Amahi Forums

Hi guys,

I'm about to rebuild my Amahi server into a new 4U rack-mounting case. Currently, it's in a 2U box, but I've ran out of drive space (well, drive bays) so need to expand.

I'm using Crashplan to back up my desktop "working drives" and "archive drive" to a single 2Tb volume in the Amahi box (let's call it Backup Master).

In my "rebuild" I want to bring into the equation a duplicate copy of that 2Tb drive so there is always a fall-back, but also maintain an off-site copy.

What I plan to do when rebuilding is install some removable drive caddies.

My plan is that Desktop backs up to Backup Master. Backup Master is duplicated to Backup Copy A, via Crashplan. Once a week, I remove Backup Copy A, and replace it with Backup Copy B, which will then be updated. Backup Copy A gets stored off-site for a week. Then the following week, I repeat the procedure, replacing B, with A. Basically always ensuring I have one copy off-site that is no more than a week out of date.

Can anyone advise on the best way to set this up? I'm a Linux newbie, so the mounting of drives/hot-swapping etc is a bit confusing, so want to make sure I can actually do this before committing to trying it and failing miserably!

It's absolutely critical for me to maintain an off-site copy - the data stored on these drives is something in the region of 250,000 digital photographs (I'm a photographer who works freelance from home).

I should also mention that I don't use greyhole - I don't want storage pooling in the setup I currently have. The backup drives are EXT3 format. My desktop and laptops are all Win7 machines, only the Amahi box is linux.

Thanks in advance for any advice, suggestions - or if I'm using the wrong app, or methods - please, let me know!

You should check with the Crashplan forum about your idea of actually swapping drives like that. Given the way that Crashplan works, I really don't think that will work. I don't think you can remove and install drives like that without messing up Crashplan's database strategy.

Are you aware that you can have Crashplan backup your files to a 'friends' remote computer? You could set up a computer at a remote site with Crashplan on it where you could store an additional copy of your backups. This way your offsite backup would be current rather than a week old.

Definitely consult with Crashplan about your idea.

Good plan there. I will look at it this evening.

I think it should be OK - I am currently "missing" one of my desktop drives normally backed up by Crashplan. It emails me to tell me that it has been unable to reach the destination, but the backup itself is intact and restorable. (the missing desktop drive has gone back to be replaced under warranty).

I will check the idea out with Crashplan.

Part of the reason I want to use Crashplan, is also to sign up for Crashplan+, which would be a second option for backup - in an ideal world, I can restore 95% of my files from my "hard copy", but if the backup is synced to the online service, I can restore the last week of data from Crashplan+ I know full well the astronomical time it would take to initially backup, and restore fully from the online solution - so the local drives should take care of the bulk of that.

(if that makes sense)

Let us know how it works out for you.

I currently use Crashplan to backup some of my systems to each other. I've been really happy with it so far - though I haven't had to a major recovery yet.

My main concern here is setting up the hot-swap drives. I struggle to get my head round the ways to mount drives in linux, and it always takes me an age to set up a new drive (I rarely ever tinkered with linux previously) - and I'm just a little bit concerned incase I find something isn't compatible/working etc.

Anyone any advice or guidelines on how to set up and manage hot-swap drives? Or am I worrying over nothing?!

As for Crashplan, I'm happy apart from one issue - I have my main 2TB volume for storing my current backups on. I set crashplan up to back up a whole 320GB drive to this backup volume. The default location is the 2TB drive, and it started the backup, only for me to discover that it's used a different drive altogether, and the backup is incomplete because the destination is out of space.

I now need to find a way of moving that partial backup to the 2TB where it's meant to be, and resuming it, otherwise I'll be starting again.