Paul Fox wrote at about 09:58:42 -0500 on Thursday, December 15, 2022: > Stefan Helfer wrote: > > Hello Iosif, > > > > Hooray, I have figured it out. It took me a few hours and more > > than 60 full backups, but I found the cause. > > I admire your perseverance! > > > My understanding and conclusion is now: > > BackupPC v3 shows the size of the backup as it would be restored to > > disk. With all hardlinks intact and used. > > BackupPC v4, on the other hand, shows the size of the backup as if > > all the hardlinks had been resolved into individual standalone > > copies of the files. > This sounds like a bug in V4, to me.
Agree that seems like a bug in v4. Makes sense though that such a "bug" could occur since v4 and v3 treat hard links very differently. Basically, v3 treats the "first" (arbitrary what is first) hard link as the "real" one and then treats the others as hard links to that "real" one via the attrib file. While in v4 they are all treated symmetrically. Perhaps this is what causes v3 to count the storage only once while v4 counts each hard link instance... > > Just to be clear, we're talking about the size number reported here, > correct, on the "Host <foo> Backup Summary" page? > ----------------------------------- > File Size/Count Reuse Summary > > > > Existing files are those already in the pool; new files are those added > to > the pool. Empty files and SMB errors aren't counted in the reuse and new > > counts. > > > > +--------------------------------------------------------------------+ > > | | Totals |Existing Files | New Files | > > |------------+-----------------------+---------------+---------------| > > |Backup#|Type|#Files|Size/MiB|MiB/sec|#Files|Size/MiB|#Files|Size/MiB| > > |-------+----+------+--------+-------+------+--------+------+--------| > > | 319 |full|226034| 76330.2| 68.89| 93| 7.5| 337| 415.0| > > |-------+----+------+--------+-------+------+--------+------+--------| > ^^^^^^ > This number, 76330.2. > ------------------------------------------------ > > If this number is wildly inflated by the presence of lots of > hard-linked files, then the results are very misleading. In the worst > case, it might not be clear that you could restore the backup to the > disk it came from. > > Am I right? > > paul > =---------------------- > paul fox, p...@foxharp.boston.ma.us (arlington, ma, where it's 34.5 degrees) > > > > _______________________________________________ > 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/ _______________________________________________ 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/
