- Annoyance
I don't know enough to give much detail besides the nvme is dropping out constantly now. Immediately after a reboot even now. I had 3.5 months of uptime and I think I updated to one of the recent updates. Out of nowhere Plex docker started crashing. I will list the troubleshooting I have now done.
Originally started with a single 4TB acer predator gm7000 nvme for cache drive/pool, due to it sitting directly underneath the a2000 GPU on the motherboard I originally assumed it was overheating and maybe even failing at this point, it would crash once in a while, then every other day.
So I replaced the nvme with a 4TB sabrent rocket4plus I had lying around, known working well I had just used it as a boot drive on a windows pc days before. Exactly the same dropouts and crashing, we're now at the point where I will reboot and it immediately crashes before I can even open the webgui for plex.
I took to unraid forums and was suggested to paste nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off in boot options and if anything this had made it worse.
I went into bios, updated it to the newest version, disabled ASPM globally and entirely. No help.
I have unplugged my UPS from the USB port as one user with similar issue reported this helped them, no change for me.
Decided perhaps the m.2 slot on my motherboard is actually failing, so I ordered a pcie to m.2 adapter and placed the m.2 in it as my motherboard only has a single m.2 slot. This changed nothing as you can see below syslog2 just happened 10 minutes ago to reboot to get plex to work again, which it is not.
So I'm not sure, I suppose I am going to try to revert back to 6.12.something if that's even possible at this point to get back to stable. Maybe unraid isn't even the problem but it doesn't appear to be hardware with it happening identically to different drives that are different brands even, in different slots. On the hardware side the final thing it could be in my opinion is the CPU (12700k) but I am highly dubious of this at this point as everything else seems to work fine.
Any input or help would be greatly appreciated I just hope I can get back to working stable because this is extremely frustrating.
My exact hardware:
ASRock Rack W680D4U-2L2T/G5 Motherboard W680 chipset
Intel i7 12700k CPU stock clocks
2x16GB DDR5 Gskill Trident Z5 6400cl32 with XMP on
5x 16TB Seagate Exos x16 HDDs in regular array (no zfs) one parity
4TB nvme cache SSD, predator GM7000 or Rocket4Plus 4TB
Nvidia RTX A2000 (ampere) GPU for transcodes
Seasonic 1000w Prime Platinum PSU
64GB USB 3.0 Flash drive for unraid
You can ask for more information I will try and check back but I've about lost any hope of help from here. I find loads of threads on the internet for multiple different linux based OSes (proxmox, unraid, truenas) with the same syslog errors and I swear each one is solved in a different way. 