Back Up Heaven:� Achieving Automation

Backupheaven-1 The more you get into back up, the more you begin to appreciate it; but however much you are glad you do it, it is still a pain in the arse when it takes forever to do. The solution to it becoming a thief of your time, or something you give up on because it’s just too much effort, is to automate it as far as you are able.

Part of the automation equation is using the correct software, and part of it is configuring your hardware, but a critical component is order in your file-system.

The software part is easy to fix – pick software that allows scheduling, and operates with minimal user input after initial configuration: for my part, I use Chronosync, which has scheduling features, for syncing between computers, and backing up to USB data drives; I use Transmit, which has minimal-click operation for on-line storage syncing; and I use Retrospect for cloning, which again allows scheduling up to a point, and requires only a couple of clicks to set things going manually.

The hardware can be tricky. If you are a Mac user, tricky doesn’t come into it, obviously. If you are a Mac user afflicted with Windows machines on your network you are going to have to let your Windows boxes see your Macs if you want to include them in the process. If you are a Windows user I offer you condolences and wish you good luck .

Seriously, though, your goal is to set up permissions and connectivity so that you can run processes without intervention, and your machines can see each other without your firewall spitting its dummy. This could be a 5 minute job or a 5 day job with calls to techies around the globe and frequent visits to support forums via Google – it seems there is really no telling; unless of course you buy a Mac – then it’s easy

Finally we get to the critical part. No amount of funky software or slickly configured hardware will cope with badly structured data: If your data is all over the place, automation is out of the window. What you are looking to do is to select one data location on one side (the primary data source) and one data location on the other side (the secondary data source or destination, depending on what you are doing). You are going to tell your software to take data source 2, and do whatever you have configured the software to do to it in relation to data source 1 (mirror sync, bi-directional sync, overwrite etc).

I achieved this as follows:

On my computer. I have a top-level directory – rather like a filing cabinet in an office, if you want an analogy. I called this directory ‘Cabinet’. ‘Cabinet’ has second level directories – data, static, local, applications, entertainment – which are rather like specific drawers in that Filing Cabinet analogy. Each second level directory has sub directories, which are, if you like, the hanging dividers in that filing cabinet drawer. In those directories there may be loose files or further sub directories. But, basically, my ‘hanging dividers’ are where I keep my hard data.

On each back up location I first set up a directory called ‘Cabinet’ and then simply mirrored the contents. Now, when I sync, I select ‘Cabinet’ on each source/destination and hit Sync. Or, if I am scheduling, I let the ‘hit sync’ part happen in the background.

It may take a while to set up – moving data around to get it ship-shape, but if automated back-up is your goal, there is no other way.

Armed with the right software, the right hardware configuration and a logical file structure, you are all set for a lifetime of painless backup.

Share this on