NVME Cache Drive disappearing

[holycrapamelon]

Hey all,

 

First time poster. Running Unraid less than a week. Really pleased with it so far, until I ran into a significant snag.

 

My NVME SSD cache drive will go offline at random intervals. It's a Crucial P1 1TB PCIe M.2 SSD that is mounted directly to the mobo in a dedicated slot (not a PCIe x4 adapter). When I cold boot the server, it comes online and works fine but after a few hours (today, it was when I woke up and noticed) the browser will issue a warning that the drive cannot be found like I yanked the thing out. A green status light on the physical SSD starts blinking and Unraid cannot see it at all. I don't have a SMART report for this drive at the moment, but I'll generate one when I can get home. Something else to note: soft rebooting the server does not bring the SSD back online. Only entirely cutting power to the mobo and cold booting allows the drive to be seen again. 

 

I'm running 9 or so dockers on the SSD, and have my downloads folder caching. SSD Trim and Folder Caching apps from CA are running. I stress tested it a bit when I migrated my Plex metadata folder onto it (over 300,000 small files). It shot up to 70 degrees C pretty quickly, but I've seen that from M.2 drives before so I didn't think anything of it. The first time this happened was during the Plex migration. It seized up and froze me out of several areas of the WebUI for Unraid, like dockers, VMs, and the dashboard. Required a cold boot to come back online. It appears like this now:

 

 

I can view the contents of it, but cannot access them. Docker appears to be running my containers, but none of them are accessible. 

 

I'm fairly certain this is a bad SSD, but I wanted to see if anyone has experienced anything that could be causing this in the OS.

[JorgeB]

Please post the diagnostics: Tools -> Diagnostics