hi,

Further to these posts - I found out that BackupExec was the faulting application which was stopping the snapshot from being released - it seemed that if the backup was still occuring whilst the snapshot was still open, and the script was closing it, it would go wrong, even though BackupExec had no business reading from that snapshot mount anyway!

The upshot of this now is that we have consistant backups of Scalix which are running and taking just over 7 hours to complete. I can cope with this.

We initially were going to put this on a Plasmon FC RAID controller, but we had serious issues with this model. Furthermore, there was no real tools that they provided for doing snapshots or live replication between the arrays, like I have seen on other systems, which is why I went down the local storage route.

We are using SATA with just under 200 users now on RAID-10 in a Dell PE2950 server, and we don't really have any IO issues.

Andrew.


Return to “Scalix Server”



Who is online

Users browsing this forum: No registered users and 1 guest

cron