Craig-
(just returning from vacation, so pardon the late response)
The issue seems to have been caused by an odd 'out of inodes' issue caused by the NFS storage appliance we use for our storage. After rebooting the storage appliance, the inode issue went away, and we've had no further instances of this issue.
However, in the meantime, a number of 'orphaned' backup directories were left behind. They exist, but are not referenced by the .../BackupPC/pc/[system]/backups file.
For example:
[backuppc@isift090 isemt101]$ pwd
/mnt/backups/BackupPC/pc/isemt101
[backuppc@isift090 isemt101]$ ls -ld 10
drwxr-x---. 5 backuppc backuppc 4096 May 31 23:05 10
[backuppc@isift090 isemt101]$ ls 10
attrib_cf06051de16efa13986a642e2dff5f6c backupInfo f%2f inode refCnt
File/directory ownership is correct, and it looks like an orphaned backup. So far, I've found that just deleting the orphaned directories seems like the best course of action, but I had to write a small script to help find them.
cd /mnt/backups/BackupPC/pc
for dir in $(for i in $(ls); do ls -d $i/[0-9]*| sort -n ; done) ; do
echo -n "$dir:"
host=$(echo $dir |cut -f 1 -d '/')
backup=$(echo $dir |cut -f 2 -d '/')
#echo $host $backup
if grep -q ^$backup $host/backups; then
echo FOUND
else
echo -n "MISSING "
ls -l $host/XferLOG.$backup.z
rm -rf $host/$backup
rm $host/XferLOG.$backup.z
fi
done
