Craig Barratt via BackupPC-users wrote at about 19:52:48 -0700 on Tuesday, October 13, 2020: > The BackupPC 3.x store has large numbers of hardlinks. What commands are > you using to migrate the data? > > You need to copy the file system in a manner that is aware of, and > reconstructs, the hardlinks (eg, rsync -aH, or a raw file system copy > provided they are compatible). Copying all the hardlinks will be slow, but > will work. > > Another reasonable option is to start over, since you can keep the old disk > around until you are comfortable you have sufficient history in the new > server. >
Two other possible solutions... 1. Many years ago, I wrote a program, BackupPC_copyPcPool.pl, to copy over v3 databases by first copying over the cpool (without worrying about hard links) and then running the program to create a text file list of file/directory names and inode numbers representing the pc tree. Then on the new machine, run the program again to re-create the pc tree and the hard links. This is still a slow process if you have lots of backups and thus lots of hard links. 2. Use this as an opportunity to move to v4. Specifically, migrate to v4 on your old machine, which basically gets rid of the hard links, then you can use regular cp or rsync to copy over the v4 migrated pool and cpool to the new machine. Not sure how much extra disk space is needed during the migration which could be a problem if you are short on disk space. CRAIG: Would it make sense to add the ability for BackupPC_migrateV3toV4 to migrate to a different filesystem than the current v3 location. I think this should be pretty simple to do given that both the pool and pc trees are re-written. _______________________________________________ BackupPC-users mailing list BackupPC-users@lists.sourceforge.net List: https://lists.sourceforge.net/lists/listinfo/backuppc-users Wiki: https://github.com/backuppc/backuppc/wiki Project: https://backuppc.github.io/backuppc/