Hey,
I've arranged my system so that i've got a disk of 320GB for my OS (linuxmce), and a disk of 1TB for all data.
What's the best way so all data is arriving on disk B?
Reason for this is to split my data from my system. Like this, i can take a complete images (without any user data) of disk A.
And later on, i would like to add a third disk of 1TB so i can mirror disk B.
Suggestions, tips, help is more then welcome!!!
QuoteWhat's the best way so all data is arriving on disk B?
Simple..... make sure it always has more free space than your os drive.
Linuxmce is smart that way.
HTH
Tim
the only component that uses that strategy is mythtv, and you can force it to use a specific drive when you create recording profiles if you want.
otherwise when ripping a disc, you tell it where to save.
True, but Orbiter's File/Save screen will always ask for the disk with the most free and use that.
-Thom
Hmmm, not sure if i like that. :'(
This means that when my 1 TB disk has fe 75% in use, he'll start to use the primary disk (os)? Regarding he still has 250 GB free?
Is there a way to change this? So that my 1TB is always the primary disk for data? I'm just a bit afraid that after a while, i'll not be aware where my data is located. Some on disk A, some on disk B. So if i need to backup stuff (os or data), i need to backup each time everytime?
My idea is for the future:
- Data backup each day (mirroring, nightly copy, to define...)
- Config backup each week (dump db's, configfiles...)
- OS backup each month (or important changes)
brononi,
backup should not be an issue, if you backup /home and don't exclude softlink'd stuff.
And the Orbiter knows where your files are, so if you look through your media library using any sort order BUT filenames, you should be fine.
Great,
So only /home to backup all my data?
I will start looking for the best way to backup this to a third disk... Some time ago, i played around with rdiff-backup. Maybe time to put it in place. ;)
Rest only linuxMCE itself. But maybe i start best a new thread for that...
Quote from: brononi on December 26, 2011, 11:34:29 AM
So only /home to backup all my data?
Backing up /home will backup all of your media, if your backup strategy follows through on symlinks. Backing up the configuration is done using a different script, as outlined in the other thread.
Thanks,
Rdiff-backup is covering symbolic links. So this should do the trick.
ps would be nice to have a complete backup system integrated in linuxmce, no?
One where you can define what/where/when/how to backup... Maybe for the next release? ;)
brononi,
feel free to architecture one, and implement it. After doing it, create a trac ticket and attach a feature patch.
Thanks in advance.
Quote from: posde on December 26, 2011, 05:06:40 PMfeel free to architecture one, and implement it. After doing it, create a trac ticket and attach a feature patch.
Love to do it. If i knew how. :(
Any guidelines/howto's around that i can use for this?
I personally don't have any clue how i could do this. I suppose it's not enough with just installing rdiff and write a short manual... :$
Well, do all the steps that are needed, create needed configuration files, put your steps into a shell script (a regular text file you can execute [look at any of the .sh files in /usr/pluto/bin]). The most simple script looks like
#!/bin/bash
echo We start a backup
/usr/pluto/bin/PlutoRestorePoint --backup
Well basiclly, this is it (an old manual i wrote for myself):
Installation of rdiff-backup
sudo apt-get install rdiff-backup
Create a backup folder (can be also remote NAS drive, external disk...). You need to do this only once. Replace everything after 'backups/' with the stuff you want (your home folder, your data folders, your...).
sudo mkdir /backups
sudo mkdir /backups/home/me
sudo mkdir /backups/home/he
sudo mkdir /backups/home/she
sudo mkdir /backups/web
Put following into a script (fe /scripts/backupData.sh)
sudo rdiff-backup /home/me /backups/home/me
sudo rdiff-backup /home/he /backups/home/he
sudo rdiff-backup /home/she /backups/home/she
sudo rdiff-backup /var/www/html /backups/web
Run now the script, you can it also put in your cron...
But isn't this very limited to my configuration?
I suggest that we put something in the field that all users can easily adapt for their needs (sources, destinations, timeframes...).
I've just noticed that several 'GUI's' exist for rdiff.
More info: http://wiki.rdiff-backup.org/wiki/index.php/GraphicalUserInterfaces#Web-based
I don't know if we can implement something like this?
Or is this a lot of manipulation on the current webmin? I don't have a clue about this. I'm happy to put some time into it. If I now how/when/where...
Look into the existing shell scripts in /usr/pluto/bin, google for bash advanced, and learn a bit of bash till new years eve :) You will be surprised what YOU can do.