Backing up to Amazon S3 using Amanda
Recently a friend of mine (Sarah) purchased a dedicated server at a hosting company and moved her data to it. Obviously she was in need of a backup solution. She chose Amanda 2.6.0 and to use S3 as her “tape” choice.
Seeing how well it worked for her I asked her to show me how it worked and now my server is also using Amanda to backup to S3. Recovery works with amrecover.
I have quite a bit of data to back up (4.5GB in /home, for example) and with a home DSL connection it takes a long time. However with a bigger pipe using S3 and Amanda would be extremely viable.
I’m very happy with the solution and will likely use it across all of my servers from now on.
For my laptop I’m still using JungleDisk, which seems to work fine. 2.0 is a very good improvement.
I’m looking at doing an S3 backup solution for my Linux and M$ PCs. Amanda and JungleDisk are my probable finalists. Why did you choose Amanda for servers and JD for your laptop? Amanda and JD can both handle server and client backups. Do you feel Amanda is better suited for automated server backups while JD is better suited for mobile PC backups? Do you use compatible buckets with JD? The only negatives I have for JD are 1) By default, JD buckets aren’t compatible with other S3 access methods (s3sync, s3fs, s3fox, etc), 2) Not OSS, 3) JD+ relies on intermediary servers for block level updates.
Comment by Jim Sifferle — July 3, 2008 @ 1:01 am
Hi Jim,
I chose Amanda for my server because it’s always on. Jungledisk for my laptop because it isn’t always on. Amanda runs from cron and I may not always have my laptop on when it would trigger. I’ve configured Jungledisk 2.0 to run when it gets a chance following a missed backup (Normal backups are scheduled for 16:00 daily).
If I had a desktop that was always on and in the same place as the server I would backup the desktop over Samba.
Amanda uses its own buckets (to represent tapes) which can be mounted through Jungledisk’s interface (e.g., exposed over webdav). I wouldn’t do access with it, though.
Jungledisk has its own bucket for use. There’s no reason to share - buckets don’t cost extra.
JD+ doesn’t matter because all of my data is encrypted before it leaves my system (same for Amanda). I don’t need to directly access my files with another non-jungledisk program because I use jungledisk to access them. I’m not tied to open source software.
Both solutions are very good at what they do. Amanda is very cool and I hope Jungledisk can incorporate some of its scheduling and backup mechanisms.
Comment by Lisa Seelye — July 3, 2008 @ 1:29 am
[...] up to Backing up to Amazon S3 using Amanda (June 28, [...]
Pingback by CrudVision - Lisa Seelye » Following up JungleDisk vs Amanda — September 2, 2008 @ 9:01 pm