This has been going on for a few months, where the log filesystem will completely fill up and CPU is consistently at 100%. It happens randomly and is only temporarily fixed by performing a reboot. Sometimes it is fixed for a week, sometimes longer.

The log is filled with the following errors:

Sep 14 02:03:56 tempest kernel: md: disk0 read error, sector=10739486256

Sep 14 02:03:56 tempest kernel: md: disk0 read error, sector=8943333208

Sep 14 02:03:56 tempest kernel: md: disk0 read error, sector=8943333008

Sep 14 02:03:56 tempest kernel: md: disk0 read error, sector=4962140440

Sep 14 02:03:56 tempest kernel: md: disk0 read error, sector=10739486256

Sep 14 02:03:56 tempest kernel: md: disk0 read error, sector=10739486256

Sep 14 02:03:56 tempest kernel: md: disk0 read error, sector=8943333016

Sep 14 02:03:56 tempest kernel: md: disk0 read error, sector=8943333216

Sep 14 02:03:56 tempest kernel: md: disk0 read error, sector=4962140248

Sep 14 02:03:56 tempest kernel: md: disk0 read error, sector=10739486256

Sep 14 02:03:56 tempest kernel: md: disk0 read error, sector=4962140448

Sep 14 02:03:56 tempest kernel: md: disk0 read error, sector=10739486256

This would indicate to me that my parity disk is potentially failing? However, after doing an extended SMART self-test, nothing immediately jumped out at me so I am skeptical.

Before I chalk this up to a bad drive and replace it, is there anything I'm missing or should test first?

Appreciate any help.

ST14000NM0288_E_ZHZ208XX0000G841014N_35000c500a7c0ca7f-20250919-2238 3.txt

Edited September 20, 2025Sep 20 by Bust

Please post the diagnostics.

Apologies for not adding in the original post. I’ve attached the diagnostics.

tempest-diagnostics-20250920-0947.zip

Not seeing any actual read errors logged so far, I assume you rebooted, but there are still what look like power/connection issues logged with parity. Replace its cables or change to a different slot and post new diags.

Yes, I had rebooted. I've not noticed there to be read errors a few at a time with the parity disk, the only time I see the read errors is when it's completely flooding the logs. An all or nothing situation if you will.

I swapped my parity drive to a known good slot in my JBOD enclosure. Posting diags again.

tempest-diagnostics-20250921-1318.zip

Not seeing any issues with parity so far, there are some with sdw, but it looks like that is not assigned.

I'll keep an eye on it, recently it hasn't gone more than a few weeks without the issue happening. If nothing happens in that time, I'll consider this solved as a bad power or data cable. Thanks for your help and potentially saving me from buying a new drive :)