In college I never had a backup for any of my computers. I knew about backups. I knew I should have one — especially given I was on Windows for the first couple of years. But I never had one. I did constantly reformat my computer (Windows), so I had backups of my working files and pictures, but that was about it.
Which really sucked, because a couple years into college my computer was stolen. Amazingly I didn’t really lose much. I lost some images, and some files, but not a whole ton of stuff.
And yet, when I got a new computer a few days later, I again didn’t backup. It wasn’t until I was out of college and getting heavier into photography when I started to back up. And then I started to get really into the nerdy side of backing things up.
Rotating cloned, bootable, drives. Offset backups. Synchronized files. If it was a good practice, I was probably doing it. And it took a lot of time to back up all that stuff.
Taking a lot of time to back things up is great, and when you don’t have a busy life you can take a couple hours on a Saturday and just back shit up. No big deal.
But when you have a house to keep maintained, a job, two kids, and a need to relax — backing shit up becomes a burden on you. So I started to let my backups slip in a big way about a year ago.
My bootable clone drives dropped to one, and it was usually a month or more out of date instead of just 24 hours out of date. I kept multiple backups of my images, but only spread across drives in my home. I still synced my primary files, but that was all the real time back up I had.
And I started to let those things slip a bit more.
Until I realized two things:
- I need to get a solid backup solution in place.
- I am way to lazy to make this anything less than automated. Even plugging in a drive is too much to ask.
The problem is there is usually no good way in which to test if alternative backup solutions actually work. With a cloned drive you can boot from the drive and test out your backup. But if you are doing something far more off the wall, what do you do?
Well for me I bought a new computer to replace my old computer. I then decided to pretend as though my old computer didn’t exist, and set up my new Mac to see what I was missing from my alternative backup system. It took a lot longer than restoring from a clone, as it took about a full day for all the data to sync over.
Having said that, I was up and running faster than if I had restored from a cloned drive, just without every file I had. In the end there was only 4 items which I had not backed up properly, and had to be manually removed from my old Mac.
In order to setup this system I had to figure out what I really needed to back up in order to start fresh on a new machine, and I pared that list down to four categories: photos, documents, apps, and miscellaneous things. Thankfully I’ve had a chance to test this out.
My photo backup system is the easiest of the lot to accomplish. I have two separate spots I keep my photos in: loose files, and iCloud.
I use iCloud Photo Library for basically every picture I take these days. The only way I back up these photos is by having my Mac mini server grab the full files (not optimized storage) and therefore I should have a rough back up in place outside of trusting Apple. That’s the easiest and most seamless system I could find since rt most photos on my iPad to begin with, and I honestly love this system.
The next task was backing up 300GB of images from before I was using iCloud Photo Library. I took these and exported them all out of Lightroom and Aperture to folders. I then took those folders and had Hazel organize all the files by dates in folders. From there I moved them all on to a 3TB hard drive. I then used Super Duper to clone that 3TB hard drive to another 3TB drive. I then shipped one of the drives to Macminicolo.net and had them plug it into my server.
Next, I opened BitTorrent Sync and told it my drive at home should sync in full with the drive on the Mac mini. After it verified everything was in sync, I was in business. Those drives never really get updated, but should I ever update them they will also automatically back up to each other (which ever one is updated), meaning I can edit files on either one, and still have my backup in place.
There’s a lot of other ways to back up photos and they are likely far more robust than what I am using. But at the end of the day this is dead simple. And dead simple is all that works for me.
This was the easiest of all to back up. Since I already have a robust system setup with BitTorrent Sync, I just started syncing my working folders (where ever they are on my Mac) to my Mac mini server and called it a day.
This is basicallylly the “private Dropbox” type of a solution. Where my Mac isn’t backing up, but syncing files to another Mac, which is then backed up.
I took stock of each app that was not purchased from the App Store when I setup my system the last time. I then got all the serial numbers for those apps outside of the App Store, and put them in 1Password. This way I don’t have to waste a ton of time requesting the serial numbers on next setup. The on going maintenance is entering any new serial numbers in as I purchase them. More burden than I would like, but not too bad overall.
Either way I can usually recover these serials if I forget in my email, or from the software vendor themselves.
There were a few other items I needed to backup, which I didn’t initially have covered, so I had to create solutions for them.
- Keyboard Maestro Macros: I now have the library auto exporting and backing up with BitTorrent Sync.
- Fonts: BitTorrent Sync.
- System Settings and Configs: I found a bunch these little things (like bash profiles) and started to back them up to GitHub. I don’t know why I went with GitHub, but I am sure I had some reason at the time.
Like I said, my system doesn’t miss too much.
How Well Does it Work?
I’m not entirely sure if I am honest. By that I mean I have not tested out the system since tweaking it more. I know that it pretty much worked fine when I setup my MacBook, but that was months ago.
I will say that it is slow to be downloading all that data and you have to be willing to choke your network and machine performance for a day or two.
The benefit is this system requires no interference by me. Everything is done automatically without me having to worry about connecting any drives. A lot of people already use Dropbox in a similar manner, or use Backblaze and like service, but those all have your data leaving your control. All my data is within my control, mostly.
My data is in a secure data center, with limited access to my hardware. It is encrypted. It is accessible to me whenever I want. I don’t have to run any backup software on my main machine.
Yes, my server runs all sorts of backups, I’m not asking for doom here. At the same time though, it all runs automatically since the drives are always plugged into it.
This all jumped into my brain when I was trying to explain to my father how I just don’t use traditional backup software anymore. That’s not to say you should dump cloned drives, it’s just to say that I am incredibly lazy.
Cloning my machine wasn’t happening anymore, nor was Time Machine backups. My system wasn’t created because I wanted something better, it was created because what I was doing was effectively nothing and I wanted to have something in place.
I think I covered all my bases. I think everything is being backed up the best it can be. I’m also fairly certain that not everything is being backed up. And that’s ok, because the irreplaceable images are being backed up. The documents I really need are backed up. Everything else is just an annoyance to recreate.
If you are wondering why I don’t use a service like Backblaze, it is simply because I don’t want me data somewhere that far out of my control. But it would be the even better lazy man solution.