Error ZFSPool and crash Proxmox

[frenk970]

Good morning,
I reinstalled Proxmox on my server and imported my zfs pool only I still get this error:

pve zed[7480]: eid=522 class=data pool='VMSSD' priority=0 err=52 flags=0x808081 bookmark=2239:1:1:650

Do you know if it can be fixed?

 

[fabian]

please post the output of "zpool status"

 

[frenk970]

please post the output of "zpool status"

Here is the status of the zfs pool

zpool status -v VMSSD
pool: VMSSD
state: ONLINE
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: https://openzfs.github.io/openzfs-docs/msg/ZFS-8000-8A
scan: scrub repaired 0B in 00:04:55 with 24 errors on Mon Feb 17 09:32:20 2025
config:

NAME STATE READ WRITE CKSUM
VMSSD ONLINE 0 0 0
sda ONLINE 0 0 7.02K
sde ONLINE 0 0 7.02K
sdc ONLINE 0 0 0

errors: List of errors unavailable: no such pool or dataset

 

[fabian]

well, your disks or memory had an issue.. and your pool setup is not redundant, so it's not possible to repair it.

 

[frenk970]

well, your disks or memory had an issue.. and your pool setup is not redundant, so it's not possible to repair it.

I know this, but I can't even move the disks on the machines and they run and never have any problems

 

[fabian]

they ran and never had problems until they did, and now you need to find out what exactly the problem is so that you can fix it.

 

[frenk970]

they ran and never had problems until they did, and now you need to find out what exactly the problem is so that you can fix it.

Is it likely a cable problem?

 

[ghusson]

Hello.
The problem can be anywhere on the chain : processor, RAM bus, RAM, PCI or SATA or whatever bus, RAM of the disk, disk chip or firmware, disk itself.
We cannot tell. But it is often a disk or RAM problem (if the RAM is not ECC).
Just to check, is your disk direct attached, it is not a RAID managed by a RAID card ?

 

[frenk970]

Hello.
The problem can be anywhere on the chain : processor, RAM bus, RAM, PCI or SATA or whatever bus, RAM of the disk, disk chip or firmware, disk itself.
We cannot tell. But it is often a disk or RAM problem (if the RAM is not ECC).
Just to check, is your disk direct attached, it is not a RAID managed by a RAID card ?

they are 3 disks connected directly to the server motherboard and are a ZFS pool and the RAM "Patriot Viper Steel RAM DDR4 3200 MHz 32GB (4x16GB) C16 Kit di Memoria Gaming XMP 2.0 Grigio"

 

[ghusson]

You can try a low level disk test and a RAM test.

 

[frenk970]

You can try a low level disk test and a RAM test.

can I do a RAM test without restarting ProxMox and using Memtest86+?

 

[ghusson]

can I do a RAM test without restarting ProxMox and using Memtest86+?

I don't know. Personally I would restart and do a full test.

 

[frenk970]

I don't know. Personally I would restart and do a full test.

Now randomly the server goes into kernel panic

 

[ghusson]

So it is pretty sure to be a hardware problem.

 

[frenk970]

So it is pretty sure to be a hardware problem.

these are the errors it keeps giving: https://pastebin.com/z8fkmbsi

 

[ghusson]

I can not tell you more.

 

[frenk970]

I can not tell you more.

from the errors I reported it seems there is a faulty SATA port, now I'm working on the server remotely, this evening when I manage to physically have the server I'll do some port change tests

 

[ghusson]

from the errors I reported it seems there is a faulty SATA port, now I'm working on the server remotely, this evening when I manage to physically have the server I'll do some port change tests

Chipset on the MB, port, cable or disk...