Re: [rdiff-backup-users] Problem

[Robert Knighten]

Paul P Komkoff Jr writes:
 > Replying to Robert Knighten:
 > > 
 > > I am in the process of re-establishing an rdiff-backup regime after some
 > > recent problems and upgrades of my computer systems.  The situation here is
 > > backing up from on Linux system (Linux version 2.6.0-test9 
 > > (address@hidden) (gcc
 > > version 3.2.3) #5 SMP Mon Nov 10 19:25:25 PST 2003) to another (Linux 
 > > version
 > > 2.4.18 (address@hidden) (gcc version 2.95.3 20010315 (release)) #10 Fri 
 > > May 31
 > > 01:36:52 PDT 2002).  Both are running rdiff-backup 0.12.5 and Python 2.3.2.
 > > The first problem comes in the case where the backup directory is actually 
 > > NFS
 > > mounted on the system which is being backed up.  When creating a brand new
 > > backup directory using the command
 > > 
 > > /usr/local/bin/rdiff-backup --exclude /cdrom --exclude /DVD --exclude 
 > > /floppy --exclude /SYSBACK --exclude /mnt --exclude /proc --exclude 
 > > /home/carolm --exclude /home/shared  --exclude /usr/local/src --exclude 
 > > /home/knighten/HADES --exclude '**.netscape.*' --exclude '**NO_BACKUP' / 
 > > gateway::/home/shared/Zeus
 > > 
 > > I got the following traceback when it appears that the backup was 
 > > essentially
 > > complete.
 > 
 > Try searching your dmesg for "Stale NFS handle" messages.
 > 
 > -- 
 > Paul P 'Stingray' Komkoff Jr // http://stingr.net/key <- my pgp key
 >  This message represents the official view of the voices in my head

No "Stale NFS handle" messages.  The only suspect item is the infamous "nfs
warning: mount version older than kernel" which results from my having to use
the 2.6.0-test9 kernel version to get driver support for the serial ATA drives
I am using.  mount doesn't actually appear to have changed, but might this
warning actually be causing the problem?

-- Bob