Re: [rdiff-backup-users] 2nd Backup fails when using 2 remote targets.

[Finn-Arne Johansen]

Elias Lopez wrote:
> Hi,
>
> I can't seem to figure out why this happens. It works the first time (no 
> backup) but the second time round it fails with:
>
> AssertionError: CCPP index out of order
>
> However, if I use the a local directory for the destination target, all 
> appears well.
>
> I'm using rdiff-backup 0.13.4 from debian backports, and python2.2.
>
> carlton# rdiff-backup --print-statistics --ssh-no-compression 
> northcote::/home port-melbourne::/rsync/rdiff/northcote/
> home
> ......
> Traceback (most recent call last):
> ..........
> AssertionError: CCPP index out of order: ('claire', '.kde', 'share', 
> 'config', 'kdeglobals') ('valerie', 'work', 'staffdevelopment', 'icams', 
> 'src', 'gcode', 'aos', 'demo', 'icams', 'web', 'CaptureBadDataEntry.gplan')
> ......

Seems very familiar. how fast are you doing this. I got the same error I 
think, when I tired to convert an old backup that used rsync and 
hardlinks to rdiff-backup. my script did something like:
 for DATE in $(ssh backup.old ls /backup) ; do
  rdiff-backup --current-time $DATE --print-statistics \
               backup.old::/backup/$DATE backup.new::/backup
 done

with each of the old backup sets sorted like
 /backup/
   2004-08-01/
   2004-09-01/
   ...
   2004-11-02/
   2004-11-09/
   2004-11-14/
   2004-11-15/

I had to something with the date, but it kind of worked anyway. I think 
i thought it was some temporary error, cause if I did things manually it 
kind of worked. But then there could be hours between each time I ran 
the new commands. I changed, and ran rdiff from backup.old instead, and 
then everythiong went fine.
Have you checked that there is no file left in /tmp on either carlton or 
 port-melbourne ? (or even northcore)

// faj