To make clear what the script does (the header of the script explains it too)
If you call:
./backupData.sh -s firstname.lastname@example.org
The scripts makes a remote replication of the zfs fs "bunas1-pool/bunas1-dataset" (available on your local nas) to the remote nas below
the zfs fs "bunas2-pool/bunas2-dataset"
- 1st the script creates the zfs fs "bunas2-pool/bunas2-dataset/bunas1-pool" and "bunas2-pool/bunas2-dataset/bunas1-pool/bunas1-dataset" on your remote machine (if they do not yet exist) (if "bunas1-pool/bunas1-dataset", contains sub-filesystems on your local nas, thoses fs are also created below "bunas2-pool/bunas2-dataset/bunas1-pool/bunas1-dataset" on your remote nas)
- then it perform a remote replication of all the snapshots available in the filesystem "bunas1-pool/bunas1-dataset" (incl the snapshots available in the subfilesystems if any) to the corresponding filesystems in your remote nas. (if some of the snapshots are already available in the remote nas, it will only replicate the newer ones that are not yet available in the remote nas)
At the end of the script you should have the following zfs fs on your remote nas (Suposing that "bunas1-pool/bunas1-dataset" does not have any sub-filesystem):
This is why you see the corresponding mounts on your destination nas
I believe that the source file system(s) should be un-mounted after the copy is complete
There is a misunderstanding: These mount are not mounts of the source filesystem, but are zfs fs available on your destination nas and containing the replicated data.
I can see in your mount list, that there are more zfs created on the remote nas than the 4 zfs fs I mentionned above:
I assume that the following mounts are due to the fact that you also called at least once:
Code: Select all
./backupData.sh -s email@example.com "bunas1-pool/bunas1-dataset" "bunas2-pool"
bunas2-pool/bunas1-pool on /mnt/bunas2-pool/bunas1-pool (zfs, local, read-only, nfsv4acls)
bunas2-pool/bunas1-pool/bunas1-dataset on /mnt/bunas2-pool/bunas1-pool/bunas1-dataset (zfs, local, read-only, nfsv4acls)
It is not clear to me why you have also the following mount:
bunas2-pool/bunas1-dataset on /mnt/bunas2-pool/bunas1-dataset (zfs, local, read-only, nfsv4acls)
bunas2-pool/bunas2-dataset/bunas1-dataset on /mnt/bunas2-pool/bunas2-dataset/bunas1-dataset (zfs, local, nfsv4acls)
Is it possible that you have created them manually?
The snapshots copy over no problem, however the terminal outputs the following error:
“Pseudo-terminal will not be allocated because stdin is not a terminal.”
I do also experience this error message. It seems that it does not have any functional consequence.
I will analyze it until the final release. Any hint would be welcome.