Mirror (RAID1) dataset went degraded and then online - how worried should I be?

TrueNAS General
1

Hey.

I was doing a cleanup of my NAS box & powered on my TrueNAS Scale Dragonfish-24.04.2 when I got the notification that:


Critical
Pool NAME state is ONLINE: One or more devices has experienced an unrecoverable error. An attempt was made to correct the error. Applications are unaffected.
2024-10-09 18:11:06
Dismiss

This is a mirrored dataset consisting of two 3TB drives. I powered off the system shortly after reading the alert to check whether some of the SATA/Power cables may have been problematic, and I noticed that one of the SATA cables was not plugged all the way in.

Plugging the cable seemed to fix the issue because after powering on, the pool was marked as healthy, without disk errors & resilvered.

My question is - how much should I trust that data, once I know the pool already went degraded because of my error? Are there any logs I can use to dig deeper into this?

I know ZFS is supposed to be fool-proof to some degree but I don’t have any backups which I can use to just restore and be confidient that my data is safe.

I know I’ll have to get some shit because of my no-backups-setup, but I just don’t have the finances right now to do that. Thank you.

2

Run a scrub. If all is fine (or corrected) you’re fine.
You’ve identified the hardware issue, and hopefully fixed it.
ZFS is designed to handle that and only ever return validated, trustable, data.

1 Like
3

Yeah, I missed that in my post, but I’m already running a scrub - I’m expecting results tomorrow as they usually take ~6h on this dataset.

Thanks for the reply.

4

Scrub passed with no errors.

Issue is solved I guess. Thank ZFS!

2 Likes