What I'm looking for is not a damn complex program, with servers+clients+daemons+agents+satellites (it's just for my box), but something that allows me to:
- have full + incremental backups;
- easy recover / exploration of the backups;
- allows different policies for different filesystems (for /home I want a retention policy of 1 year, while for /media 1 month is more than enough);
- efficiently store big files (like films, big archives, etc);
- it should recognize when a file is moved from a location to another, and avoid the del+cp again;
- possibly start when the external disk is connected, or only if it's there;
- possibly a nice gui/html frontend to allow for easy administration but NOT to schedule backup (it must run as deamon).
What do you have to suggest to me about it?
I , by myself, am considering backuppc, amanda and bacula, while the latter seems a lot complex.
I wait for your reply :)
20 comments:
try rsnapshot. AFAIK it doesn't have a GUI but it's fast and reliable.
ciao
Alessandro
I'm using Bacula to backup two machines - my own Debian box and my wife's WinXP laptop.
May not be exactly up to spec with your requirements, but it works for me - it saved my arse several times.
This comparison may be helpful.
Another vote for 'rsnapshot'.
Or perhaps 'sbackup'. On my sister-in-law's Ubuntu box, I've also been impressed with 'Simple Backup' (I think it's actual package name is sbackup).
I've used Bacula for a while as well, and it's great for big, complex systems/networks. I think it's overkill for a personal solution though.
I tried sbackup, but then I settled for backintime (http://www.le-web.org/back-in-time/ , but the webpage sucks).
It's basically an interface to rsync, but with way more options, nice GUI etc.
I'm very happy with backuppc:
BackupPC is disk based and not tape based. This particularity allows features not found in any other backup solution:
* Clever pooling scheme minimizes disk storage and disk I/O. Identical files across multiple backups of the same or different PC are stored only once resulting in substantial savings in disk storage and disk writes. Also known as "data deduplication".
* Optional compression provides additional reductions in storage. CPU impact of compression is low since only new files (those not already in the pool) need to be compressed.
* A powerful http/cgi user interface allows administrators to view log files,
configuration, current status and allows users to initiate and cancel backups and browse and restore files from backups very quickly.
* No client-side software is needed. On WinXX the smb protocol is used. On Linux or Unix clients, rsync or tar (over ssh/rsh/nfs) can be used
* Flexible restore options. Single files can be downloaded from any backup directly from the CGI interface. Zip or Tar archives for selected files or directories can also be downloaded from the CGI interface.
* BackupPC supports mobile environments where laptops are only intermittently connected to the network and have dynamic IP addresses (DHCP).
* Flexible configuration parameters allow multiple backups to be performed in parallel.
* and more to discover in the manual...
Please let us know what you find after evaluating these possibilities, Sandro.
Cheers.
I use Dirvish for backups. I don't know how it does with move detection and there's no graphics frontend (everything is done with config files), but it definitely satisfies the rest of your requirements.
Hard links are used to allow large numbers of incremental backups to be stored without having to copy the unmodified files.
Retention policies are quite flexible, and can be set based on filesystem and date/time of day (i.e. you can do things like have friday evening backups stay around for a month and monday noon backups stay around for a week).
Recover is easy as your backup directory ends up holding what are essentially a series of snapshots of the directories you've backed up. The hard-linked directory system also allows you to selectively remove part of all of an unneeded backup (if, say, you don't really need all three versions of that movie you made, but you want to keep the rest of those three backups) without effecting backups made before or after.
I like rdiff-backup. It's simple and with a little scripting could do most of what you want.
I second BackupPC. It is really nice and through the web interface can even be maintained from work :-)
I am also using rsnapshot for a scenario similar to yours.
A vote for rdiff-backup and a vote against rsnapshot: using a hard link tree for each increment does not scale. I'm in the middle of packaging "archfs" which is a FUSE-powered program for providing a filesystem view of an rdiff-backup backup. I'd recommend giving that a look too. Check out Ted T'so's blog, where he had a good set of comments on this subject.
I use "Simple Backup". It is simple. You can setup inclusions and exclusions. It runs on a schedule. You can tell it when to do incremental and full backups.
Pretty happy so far... though haven't had to restore from backup and that is always the big test.
I'm using rdiff-backup as well, but for complex setups it would be nice to have a gui or a configuration tool. I don't know how to remove specific incremental backups, for example.
Maybe you can try with a scheduled rsync!
maybe not the most user-friendly software,
but the most flexible, stable and secure.
You can find some scripts too on the net, like rsyncbackup. bye
rsnapshot is really great for personal backup.
On my laptop I use it with an external harddrive. I set it to a certain udev device and made a basic script that mounts it, backs up, and unmounts it whenever I plug it in.
I'ld recommend backup-ninja with rdiff-backup; I use it with LVM snapshots on a 2TB disk and it works great!
I've had decent success with backup2l, although I'm not entirely satisfied with it. Haven't found anything better yet, though.
I am having one Ubuntu server that is backing up files using backup2l. I am not very much fluent with backup2l.
However, it is taking too much spaces.
I would like to minimize the space.
Is it possible to backup only last one month's data using backup2l? Which means, after every month the previous data will be removed and current data with full backup will be there?
Any help will be highly appreciated.
M.Islam
tarazia@gmail.com
@Mithun: and you ask for assistance in a blog post where I ask for suggestions? bah...
Post a Comment