TLDR;
So the question is: Why are two of these datasets replicating more than once to different datasets? And most importantly, how do I stop it from happening?
I have 6 replication jobs set up and two of them are acting strangely. They are replicating themselves to the wrong dataset, in addition to, their correct datasets. If I may try to demonstrate. Only one of these is set to both snapshot and replicate recursively: #5 iocage. This is to backup all my jails in one job so I have all their config data. The other 5 snapshots/replication tasks that match them, are NOT recursive.
The source data is on a 2TB mirror, they each have a periodic snapshot task once weekly, then a replication task to send each one to a 6TB single disk on the same box (for redundancy). The source is one pool and the destination is a different pool: 2TB mirror is Pool1 and 6TB single disk is Pool2.
#1/#2/#6 are datasets with no child datasets, #5 is a dataset with many child datasets, #3/#4 are zvols that are mounted to different virtual machines.
Pool1 <----------------------------------------------------------------------|-----------------------------> Pool2
This is what SHOULD happen :
1) homeshares (for freenas custom users) --> weekly periodic snapshot ------------> replication task to homeshares
2) linuxshares (for smb sharing) --------------> weekly periodic snapshot ------------> replication task to linuxshares
3) linux-vm ------------------------------------> weekly periodic snapshot ------------> replication task to linux-vm
4) pihole-vm ----------------------------------> weekly periodic snapshot ------------> replication task to pihole-vm
5) iocage (top level dataset for jails) ---------> weekly periodic snapshot ------------> replication task to iocage
6) nextcloudfiles ------------------------------> weekly periodic snapshot ------------> replication task to nextcloudfiles
This is what ACTUALLY happens (bold for the 2 that are misbehaving):
1) homeshares (for freenas custom users) --> weekly periodic snapshot -----------> replication task to homeshares
2) linuxshares (for smb sharing) --------------> weekly periodic snapshot -----------> replication task to linuxshares
3) linux-vm ----------------------------------> weekly periodic snapshot ---------> replication task to linux-vm AND homeshares
4) pihole-vm ----------------------------------> weekly periodic snapshot -----------> replication task to pihole-vm
5) iocage (top level dataset for jails) -----> weekly periodic snapshot ----------> replication task to iocage AND pihole-vm AND nextcloudfiles AND linuxshares
6) nextcloudfiles ------------------------------> weekly periodic snapshot -----------> replication task to nextcloudfiles
Here is a screenshot of my pools showing Pool2 - the replication destination (numbers on the photo should match 3 or 5 relating to which job the error seems related to). So, each box has a number corresponding to which job I think it is being replicated from. Number 3 is replicating TWICE, one correct and one not correct. Number 5 is replicating FOUR times, only one is correct, the other three are wrong. I've checked the replication tasks and they have the correct datasets listed for replication. Not to mention it is not possible to set more than one dataset to replicate to.
So the question is: Why are two of these datasets replicating more than once to different datasets? And most importantly, how do I stop it from happening?
So the question is: Why are two of these datasets replicating more than once to different datasets? And most importantly, how do I stop it from happening?
I have 6 replication jobs set up and two of them are acting strangely. They are replicating themselves to the wrong dataset, in addition to, their correct datasets. If I may try to demonstrate. Only one of these is set to both snapshot and replicate recursively: #5 iocage. This is to backup all my jails in one job so I have all their config data. The other 5 snapshots/replication tasks that match them, are NOT recursive.
The source data is on a 2TB mirror, they each have a periodic snapshot task once weekly, then a replication task to send each one to a 6TB single disk on the same box (for redundancy). The source is one pool and the destination is a different pool: 2TB mirror is Pool1 and 6TB single disk is Pool2.
#1/#2/#6 are datasets with no child datasets, #5 is a dataset with many child datasets, #3/#4 are zvols that are mounted to different virtual machines.
Pool1 <----------------------------------------------------------------------|-----------------------------> Pool2
This is what SHOULD happen :
1) homeshares (for freenas custom users) --> weekly periodic snapshot ------------> replication task to homeshares
2) linuxshares (for smb sharing) --------------> weekly periodic snapshot ------------> replication task to linuxshares
3) linux-vm ------------------------------------> weekly periodic snapshot ------------> replication task to linux-vm
4) pihole-vm ----------------------------------> weekly periodic snapshot ------------> replication task to pihole-vm
5) iocage (top level dataset for jails) ---------> weekly periodic snapshot ------------> replication task to iocage
6) nextcloudfiles ------------------------------> weekly periodic snapshot ------------> replication task to nextcloudfiles
This is what ACTUALLY happens (bold for the 2 that are misbehaving):
1) homeshares (for freenas custom users) --> weekly periodic snapshot -----------> replication task to homeshares
2) linuxshares (for smb sharing) --------------> weekly periodic snapshot -----------> replication task to linuxshares
3) linux-vm ----------------------------------> weekly periodic snapshot ---------> replication task to linux-vm AND homeshares
4) pihole-vm ----------------------------------> weekly periodic snapshot -----------> replication task to pihole-vm
5) iocage (top level dataset for jails) -----> weekly periodic snapshot ----------> replication task to iocage AND pihole-vm AND nextcloudfiles AND linuxshares
6) nextcloudfiles ------------------------------> weekly periodic snapshot -----------> replication task to nextcloudfiles
Here is a screenshot of my pools showing Pool2 - the replication destination (numbers on the photo should match 3 or 5 relating to which job the error seems related to). So, each box has a number corresponding to which job I think it is being replicated from. Number 3 is replicating TWICE, one correct and one not correct. Number 5 is replicating FOUR times, only one is correct, the other three are wrong. I've checked the replication tasks and they have the correct datasets listed for replication. Not to mention it is not possible to set more than one dataset to replicate to.
So the question is: Why are two of these datasets replicating more than once to different datasets? And most importantly, how do I stop it from happening?
