rdiff-backup bug or error? Disk quota exceeded but that is not true

Question

I want to make a backup with rdiff-backup.

I have a backup VPS and I want to get the datas from remote host.

Here it is my command what I used:

rdiff-backup root@**RemoteIP**::/remotesourcefolder/ /targetlocalfolder/ 

After some seconds, they throw some error and stopped with this;

OSError: [Errno 122] Disk quota exceeded: '/demo/rdiff-backup.tmp.5' [root@mn6 ~]# Fatal Error: Lost connection to the remote system 

.

Before you write anything, I have a bounch of free space, and the source folder under 50 MB...

df -h Filesystem Size Used Avail Use% Mounted on /dev/simfs 1.0T 137G 888G 14% / none 512M 4.0K 512M 1% /dev none 512M 0 512M 0% /dev/shm 

What I'm doing wrong? (Python version: 2.6.6, rdiff-backup: 1.2.8)

Here is the full raw output: http://pastebin.com/raw/Bs4VRdwX

Answer 1

OK, there is some space on the server (I'm assuming your "df -h" was executed on the backup server, not on the client) but maybe you have no free inode on the server (i.e. the maximum number of files has been reached on this FS). This is unlikely but you can check this with:

df -i 

Else, have you checked your quotas on the server?

quota 

An other reason could be that your rdiff-backup command, on the server side, is limited, especially if this server is shared with several users. It can be limited with the number of open files. Check this with:

ulimit -a 

There may be other limitations you don't see on a shared platform (e.g. cgroup). But first try the above commands.

[edit] As a last resort, you should contact your provider. See this answer (your provinder is using simfs too).