kiwijunglist Posted June 27 Share Posted June 27 (edited) Hi A couple of times I noticed all my dockers were down. When I went into unraid gui I noticed the cache_nvme drive was still listed, but the temp is * instead of a number, and the share that uses the "cache_nvme" drive is missing and I can't do smart tests etc. If I reboot the system the cache_nvme drive comes back and everything works fine. I have two nvme drives "Cache_nvme" and "Samsung Pro". "Cache_nvme" which is an Intel 670P 1TB nvme drive seems to be having some problems? My other nvme drive is Samsung 980 Pro 1TB. This has happened 2 or 3 times over the last 3 months. I have attached my diagnostics file. No issues with smart testing of the nvme drive (attached) I have also down some googling and added the following to the flash syslinux configuration: I'm not sure if this powersaving disabling tweak will solve my issue (or not). Can anyone please give me any insight into the issue? Thanks manitoba-diagnostics-20240121-1712.zip cache_nvme - smart test download.txt Edited June 27 by kiwijunglist Quote Link to comment
JorgeB Posted June 27 Share Posted June 27 29 minutes ago, kiwijunglist said: I'm not sure if this powersaving disabling tweak will solve my issue (or not). It's worth a try, you can also try a different m.2 slot if available, even if you have to swap with a different device, it has been known to help. 1 Quote Link to comment
kiwijunglist Posted June 27 Author Share Posted June 27 Yes thanks. That is my next step (swap m2 slot). Could also try swap the samsungpro and the intel 670p around, and if I start getting an issue with the samsungpro then I know it's not the intel drive. Quote Link to comment
kiwijunglist Posted July 5 Author Share Posted July 5 It happened again, "cache_nvme" drive is "offline". manitoba-diagnostics-20240705-1806.zip Quote Link to comment
JorgeB Posted July 5 Share Posted July 5 Jul 5 02:13:56 manitoba kernel: nvme nvme0: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0x10 Jul 5 02:13:56 manitoba kernel: nvme nvme0: Does your device have a faulty power saving mode enabled? Jul 5 02:13:56 manitoba kernel: nvme nvme0: Try "nvme_core.default_ps_max_latency_us=0 pcie_aspm=off" and report a bug Still dropping, if the kernel option doesn't help recommend trying a different model device (or a different board), assuming m.2 slots have already been swapped. Quote Link to comment
Recommended Posts
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.