mchinn
July 14, 2026, 9:42pm
1
I’ve had a FreeNAS system at home for some time. About 2 1/2 years ago after a thunderstorm / power outage it no longer worked. Fast-forward the 2 1/2 years I’ve moved across the country and decided to get it our and work on it.
The o/s is fine however I cannot add the pool any longer. The o/s can see the pool but I receive an i/o error on the pool add. The drives themselves seem physically fine, just that the pool is corrupt.
I do not know my way around Linux types of devices but I’m fairly handy with google. I’ve tried the command line pool add but that failed even with the force command(s).
So before I call it a day and recreate the pool from scratch, is there a way to determine what drive might be failed out. From my PC experience, this seems like a case that one of the drives may have lost its FAT table and the data is indeed there but lost its index.
The machine contains 4 physical drives (3TB) in ZFS so I was working with the understanding that it would survive a single physical drive failure.
If it is possible to determine what drive has failed, if I remove that drive would it rebuild/resilver with the installation of a new drive or could it run on 3 drives for a temp time in order to backup the information before rebuilding?
I look forward to the replies, just never dealt with a recovery situation on the device before.
Thank you in advance,
Michael
run a sudo zpool status -v at the shell and paste the result here in codeblocks please.
mchinn
July 14, 2026, 10:00pm
3
root@NAS01:~ # sudo zpool status -v
NAME STATE READ WRITE CKSUM
boot-pool ONLINE 0 0 0
ada4p2 ONLINE 0 0 0
errors: No known data errors
Note: the pool in question is not listed named “FreeNAS01”
Arwen
July 15, 2026, 4:26am
4
Actually that should have been:zpool importlsblk -o NAME,MODEL,SERIAL,LABEL,UUID,PARTUUID,TYPE
Yes, normally a RAID-Z1 would be able to survive a single disk failure without problems. However, on consumer hardware, and power loss, it is possible that more than 1 disk failed.
Also, if you can, please use the correct terminology for the pool layout. Calling it a 4 x 3TB ZFS does not completely describe the layout. We can assume you mean RAID-Z1. But, we have seen people think they have a RAID-Z1 because it’s ZFS, when in reality they have a multi-disk striped pool which has zero redundancy. In that case, loss of a disk is total loss.
1 Like
mchinn
July 15, 2026, 5:04am
5
the lsblk command doesn’t seem to work in the shell.
from the zpool import command I can confirm that the pool is DEGRADED
One drive is UNAVAIL cannot open
Also that it is a “raidz1-0”
Arwen:
lsblk -o NAME,MODEL,SERIAL,LABEL,UUID,PARTUUID,TYPE
This works for me on 25.10.4. What is your OS version, Core or Scale?
mchinn
July 15, 2026, 6:00am
7
TrueNAS Core
I think this shows disk info on Coregeom disk list & geom part list
mchinn
July 15, 2026, 1:57pm
9
root@NAS01:~ # geom disk list & geom part list
Name: ada0p1
Name: ada0p2
Name: ada0
Geom name: ada2
Name: ada2p1
Name: ada2p2
Name: ada2
Geom name: ada4
Name: ada4p1
Name: ada4p2
Name: ada4p3
Name: ada4
Geom name: ada3
Name: ada3p1
Name: ada3p2
Name: ada3
Geom name: ada0
Name: ada0
Geom name: ada1
Name: ada1
Geom name: ada2
Name: ada2
Geom name: ada4
Name: ada4
Arwen:
zpool import
What is the output for this?
mchinn
July 15, 2026, 3:05pm
11
root@NAS01:~ # zpool import
FreeNAS1 DEGRADED
raidz1-0 DEGRADED
gptid/b664a14e-b66f-11ea-b813-408d5c424ff4 ONLINE
gptid/b8952653-4156-11e6-8878-408d5c424ff4 UNAVAIL cannot open
gptid/f38c854b-f4f0-11e7-b0c9-408d5c424ff4 ONLINE
gptid/b9476c90-4156-11e6-8878-408d5c424ff4 ONLINE
Arwen
July 15, 2026, 7:53pm
12
If there is no other damage, then it should import with just 1 failed / unavailable disk. Use:zpool import -R /mnt FreeNAS1
However, if this fails, their could be additional damage on 1 or more of the other 3 disks.
You’re missing an entire drive which has the partition b8952653-4156-11e6-8878-408d5c424ff4.
The other 3 drives and the boot drive are all accounted for.
Is it physically unplugged?
