Lexar NM790 2TB

Same here: Lexar NM790 4 TB

The patch didn't apply with the patch file, so I changed it directly in the code file.

Kernel 2.6.2-19 seems to work now.
Unfortunately my patched kernel 2.6.2-19 did not start today.
 
This seems to be the reason why my Proxmox host crashes about every 1-2 days, both with kernel versions 6.2.16-19 and 6.5.3.
Fixed daily crashes on Proxmox host with MSI PRO Z790-P WIFI and Lexar NM790 4TB SSD by updating BIOS. Original BIOS version was early 2023. Now running Linux 6.5.3-1-pve, host stable for 7 days without further modifications. Previous setup with old BIOS and kernels 6.2.16-19 or 6.5.3 experienced daily crashes.

Always good to update BIOS for stability with latest hardware.
 
  • Like
Reactions: fiona
So the good news is that with proxmox Version 8.1-1, i can confirm that the drives are now accressible without problems. Also in the intaller, which is great.

The bad news is that i see periodically this error on all systems with Lexar NM790 .. no crashed or other problems.. but quite concerning, isn't it?
nvme 0000:01:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x000f address=0xfee00000 flags=0x0000]
 
So the good news is that with proxmox Version 8.1-1, i can confirm that the drives are now accressible without problems. Also in the intaller, which is great.

The bad news is that i see periodically this error on all systems with Lexar NM790 .. no crashed or other problems.. but quite concerning, isn't it?
I don´t get this error with proxmox 8.1.3.

But in the last days I got two system freezes without any logging showing the following error on the display. I don´t know if this is a specific NM790 problem because I found similar reports for other nvme drives.
Code:
[...] device-mapper: thin: process_cell: dm_thin_find_block() failed: error = -5
 
Updated kernel to Linux pve 6.5.11-7-pve
Removed the kernel cmd line workaround (nvme_core.default_ps_max_latency_us=0) which worked for me with an older kernel and a similar issue is back.
It takes less than a day for the issue to appear, similar to the way it was before I used the workaround with older kernel.

screen.pngscreen2.pngscreen3.png
 
Last edited:
Updated kernel to Linux pve 6.5.11-7-pve
Removed the kernel cmd line workaround (nvme_core.default_ps_max_latency_us=0) which worked for me with an older kernel and a similar issue is back.
It takes less than a day for the issue to appear, similar to the way it was before I used the workaround with older kernel.
I have the same issues with 6.5.11-7-pve.

Which kernel version an workaround do you use to have a stable system?
 
Last edited:
@fiona the workaround (nvme_core.default_ps_max_latency_us=0) seems to work for the new kernel (Linux pve 6.5.11-7-pve) too. Uptime 2.5 days :)
 
  • Like
Reactions: fiona
@fiona the workaround (nvme_core.default_ps_max_latency_us=0) seems to work for the new kernel (Linux pve 6.5.11-7-pve) too. Uptime 2.5 days :)
I would like to try this too. I will report the result here after a few days.

To understand it clearly, would this the correct entry in my /etc/default/grub file?
Code:
GRUB_CMDLINE_LINUX="nvme_core.default_ps_max_latency_us=0"
 
Where would one put `nvme_core.default_ps_max_latency_us=0` if using systemd-boot?
 
Did anyone try kernel 6.8 provided by Proxmox without the workaround? (with Lexar NM790 or any other SSD with Maxiotech MAP1602A controller)
 
PC: Lenovo Thinkstation P520 (Intel Xeon W-2135)
Proxmox: 8.2.2
Kernel: 6.8.4-2-pve
SSD: same (NM790 2TB)

No need for the workaround.
 

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!