How to remove encryption from a ZFS volume (while keeping the data)

[Muddro]

Sorry, I don't understand. Redundancy (mirror, RAIDZn) is per vdev. You cannot mirror 2 or more vdevs but the vdev can be a mirror of 2 or more disks. So if (I hope so) you have two vdevs, each a RAIDZ2, then of course you can replace a disk from each vdev and still sleep well, because each vdev keeps one disk redundancy. Lose one vdev, lose the entire pool ...

HTH,
Patrick

Your correct. I think I misread some other forum posts. In any case it's moving along, a little slower than before but I think total time will be shorter.

 

[martin4897]

Hm my Server starts the resilver as soon as i push a drive OFFLINE is this a problem? i am on 12.0 U2.1

pool: Daten
state: DEGRADED
status: One or more devices is currently being resilvered. The pool will
continue to function, possibly in a degraded state.
action: Wait for the resilver to complete.
scan: resilver in progress since Tue Mar 30 23:36:09 2021
2.29T scanned at 17.9G/s, 86.7G issued at 678M/s, 10.3T total
6.91G resilvered, 0.83% done, 04:22:12 to go
config:

NAME STATE READ WRITE CKSUM
Daten DEGRADED 0 0 0
raidz1-0 DEGRADED 0 0 0
gptid/e56185cf-917c-11eb-a63f-a85e453d4185 OFFLINE 0 0 0 (resilvering)
gptid/b31e3355-9148-11eb-ad1f-a85e453d4185.eli ONLINE 0 0 0
gptid/12659858-90ef-11eb-95ac-a85e453d4185.eli ONLINE 0 0 0
gptid/c5c53b07-90ae-11eb-a648-a85e453d4185.eli ONLINE 0 0 0

errors: No known data errors

 

[Patrick M. Hausen]

@martin4897 Why are you posting the same thing two times?

Did you have a hot spare disk configured?

 

[martin4897]

@martin4897 Why are you posting the same thing two times?

Did you have a hot spare disk configured?

No idea why this was posted 2 times, cant delete one of them either.

Nope no hot spare has been configured. I dont understand why but as soon as i push one drive OFFLINE it starts resilvering it in OFFLINE state, but the following commands are working while it resilvers the drive...

 

[martin4897]

It seems to have finished without Problems, can i trust it and start resilver the next drive without killing my Pool, or should i better stay at this point and try to figure out why this is happening?

Code:

 pool: Daten
 state: ONLINE
  scan: resilvered 2.57T in 04:12:06 with 0 errors on Wed Mar 31 08:06:14 2021
config:

        NAME                                                STATE     READ WRITE CKSUM
        Daten                                               ONLINE       0     0     0
          raidz1-0                                          ONLINE       0     0     0
            gptid/e56185cf-917c-11eb-a63f-a85e453d4185      ONLINE       0     0     0
            gptid/b31e3355-9148-11eb-ad1f-a85e453d4185.eli  ONLINE       0     0     0
            gptid/12659858-90ef-11eb-95ac-a85e453d4185.eli  ONLINE       0     0     0
            gptid/c5c53b07-90ae-11eb-a648-a85e453d4185.eli  ONLINE       0     0     0

errors: No known data errors

 

[Ericloewe]

Mod note: deleted the duplicate post

 

[Patrick M. Hausen]

So you are saying that the moment you do e.g. zpool offline gptid/b31e3355-9148-11eb-ad1f-a85e453d4185.eli and nothing else, the resilvering starts? Weird. Never experienced that.

My guess is that you can continue. You are aware that you are running at a rather high risk, anyway, using this procedure with a RAIDZ1?

 

[martin4897]

Yes it started directly after zpool offline gptid/e56185cf-917c-11eb-a63f-a85e453d4185.eli. Yes I am aware of this, but all drives are just one week old, because i upgraded to larger ones, so i think the risk should be pretty small and I have a backup of all important data. Though they get really hot now 60°C, i think i should try to increase the cooling which is as good as impossible in this small case.

 

[Steve M]

@Patrick M. Hausen Thank you so much for creating this valuable guide. Finally I have upgraded my FreeNAS 9.3 to TrueNAS 12.0 and increased the disk capacity applying the procedure as you describe to remove the legacy encryption.

My experience varied with TrueNAS 12.0-U4.1 slightly which even simplified the process.
0. run Disklist (see below) to capture all the necessary ids.
1. Use GUI to take one disk offline from the pool.
2. Command line

Code:

zpool replace <poolname> gtpid/<id>.eli gtpid/<id>

. Wait for resilvering to complete.
Repeat for all disks.
Export/Disconnect on the GUI (not to destroy data (off by default))
Reboot and Import.

Entries were removed from the freenas-v1.db by the process automatically.

One tool I found valuable is https://github.com/nephri/FreeNas-DiskList with the discussion around the tool is on ServeTheHome forum.

-------------------------
OS Version: TrueNAS-12.0-U4.1
Model: ProLiant MicroServer