16
02
2012
With that mess complete, I could now SSH into my Mac Mini from the office using a command like the one below (replacing the X’s with your Time Capsule IP address):
I could now complete a successful rsync sync for the first time as well (command is all one line):
rsync -avz --progress /Users/jv/Desktop/forotherbooks/ jvmini@XXX.XX.X.X:/Users/jmini/Desktop/fromotherbooks Step 2: Public Key Authentication
Hooray! Since I knew I wanted to end up being able to do this automatically, entering the password each time wasn’t going to be an option. We use public key authentication for a few things in the office, so I had an idea of getting that set up, but if you are new to that game, check out this guide. Here’s the run-down of the commands used:
ssh-keygen -t dsa
scp ~/.ssh/id_dsa.pub jvmini@XXX.XX.X.X:.ssh/authorized_keys
ssh jvmini
Step 3: Enter the Cron
Now I had sync set up and working, but the automated part was still missing. For that, I needed cron. Cron is a UNIX tool for scheduling jobs – pretty cool if you want to automate stuff and have it run on schedule. I didn’t have any experience with cron at all, so it took me a little while to get a handle on it. Basically, there are two parts: the script and the cron table.
The script I wrote wasn’t too complicated – essentially sync everything up, and the delete all the old files from the local directory.:
#!/bin/bash
RSYNC=/usr/bin/rsync
SSH=ssh
LPATH=/Users/jeff/Desktop/otherbooks/
RPATH=jeffmini:/Users/jeffmini/Desktop/fromotherbooks
echo "Syncing and it feels so good"
$RSYNC -avz --progress $LPATH $RPATH >> /tmp/output.txt
cd $LPATH
rm *
I saved this under the title sync_books.sh and saved it for now in my Home directory.
Next was setting up the cron table. Cron table is a config file where your list of jobs are for executing. This step was a bit tricky, since the normal command
crontab -e
Wasn’t working properly for whatever reason. I blame Apple but that’s ok. I found a workaround, which is to use the editor Nano, rather than VIM, to edit the crontab file. In retrospect, this makes sense, since Apple has been tried deprecating the use of cron in favor of something called launchd. Anyway, here’s the command for opening the crontable in Nano:
EDITOR=nano crontab -e
This opened up the cron table, so I could add a new line to it. Cron tables (or at least my limited understanding of them) have syntax like so:
* * * * * command to be executed
The asterisks in this case stand for [min (0-59), hour (0-23), day (1-31), month (1-12), day of week (0-7, 0 and 7 are Sunday). A great tutorial for Cron table syntax is over at Admins Choice.
Step 4: Testing
Since I was setting up the cron job to run once a week (2am Sundays), I didn’t want to wait that long to make sure it worked. A one-line addition to my cron table file ran the cron script immediately:
*/1 * * * * /Users/jv/sync_books.sh
And then running a tail command in the terminal, to track updates to that “output.txt” file:
tail -f output.txt
And a minute later, voila! Syncing was complete! Now I can dump all the stuff I want to access on the Mac Mini (or archive) without clogging the network, or dumping it into the Time Machine abyss. Enjoy, and let me know if you have any questions!
