Remote replication of large datasets

[MrNaz]

I have a TrueNAS Core box with a 5tb dataset. I would like to replicate that to a remote TrueNAS Core box. Both boxes have similar specs and space.

I would like the replication to happen every 12 hours, which matches the period snapshot we have set up for every 12 hours.

The problem is that the internet connection at the source is only 40mb up, which is fine for the daily changes, but not enough to do the initial seeding. We'd be OK with the initial seed taking a week or two, but the problem is that if there is anything that disrupts the connection, it seems to have to start over. Is there a setting that will allow replication jobs to be fault tolerant? If not, what's the solution?

Physically taking the boxes together is not feasible.

 

[spuky]

use a 10TB+ external drive send snapshots there ship / bring it to the remote location import snapshots... after that set up the sync... ?

 

[MrNaz]

use a 10TB+ external drive send snapshots there ship / bring it to the remote location import snapshots... after that set up the sync... ?

That is not the right answer. It's 2023 ffs. I remember being excited about FTP getting resume support around a quarter of a century ago.

Is there no proper way of doing this reliably?

 

[winnielinnie]

That is not the right answer. It's 2023 ffs. I remember being excited about FTP getting resume support around a quarter of a century ago.

Is there no proper way of doing this reliably?

I don't use the GUI to do send/recv (for other reasons that are out of scope.)

But I assumed the Replication Tasks configured in the GUI would use a "resume token"? It would be very odd if they don't.

Otherwise, you can do the first replication in the command-line (via SSH + tmux) and manually invoke a resume token; which will allow you to resume if the process aborts or gets interrupted.

Here's a forum post for reference:

can't maintain ssh connection to perform large zfs send

I'm trying to zfs send a 6 TB pool from some old disks to some new disks. It gets a few 100 GB through, but then the ssh connection always drops with: client_loop: send disconnect: Broken pipe I've set ServerAliveInterval in my client's ~/.ssh/config but that doesn't seem to help. Could it...

www.truenas.com

(The linked post doesn't assume a remote transfer over SSH, so you'll have to include the "ssh" portions in your own commands. I left out the SSH stuff so that the commands don't get over-cluttered in the examples.)

 

[bluonek]

I've been testing replication strictly using the TrueNAS replication interface for the past week. It works excessively well and yes it does include a resume token.

OP - what issue did you run into? Are you saying after sending X GB's and losing connection, the retry started from scratch? Post a screenshot of your replication setup.

 

[bluonek]

shit, i just unknowingly necro'd - oh well.................................