[rdiff-backup-users] rdiff-backup fails

[mcgregor]

Dear all,

I've noticed that some of our rdiff-backup jobs are failing, while others just 
work. I have attached the most recent rdiff-backup output in the hope that 
someone can shed some light on what is going on. The log implies a previous 
backup has failed but unfortunately I do not have access to the log when that 
happened (nothing in the rdiff-backup-data dir looks like it reports any 
serious errors - the logs just stop).

I'm running rdiff-backup version 0.12.6 with python 2.2 and librsync v0.9.6 . 

In the first log it does mention insufficient disk space, and I've read some 
posts about this in the archives. I think the system is running out of temp 
space, so just to clarify, does rdiff-backup require temp space on the source, 
destination, or both? (And how much?)

The second log is a mystery to me.

The two backups which fail are the largest (150G and 50G respectively). I doubt 
this matters, but these backups are run using cron (excerpt shown below):

0  1 * * * root /usr/bin/rdiff-backup /usr/cfengine 
sequoia0::/home/backup/needles/usr/cfengine
10 1 * * * root /usr/bin/rdiff-backup /usr/people 
sequoia0::/home/backup/needles/people   <- THIS ONE FAILS
30 1 * * * root /usr/bin/rdiff-backup /usr/boone 
sequoia0::/home/backup/needles/boone
45 1 * * * root /usr/bin/rdiff-backup /root sequoia0::/home/backup/needles/root
0  2 * * * root /usr/bin/rdiff-backup /var/yp 
sequoia0::/home/backup/needles/var/yp

And I'm wondering what happens if one rdiff-backup starts running before the 
previous finishes. Would this cause a problem?

Any ideas are welcome. For the moment (and historically) I am just starting a 
fresh backup destination. This seems to work.

Gordon McGregor.

rdiff-fails
Description: Binary data