Cache drive suddenly unreadable / unmountable

[Zenichiro]

I've been running Unraid for 18 months / 2 years now and up until I think the release of 6.12.0, it had been completely rock solid for me, literally zero problems with stability. However, since the upgrade, after a short period of time, from 1-3 days, Docker would crash and wouldn't be able to be restarted without a reboot. Parity check would always show no errors after the boot and it'd run fine again. I realised that turning off Docker would prevent any issues, and most recently I noticed that just having a few Docker containers running also seemed to prevent problems. I had it running mostly just the basic requirements for Sonarr etc for the last few weeks or so and it has remained stable.

 

Until today.

 

I just logged into the server and it was showing alerts saying that the cache drive was full / unreadable - although the info bars were showing it still at around 23%. I thought I'd try a reboot to see if it'd fix it - no luck - now it won't mount at all.

 

All the drives were brand new when I built the server and given the timing of the problems began with the upgrade, I don't think it's a hardware issue.

 

What can I do to resolve this? I've attached my diagnostics bundle.

spacestation01-diagnostics-20230813-1810.zip

[JorgeB]

If the log tree is the only issue this may help:

btrfs rescue zero-log /dev/sdf1

Then re-start array

[Zenichiro]

That seems to have worked, it's back online. I'll run through a full parity check and then restart to bring up Docker - fingers cross, thanks! I'll mark once I've run the check.