Introduction
rsnapshot is a handy tool to manage backups using rsync and hard links on the
filesystem. rsnapshot will copy folders and files but it will skip duplication
over backups using hard links for files which has not changed.
This kinda create snapshots of your folders you want to backup, only using
rsync, it’s very efficient and easy to use, and getting files from backups is
really easy as they are stored as files under the rsnapshot backup.
Installation
Installing rsnapshot is very easy, on most systems it will be in your official
repository packages.
To install it on OpenBSD: pkg_add rsnapshot (as root)
Configuration
Now you may want to configure it, in OpenBSD you will find a template in
/etc/rsnapshot.conf that you can edit for your needs (you can make a backup
of it first if you want to start over). As it’s stated in big (as big as it can
be displayed in a terminal) letters at the top of the configuration sample
file, you will see that things must be separated by TABS and not spaces. I’ve
made the mistakes more than once, don’t forget using tabs.
I won’t explain all options, but only the most importants.
The variable snapshot_root is where you want to store the backups. Don’t put
that directory in a directory you will backup (that will end into an infinite
loop)
The variable backup is for telling rsnapshot what you want to backup from
your system to which directory inside snapshot_root
Here are a few examples:
backup /home/solene/ myfiles/
backup /home/shera/Documents shera_files/
backup /home/shera/Music shera_files/
backup /etc/ etc/
backup /var/ var/ exclude=logs/*
Be careful when using ending slashes to paths, it works the same as with rsync.
/home/solene/ means that into target directory, it will contains the content
of /home/solene/ while /home/solene will copy the folder solene within the
target directory, so you end up with target_directory/solene/the_files_here.
The variables retain are very important, this will define how rsnapshot keep
your data. In the example you will see alpha, beta, gamma but it could be hour,
day, week or foo and bar. It’s only a name that will be used by rsnapshot to
name your backups and also that you will use to tell rsnapshot which kind of
backup to do. Now, I must explain how rsnapshot actually work.
How it work
Let’s go for a straighforward configuration. We want a backup every hour on the
last 24h, a backup every day for the past 7 days and 3 manuals backup that we
start manually.
We will have this in our rsnapshot configuration
retain hourly 24
retain daily 7
retain manual 3
but how does rsnapshot know how to do what? The answer is that it doesn’t.
In root user crontab, you will have to add something like this:
# run rsnapshot every hour at 0 minutes
0 * * * * rsnapshot hourly
# run rsnapshot every day at 4 hours 0 minutes
0 4 * * * rsnapshot daily
and then, when you want to do a manual backup, just start rsnapshot manual
Every time you run rsnapshot for a “kind” of backup, the last version will be
named in the rsnapshoot root directory like hourly.0 and every backups will be
shifted by one. The directory getting a number higher than the number in the
retain line will be deleted.
New to crontab?
If you never used crontab, I will share two important things to know about it.
Use MAILTO=“” if you don’t want to receive every output generated from scripts
started by cron.
Use a PATH containing /usr/local/bin/ in it because in the default cron PATH it
is not present. Instead of setting PATH you can also using full binary paths
into the crontab, like /usr/local/bin/rsnapshot daily
You can edit the current user crontab with the command crontab -e.
Your crontab may then look like:
PATH=/bin:/sbin:/usr/bin:/usr/sbin:/usr/X11R6/bin:/usr/local/bin:/usr/local/sbin
MAILTO=""
# comments are allowed in crontab
# run rsnapshot every hour at 0 minutes
0 * * * * rsnapshot hourly
# run rsnapshot every day at 4 hours 0 minutes
0 4 * * * rsnapshot daily