Hi,
I finally build a backup server for replication but the replication task can't sync all datasets because the file names would be too long. Sure, the paths are a bit longer because I replicate "SSDpool/Stuff/..." to "HDDpool/SSDpool_backup/Stuff/..." and also "HDDpool/..." to "BACKUPpool/HDDpool_backup/...". So the "SSDpool/Stuff/..." becomes "BACKUPpool/HDDpool_backup/SSDpool_backup/Stuff/..." later.
How do you handle this?
I want to make sure that the replication always works even with new long file names added in he future without running into problems.
The only option that comes to my mind is to delete and recreate all datasets beginning with a extra long name so all paths in backups are shorter than the original dataset. Like "HDDpool/UselessDatasetWithExtraLongNameToPreventToòLongPaths/MyShares/..." and "SSDpool/UselessDatasetWithExtraLongNameToPreventTooLongPaths/MyShares/...".
So I can replicate:
So that the path of original datasets are always longer then the paths of the backup datasets.
But is there a way to clear this problem without deleteing and recreating all datasets? Not only that I need to copy dozens of TB I also need to modify all my jails and scripts to match the new datasets.
Any ideas?
I finally build a backup server for replication but the replication task can't sync all datasets because the file names would be too long. Sure, the paths are a bit longer because I replicate "SSDpool/Stuff/..." to "HDDpool/SSDpool_backup/Stuff/..." and also "HDDpool/..." to "BACKUPpool/HDDpool_backup/...". So the "SSDpool/Stuff/..." becomes "BACKUPpool/HDDpool_backup/SSDpool_backup/Stuff/..." later.
How do you handle this?
I want to make sure that the replication always works even with new long file names added in he future without running into problems.
The only option that comes to my mind is to delete and recreate all datasets beginning with a extra long name so all paths in backups are shorter than the original dataset. Like "HDDpool/UselessDatasetWithExtraLongNameToPreventToòLongPaths/MyShares/..." and "SSDpool/UselessDatasetWithExtraLongNameToPreventTooLongPaths/MyShares/...".
So I can replicate:
"SSDpool/UselessDatasetWithExtraLongNameToPreventTooLongPaths/MyShares/..." -> "HDDpool/SSDpool_backup/MyShares/..."
"HDDpool/SSDpool_backup/MyShares/..." -> "BACKUPpool/HDDpool_backup/SSDpool_backup/MyShares/..."
"HDDpool/UselessDatasetWithExtraLongNameToPreventTooLongPaths/MyShares/..." -> "BACKUPpool/HDDpool_backup/MyShares/..."
So that the path of original datasets are always longer then the paths of the backup datasets.
But is there a way to clear this problem without deleteing and recreating all datasets? Not only that I need to copy dozens of TB I also need to modify all my jails and scripts to match the new datasets.
Any ideas?
Last edited: