Hi all,

The other day Plex wasn't working and I discovered that my Cache drive (which stores my docker image) was no longer detected by unraid, it had disconnected.

Rebooting unraid fixed it, but the issue occurred again today.

I'm not sure if this is a hardware issue, but I thought before buying another 1TB drive I should post my diagnostics here. Is anyone able t take a look please and see if there's any useful info in there?
I ran a short smart diagnostics on the drive and it completed without error as far as I can see.

Thanks

tower-diagnostics-20250903-2029.zip

Sep 3 09:23:41 Tower kernel: nvme nvme1: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0x10

Sep 3 09:23:41 Tower kernel: nvme nvme1: Does your device have a faulty power saving mode enabled?

Sep 3 09:23:41 Tower kernel: nvme nvme1: Try "nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off" and report a bug

Try adding those options to syslinux.cfg, if you are not sure how to, please ask.

I did that today and rebooted, now my dockers won't start and my shares are not showing.

I removed the lines (that I had added) and rebooted again, still having massive issues. I think the issue i'm experiencing now is a coincidence and unrelated to the lines that I added, but I could be wrong.

I've attached new diagnostics, can someone please review?

tower-diagnostics-20250918-2039.zip

Fix common problems shows my cache as read only or full...it's not full though. Have >500GB free.

Edited September 18, 2025Sep 18 by mintjberry

The device is still dropping offline, but the kernel options are not being used at the moment:

Sep 18 20:33:50 Tower kernel: nvme nvme0: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0x10

Sep 18 20:33:50 Tower kernel: nvme nvme0: Does your device have a faulty power saving mode enabled?

Sep 18 20:33:50 Tower kernel: nvme nvme0: Try "nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off" and report a bug

Sep 18 20:33:50 Tower kernel: nvme 0000:01:00.0: enabling device (0000 -> 0002)

Sep 18 20:33:50 Tower kernel: nvme nvme0: Removing after probe failure status: -19

I think I've added the line correctly now and have rebooted..

My shares still aren't showing. Having docker issues as well which I'm sure is related. New diagnostics attached.

tower-diagnostics-20250919-1138.zip

Further to this, when I stop the array now TWO drives are missing, both are NVME I think.

It's weird as they only show as missing when I stop the array. They seem to start fine with unraid and don't display any issues until I stop the array.

More info - my array is taking forever to start, looks like there's USB issues

I moved the unraid flash drive to a USB port, and also switch my USB A to C portable monitor cable from a USB 2.0 port to a USB 3 (was possibly trying to pull too much power and causing USB instability issues on my motherboard).

Now everything is working properly.

Sep 19 10:37:08 Tower kernel: nvme nvme1: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0x10

Sep 19 10:37:08 Tower kernel: nvme nvme1: Does your device have a faulty power saving mode enabled?

Sep 19 10:37:08 Tower kernel: nvme nvme1: Try "nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off" and report a bug

Sep 19 10:37:08 Tower kernel: nvme1n1: I/O Cmd(0x2) @ LBA 434207336, 32 blocks, I/O Error (sct 0x3 / sc 0x71)

Sep 19 10:37:08 Tower kernel: I/O error, dev nvme1n1, sector 434207336 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 2

Sep 19 10:37:08 Tower kernel: nvme 0000:04:00.0: enabling device (0000 -> 0002)

It's still dropping; I suggest trying a different brand/model device, or a different board.

Sorry to revive this hopefully thats ok but I'm having very similar issue to this out of nowhere so for you it ended up being some issue with USB ports and what was plugged into it causing nvme controller to crash?

I switched nvmes even and the problem persists. Thanks