I'm spread across three computers at the moment. This was not
planned, nor is it desirable. But they're all serving a purpose
here at home, so I find myself jumping from machine to machine.
The situation with my password safe, personal wiki, and
so-called 'dot files' being out of sync is getting really
painful. (I've used rsync and scp, but I've also _emailed_ files
to myself on occasion.)
So I finally installed Syncthing today.
Wow, what a great piece of software.
Sure, I'll have to live with it for a couple of months before I
can give a proper assesment, but, I'm just really impressed so
far.
The security/discovery model for connecting computers is
brilliant: each machine gets an extremely long, random ID. To
connect two machines requires that _both_ of them are explicitly
instructed to connect to the other one.
The ID of a machine is useless to a hacker because it only
allows them to _allow you_ to connect to them. :-)
Now, having a ~/sync directory that updates between two of these
machines has lit a fire under me to automate some of my setup.
Like: script the symbolic linking of .bashrc to my new
sync/dotfiles/bashrc. This is stuff I've intended to do for a
long time, but hadn't gotten around to it.
Of course, I've been breaking stuff left and right tonight.
In fact, I've only just now recovered enough functionality to be
able to post this. :-)
It reminds me of all of the lessons I learned when I played
DevOps a couple years ago and scripted the provisioning of a web
server for a software project: when you automate things, you
find all of the "stuff" that has fallen through the cracks,
forgotten and undocumented, but vital for functionality you
depend on.
Once I hook up the third machine to sync with these two, I'll
see what has fallen through the cracks.
Oh, the other interesting thing is deciding what should be
synced and what shouldn't be. I'm up to 13Mb so far.
