Help, I’m at a loss, I must have done something seriously wrong, it seems like I can’t replicate the pool with all the jails anymore (named MARS).
I did some snapshot pruning that caused some data loss, lucky not all the jails were lost in the process because 4 of them were busy.
I’m trying to perform a full replication to a pool called BMARS.
When I try to perform a full replication from scratch, it fails (log attached at the bottom). I proceeded to erase the entire content of BMARS (created a new pool with same name), but even now when I ask to perform a full replication from scratch, I still get the same error (log attached att the bottom).
What can I do to backup my data before it disappears? Do I have to perform some correction on MARS (I’m terrified of this, I might lose data)?
My only idea is to clone only the snapshots named francesco-xxx instead of auto-xxx since those seems to be the source of the problem, but I don’t know the consequence of this. There also seems to be a few snapshots named after my jails (nextcloud_1, znc_1 and 2 more) and those are the things that hold my jails configuration, so I don’t know if I should copy those snapshots.
EDIT: I tested my idea and it didn’t work. I still get an error.
EDIT 2: Analyzing the snapshots, it seems like any snapshot after the pruning is causing some issues because a snapshot in the past is “missing”. These are some relevant lines I extracted from the “ERROR” entry in the long, formatted nicely
skipping snapshot MARS/mymedia@francesco-202408021221 because it was created after the destination snapshot (francesco-202407311813)
skipping snapshot MARS/iocage/jails/znc_1@francesco-202408021221 because it was created after the destination snapshot (francesco-202407311813)
cannot send MARS@francesco-202407311813 recursively: snapshot MARS/iocage/jails/znc_1/root@francesco-202407311813 does not exist
warning: cannot send 'MARS@francesco-202407311813': backup failed
cannot receive: failed to read from stream")
Obviously I want to keep both francesco-202407311813 and francesco-202408021221. I’m fine just keeping the data that’s on disk right now and losing all the snapshots but francesco-202408021221, but I’m terrified that if I delete a snapshot in the past, I also lose data, how it happened earlier.
Is my only option to rsync all the data to a different drive?
Log with error: Zeus replication log · GitHub
EDIT 3:
I was able to initiate a replication but with caveats. I created a brand new snapshot with a different naming schema francesco-fix1-%Y%m%d%H%M, then I created a new replication task that clones ONLY that snapshot. This replication task is working and it’s now copying 2TB of data to the pool.
My “origin” pool remains in a bad shape in the sense that I cannot replicate my auto snapshots anymore and I don’t know what happens if I ever try to rollback some of those. One of my thoughts is:
Delete znc_1 snapshot or create a new snapshot named francesco-202407311813 for dataset MARS/iocage/jails/znc_1/root. This would be created based on current data, so I’m not sure what’s the consequence of doing it.