Help with transfer of data from FreeNAS 9.3 to TrueNAS (creative solutions needed)

[Johev]

Dear Community, I am coming to you with a request for assistance. I have an old FreeNAS box, which has been rock stable for a long time (sometimes more than 1 year before a reboot), however I would like to transfer the data from it to a new box. Unfortunately I do not manage to make the two systems talk to each other. I tried to follow all the tutorials, however they just don't seem to work.

Old FreeNAS specs:
I have a DELL T20 with 12GB of ECC RAM and 3 - 4TB Disks (in RAIDZ1)
The pool is currently degraded, even though I have changed the failing HDD and it did re-silver the pool 100%. However all 3 disks are marked as online and show that they have passed in the short SMART test. The data is also fully accessible.
Current version FreeNAS-9.3-STABLE-201511040813

I tried to update it several times, however the systems refuses to post and I end up reverting to this version. That is the reason it is on this version since 2015.

My TureNAS box is a Supermicro X10SL7-F with a Xeon E3-1265L v3 and 32GB of ECC RAM. It has 8 (4TB Ironwolf drives in a RAIDZ2).
I already tried all the tutorials to make the data transfer between the two boxes, however they have all failed. That is why I am now asking for creative solutions.

I was thinking of installing a TrueNAS on a new USB drive and plugin it into the Dell T20. Would I be able to import the pools and then transfer the data between the two boxes?

The data is not backed-up anywhere else and that is why I prefer to consult with the experts, before I do something irreversible to it. The idea is to have then 2 boxes, with one being the main and the other having a backup of the data for recovery purposes.

 

[elvisimprsntr]

1. Configure new system
2. Use RSYNC command from one system to the other
3. Implement 3-2-1 backup strategy. RAID is not a backup, RAID is resiliency.

 

[Johev]

Thank you @elvisimprsntr for your reply. The problem I have is that RSYNC does not seem to work between the two boxes. I have followed many guides and it just fails.

The new System is up and running, with no data on it and I would just like to transfer everything to it. If you have any tutorial which I could try, please link it as I was unable to make this work.

Many thanks for all your assistance.

 

[danb35]

So many possible ways to do this. rsync is one. ZFS replication would probably be the most preferred. Or you could just connect the old disks to the new server, import that pool, and copy the data over with any command-line tools you like. Or doubtless dozens of other possibilities.

 

[elvisimprsntr]

The problem I have is that RSYNC does not seem to work between the two boxes. I have followed many guides and it just fails.

RSYNC is a universally available protocol that works on all Linux systems when set up correctly.

1. Did you enable RSYNC server on the target system?
2. Were the same credentials and data pool ACLs set up on both systems?
3. What commands have you tried?
4. What were the error messages?

 

[Johev]

First of all I would like to thank both of you for your replies.
@danb35
If it would be as easy as detaching the 3 HDDs and importing them in the new server, I would definitely do it. I am just afraid that given the 6 years apart from my version to the current version of FreeNAS/TrueNAs, the ZFS has evolved so much that I would not be able to import it correctly and I could lose the data. I am also afraid that after trying that approach and failing, I could no longer re-install them in the old server and access the data as something could have been changed in following this path.

As for ZFS replication, it would be excellent if I could make it work.

@elvisimprsntr
I always try to configure it thorough the web GUI of both servers. No commands were tried. I'll try it again today and I'll post some screenshots. If you or anyone from this community could guide me or give me any good advice, I would greatly appreciate it.

Thanks once again to both for all your time and advice.

 

[Spearfoot]

Dear Community, I am coming to you with a request for assistance. I have an old FreeNAS box, which has been rock stable for a long time (sometimes more than 1 year before a reboot), however I would like to transfer the data from it to a new box. Unfortunately I do not manage to make the two systems talk to each other. I tried to follow all the tutorials, however they just don't seem to work.

Old FreeNAS specs:
I have a DELL T20 with 12GB of ECC RAM and 3 - 4TB Disks (in RAIDZ1)
The pool is currently degraded, even though I have changed the failing HDD and it did re-silver the pool 100%. However all 3 disks are marked as online and show that they have passed in the short SMART test. The data is also fully accessible.
Current version FreeNAS-9.3-STABLE-201511040813

I tried to update it several times, however the systems refuses to post and I end up reverting to this version. That is the reason it is on this version since 2015.

My TureNAS box is a Supermicro X10SL7-F with a Xeon E3-1265L v3 and 32GB of ECC RAM. It has 8 (4TB Ironwolf drives in a RAIDZ2).
I already tried all the tutorials to make the data transfer between the two boxes, however they have all failed. That is why I am now asking for creative solutions.

I was thinking of installing a TrueNAS on a new USB drive and plugin it into the Dell T20. Would I be able to import the pools and then transfer the data between the two boxes?

The data is not backed-up anywhere else and that is why I prefer to consult with the experts, before I do something irreversible to it. The idea is to have then 2 boxes, with one being the main and the other having a backup of the data for recovery purposes.

Can you please give us more details about how you attempted to use rsync and replication to transfer the data? Perhaps we can help you get one of those two to work. Note that both approaches will be relatively slow, even if you have 10G networking.

You have 8 disks installed on your X10SL7, so you're using all 8 ports of the built-in SAS2308 controller, right? But the motherboard has 6 SATA ports as well. Are 4 of those ports available? Because if they are, you could plug your older pool into them; import it; and then use replication to transfer the data directly on the machine with no networking involved. This will be much faster than rsync or networked replication.

One caveat to this approach: the pool names must be different. But a pool can be renamed, so this isn't a show-stopper.

