Local Replication high disk usage

[rwgs]

HI, running latest version of TrueNAS core, I have setup a local replication task of 2 datasets from one disk pool to another (on the same physical appliance). Both pools have the same number (3) and model of disk and are using RAIDZ1.

For reasons I can't figure out, the destination datasets started off being equal, but overtime are now growing more than the source datasets. The replication task is also set to synchronize destination snapshots with the source.

Given that I just want both pools to be mirrored - what is happening and how can I make sure they are the same? I'm still fairly new to TrueNAS, so Is there something I'm missing here?

Thanks

 

[NugentS]

are you snapshotting the destination pool?

 

[rwgs]

are you snapshotting the destination pool?

Nope, just the source. Although now you mention it, that does appear to be the case - even though I don't have anything turned on for the destination pool. Is it doing that because of the replication?

 

[rwgs]

To add to my previous post (as I can't edit it yet and still waiting for moderation). The snapshots which are in the destination pool appear to be simply duplicates of the ones from the source - so I'm not sure if it's actually snapshotting the destination, or simply replicating them from the source.

In any case, the only snapshotting actually enabled is on the source pools (under the "Periodic Snapshot Tasks" menu).

 

[rwgs]

More info - the snapshots in the destination have the same type as those from the source, so I believe these are just being copied as part of the replication (rather than new snapshots entirely).

 

[rwgs]

*That should say same time stamp.

 

[NugentS]

By how much are we talking growth. A lot or a little.
I am replicating a bunch of smapshots from 1 nas to another and the sizes are never quite the same, but fairly close

 

[rwgs]

By how much are we talking growth. A lot or a little.
I am replicating a bunch of smapshots from 1 nas to another and the sizes are never quite the same, but fairly close

Currently about 25% so it's significant. Multiple terabytes in this case.

 

[NugentS]

Mine definately isn't 25%. A few % - yes - but no more than that.
How about different retention periods for the destinatation dataset vs the source dataset?

 

[rwgs]

Mine definately isn't 25%. A few % - yes - but no more than that.
How about different retention periods for the destinatation dataset vs the source dataset?

Nope, those are both the same.

What would be the easiest way to figure out what the space is being used by?

 

[NugentS]

Something at the commadline I would have thought - and you will need someone else to answer that I am afraid as I have reached my limit of knowledge / guesswork