File server redundancy

[DrCheese]

Hi there,

 

How do you guys deal with making file servers redundant? We have a VMCluster with a large SAN and the file server runs on a VM.

 

However, naturally it's something that can't ever really be rebooted apart from late in the evenings & if it went down (for whatever reason) we'd lose it. It's not ever really been a problem but I was wondering if others have managed to make their's redundant.

 

I can do it via a cluster, or something like DFS but I'm not really sure how to proceed. Does it even really matter?

[synaesthesia]

Depends what you want to achieve. I like using DFS namespaces & replication as it means there can be a failure and noone need ever know.

[DrCheese]

Well, I'd like to be able to reboot the file server without anyone really noticing... Downside of DFS is that it's replicating the data so taking up twice the space on our SAN

[synaesthesia]

Indeed, it's not suitable for all. We use local storage on 2 servers in 2 different sites, so ideal here. You can offset some of that loss though if you're on 2012 by looking at de-duplication too.