Example... assuming you don't have any data on your new system; your old pool is named tank; and you want your new system's pool to have the same name as the old one:

• In the GUI, create a pool using your 8 disks in RAIDZ2; name it newtank
• Shut down the machine and plug in the 4 disks from your old system. Power it back up.
• In the GUI, import your old pool (tank)
• At a shell prompt, run these commands to replicate tank to newtank:
  • zfs snapshot -r tank@clone_snap
  • zfs send -R tank@clone_snap | pv | zfs receive -Fv newtank
• When this completes, use the GUI to detach your old pool (tank); shut down the machine; remove the old pool's disks; then power it back up
• In the GUI, detach your newtank pool
• At a shell prompt, run these commands to rename it to tank
  • zpool import newtank tank
  • zpool export tank
• In the GUI, import the tank volume

You should now have all of your data from the old pool on your new pool, with name tank.

 

[Johev]

@Spearfoot Thank you very much for your advice. In fact and as you have pointed out, I have several SATA ports available on my Motherboard. However before trying this, could you please look at the "zpool status -v" result of my pool. I've tried to replace the failed disk, however even after the resilvering it still shows that the pool is degraded.

Code:

[root@freenas] ~# zpool status -v v1
  pool: v1
 state: DEGRADED
status: One or more devices has experienced an error resulting in data
        corruption.  Applications may be affected.
action: Restore the file in question if possible.  Otherwise restore the
        entire pool from backup.
   see: http://illumos.org/msg/ZFS-8000-8A
  scan: resilvered 6.82G in 11h55m with 2 errors on Tue Aug 31 23:29:40 2021
config:

        NAME                                              STATE     READ WRITE C                                         KSUM
        v1                                                DEGRADED     0     0                                              2
          raidz1-0                                        DEGRADED     0     0                                              4
            replacing-0                                   DEGRADED     0     0                                              0
              11787595691211847657                        UNAVAIL      0     0                                              0  was /dev/gptid/af39dcf0-f73d-11e4-b3c4-989096a87096
              gptid/0a5e00ef-09e2-11ec-8625-989096a87096  ONLINE       0     0                                              0
            gptid/afb610ef-f73d-11e4-b3c4-989096a87096    ONLINE       0     0                                              0
            gptid/11fe4998-3efb-11eb-b64c-989096a87096    ONLINE       0     0                                              0

errors: Permanent errors have been detected in the following files:

        <0x4d9>:<0x1c7d1>

I also don't understand how one file got corrupted, as I have ECC memory in the server. I have deleted the file in question, however it still shows the error.

Do you think that I would be able to import this zpool named "v1" into a Truenas 12U5 version and copy the data that way?

Many thanks once again for all the advice.

 

[Spearfoot]

@Spearfoot Thank you very much for your advice. In fact and as you have pointed out, I have several SATA ports available on my Motherboard. However before trying this, could you please look at the "zpool status -v" result of my pool. I've tried to replace the failed disk, however even after the resilvering it still shows that the pool is degraded.

Code:

[root@freenas] ~# zpool status -v v1
  pool: v1
state: DEGRADED
status: One or more devices has experienced an error resulting in data
        corruption.  Applications may be affected.
action: Restore the file in question if possible.  Otherwise restore the
        entire pool from backup.
   see: http://illumos.org/msg/ZFS-8000-8A
  scan: resilvered 6.82G in 11h55m with 2 errors on Tue Aug 31 23:29:40 2021
config:

        NAME                                              STATE     READ WRITE C                                         KSUM
        v1                                                DEGRADED     0     0                                              2
          raidz1-0                                        DEGRADED     0     0                                              4
            replacing-0                                   DEGRADED     0     0                                              0
              11787595691211847657                        UNAVAIL      0     0                                              0  was /dev/gptid/af39dcf0-f73d-11e4-b3c4-989096a87096
              gptid/0a5e00ef-09e2-11ec-8625-989096a87096  ONLINE       0     0                                              0
            gptid/afb610ef-f73d-11e4-b3c4-989096a87096    ONLINE       0     0                                              0
            gptid/11fe4998-3efb-11eb-b64c-989096a87096    ONLINE       0     0                                              0

errors: Permanent errors have been detected in the following files:

        <0x4d9>:<0x1c7d1>

I also don't understand how one file got corrupted, as I have ECC memory in the server. I have deleted the file in question, however it still shows the error.

Do you think that I would be able to import this zpool named "v1" into a Truenas 12U5 version and copy the data that way?

Many thanks once again for all the advice.

I recommend you get your pool repaired before trying to clone it.

 

[Johev]

I recommend you get your pool repaired before trying to clone it.

How do I understand what the problem is? It has 3 4TB HDDs, 2 of them are new as the old ones started to fail after 4,5 years of non-stop use. I don't understand how to remove the old unplugged disk. I did follow the procedure to replace the disk in the GUI and it all went well, except now I am left with the "unavailable" disk.

It also shows me that there are 2 erros, without any information on what is wrong. Any advice would be greatly appreciated.

 

[Spearfoot]

How do I understand what the problem is? It has 3 4TB HDDs, 2 of them are new as the old ones started to fail after 4,5 years of non-stop use. I don't understand how to remove the old unplugged disk. I did follow the procedure to replace the disk in the GUI and it all went well, except now I am left with the "unavailable" disk.

View attachment 49187

It also shows me that there are 2 erros, without any information on what is wrong. Any advice would be greatly appreciated.

I hesitate to give any advice because I have little experience with this type of problem and I don't know exactly what you did.

I'm not even certain whether you have 3 disks in your raidz1 pool, or 4.

Perhaps someone more knowledgeable will stop by and give us a clue.

 

[Etorix]

There are three drives, one of which is in the process of being replaced. If @Johev still has the old drive, he may try to put it back and scrub again.

Then there is an error in a metadata file. I know of no way to repair that; human checking of the data will be required.