Disk disappeared, pool offline. Where to look for logs?

[Supose]

One of my disks suddenly disappeared from Storage/Disks. Accordingly the pool is marked as offline. I didn’t change any hardware nor did I change anything in the configuration of my TrueNAS. The disk is spinning, I tried different SATA and power cables.

Don’t know what else I can do.

Are there any logs, so I at least could find out when this happened? The system dataset is on the disk that disappeared.

Would be greatly happy for all advice.

 

[hugovsky]

What's your hardware? Can you use that disk with another machine and test smart?

 

[Supose]

It Seagate 4TB NAS HDD (ST4000VN000). Don’t have any other machine capable of reading ZFS. I put it in an two different external cases connecting it via USB to the TrueNAS machine. Isn’t listed when trying to import a pool and isn’t listed under Storage/Disks. Actually it is spinning down after a few seconds. So, probably a case for some hardware specialist or the bins :/

Are there any logs saved outside the system dataset?

 

[hugovsky]

By hardware, I mean your board, cpu, etc. Don't connect disks thru usb. You don't need to be able to read ZFS to test smart. Just connect one of the drives to anohter pc and run smart tests to rule out disk problems.

 

[Supose]

Sorry for the delayed reply. The harddrive seems mechanically broken. Independently from the hardware I use to connect it to. It spins up, makes unusual noise and spins down again, not being detected by several OS and hardware configurations.

I would like to go back to my initial question: Is there a place to look for logs outside the system dataset (as the system dataset is sadly on the broken disk)?

 

[Patrick M. Hausen]

One of my disks suddenly disappeared from Storage/Disks. Accordingly the pool is marked as offline.

Did you build the pool without any redundancy?

As for the logs: if the disk is making "funny noises" it just went kaputt from one second to the next. What do you expect the system to log in this case? You won't get a log entry other then "disk adaXY disconnected". After all that is what the operating system sees. You won't get a reason what failed. Disks do fail sometimes. That's why you build a pool with redundancy.

If your pool has got some redundancy, it should not go offline. One could then examine the status with zpool status -v and replace the broken disk with a new one.

 

[Supose]

Did you build the pool without any redundancy?

Without ZFS redundancy, single disk. I have backups though (to a certain degree).

As for the logs: if the disk is making "funny noises" it just went kaputt from one second to the next. What do you expect the system to log in this case? You won't get a log entry other then "disk adaXY disconnected". After all that is what the operating system sees. You won't get a reason what failed. Disks do fail sometimes. That's why you build a pool with redundancy.

"disk adaXY disconnected" is exactly what I expect. Plus a time stamp, so I know when it happened. That way I can make some assumptions and conclusions about my data loss (my replication setup is somewhat cumbersome and any try to explain would just be a waste of everybody’s time; I will reconsider my setup). Anyhow, a time stamp of the accident would be of great help.

 

[Patrick M. Hausen]

"disk adaXY disconnected" is exactly what I expect. Plus a time stamp, so I know when it happened.

That's logged in the system dataset. Sorry.

 

[elvisimprsntr]

Without ZFS redundancy, single disk.

 

[Supose]

That's logged in the system dataset. Sorry.

Thank you very much