From: Long V <lon...@ya...> - 2011-12-01 20:21:50
|
Hi, I have set BackupPC to backup a directory where the content is always wiped and re-populated with 99% of the time same filename and same content inside the files. File timestamps would be different as they have been re-generated. I expected that BackupPC would detect it's the same filename and same content and use hardlink instead of re-copying the file again. However, I've been monitoring /var/lib/BackupPC/cpool using `du -sh /var/lib/BackupPC/cpool` and it seems to be growing at a steady state that is much faster than what I expected. Either I am not looking at the right pool or the way I measure the pool size (du -sh) is wrong or BackupPC is not using hardlink in my case. To confirm my suspicions: - is /var/lib/BackupPC/cpool the pool I need to look for - is `du -sh` the proper way to measure the size of the pool - how do I know how many hardlinks are pointing to the same file in the pool? - how do I calculate the saving ratio of using hardlink vs simple copy the file again? (this would confirm if my instance of BackupPC is exploiting the saving advantage of hardlink or not) Thanks, Long |