Truth be told I'm not 100% sure if this is due to 6.9.x, but the issues did start when I upgraded so I figured I'd give it a shot in here. Prior to 6.9.x, I had zero locks since deploying my current hardware from a year ago. The only material config change since the upgrade was that I began using my nvme drive as my primary cache drive, whereas before it was sitting idle in my server via unassigned devices. Also I no longer use unassigned devices but single- and multi-disk pools.
At least one of the prior locks showed a CPU_CATERR event via my IPMI, but the last one did not (the forum thread on my motherboard had others running into this with prior BIOS versions and Intel Turbo Boost enabled, but I never experienced it). I just lost all connectivity and was unable to access the system via IPMI remote control.
After one of the prior locks, I began running a local syslog server on the unraid server. This is what was captured prior to the lock:
https://pastebin.com/raw/afA6Wd5a
The only thing that stands out to me is CPU tainted errors. It does look like something similar happened earlier in the day, but the system continued to function. When this lock occurred, I believe there was writes occurring to the cache drive. I have since disabled turbo boost in the Tips and Tweaks plugin, and removed some plugins that are not commonly used.
Diagnostics attached.
Recommended Comments
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.