I really wanted to like BitTorrent Sync when it came out, but after it came out and hearing that it was going to be closed source turned me away from it. Even before BTSync was a thing I was using Dropbox for the longest time. I would always get close to my limit and figure out a way to move data around (deleting old data) to fit new data. Upgrading wasn't something that I wanted to do because I only needed 20 Gb of cloud storage, so paying for 100 Gb was a waste for me, especially since I was in college. This resulted in a lot of data just getting pushed into other cloud storage providers with little organization. This couldn't last…
Google Drive came out and I signed up for it as soon as I could. It solved my data problems for a little bit, but lacked one major thing, didn't have good Linux support. Didn't use this for much of anything other than for scanning documents for my paperless office. Reminder I should write a blog post on my paperless workflow.
Then came OneDrive. It worked well for Windows, and there were hacks for Linux. They gave out 30 Gb for free which solved my data problems for a little, but since it wasn't easy to use on all three major OS platforms it didn't get much use.
I learned about Syncthing relatively recently. Caught my eye because it was a Golang program which allowed it to run on all platforms easily. WebUI allowed me to easily configure it for headless machines, no clunky GUI involved. After trying it out with a couple computers I've been won over. Dropbox was gone, almost.
With so much more control over the sharing that with Dropbox and the fact that it isn't federated it was challenging to try to figure out the best folder structure. First mistake I made, and still paying for, is dedicating a folder for a specific use and sharing that folder only. I have a
3dprinting folder that I use for syncing up all the
.gcode files for my 3D printers. This works nicely because not all my computers need and want to share those files. What I should have done is used the default share that Syncthing provides to share everything with, very Dropbox-like, then slowly move to individual per use folders for more specific things.
My current workflow:
My server will house all the shared folders. This makes it easy to have new machines subscribe to folders even if the originating host is down.
BTW, the documentation is pretty awesome and explains all the cool features.
My next test will be testing syncing outside of my office.