SSD Disconnects randomly

Jhin · 2025-04-15 05:36:10

Hi, I don't entirely know if its an Kernel problem or an Hardware problem but I'm posting to see if I get any direction.

I'm having a problem with my ssd acting like it was removed from my computer at random times, only fix is to shutdown completely and turn back on, if I press the reset button it wont detect the ssd.
I keep a

sudo dmesg -w

terminal open in case it happens to see the logs because when the ssd disconnects it cant write the error log to it.

It happens aprox once a week and the last time it disconnected I could only film the monitor because I can't open files to save the log or open another program to save somewhere on the cloud.
Couldn't find a way to write the entire log as text, so I'm using the yt as a video holder, if it's against the forum rules I could remove it or try an IA text recognition.

YT link of the screen: https://youtu.be/9bsqqQmx6AE
There's a msg telling me to try

nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off

and report a bug, the first two options were in effect at the moment of the problem, I just added the third one now but idk if it will make any effect.

SSD: 01:00.0 Non-Volatile memory controller: ADATA Technology Co., Ltd. ATOM 50, LEGEND 840 NVMe SSD (DRAM-less) (rev 01) [1cc1:613a]

sudo lspci -vvnn

: https://pastebin.com/DCn1gCVt

Last edited by Jhin (2025-04-16 19:27:57)

wannabeeeearchieeee · 2025-04-15 16:01:58

Yeah, I have the same issue since I'm using a laptop.
I'm using Debian, though. I think it's more of a hardware issue—at least in my case.
I have an SSD in a caddy by removing the CD-ROM, and the hard drive is in the main storage bay.

The funny part is, in my case, the SSD doesn't disconnect when I reboot the system through the distro,
but it does disconnect when I reboot using the power button.
The hard drive, on the other hand, never disconnects at any point.

I don’t think the SSD itself is the problem, because I used to run Linux on my desktop using the same SSD,
and I never had a single issue.
So I think it’s more likely related to the caddy or something similar.

Last edited by wannabeeeearchieeee (2025-04-15 17:03:23)

d_fajardo · 2025-04-15 16:46:43

Have you installed nvme-cli? What is the output of nvme smart-log or nvme error-log?
https://wiki.archlinux.org/title/Solid_state_drive/NVMe

Jhin · 2025-04-16 01:04:43

d_fajardo wrote:

Have you installed nvme-cli? What is the output of nvme smart-log or nvme error-log?
https://wiki.archlinux.org/title/Solid_state_drive/NVMe

Yes, here is the output:

❯ sudo nvme smart-log /dev/nvme0
Smart Log for NVME device:nvme0 namespace-id:ffffffff
critical_warning			: 0
temperature				: 114 °F (319 K)
available_spare				: 100%
available_spare_threshold		: 10%
percentage_used				: 2%
endurance group critical warning summary: 0
Data Units Read				: 25975969 (13.30 TB)
Data Units Written			: 32209273 (16.49 TB)
host_read_commands			: 342887942
host_write_commands			: 664001866
controller_busy_time			: 33
power_cycles				: 1524
power_on_hours				: 407
unsafe_shutdowns			: 106
media_errors				: 0
num_err_log_entries			: 0
Warning Temperature Time		: 0
Critical Composite Temperature Time	: 0
Temperature Sensor 1			: 114 °F (319 K)
Temperature Sensor 2			: 113 °F (318 K)
Thermal Management T1 Trans Count	: 0
Thermal Management T2 Trans Count	: 0
Thermal Management T1 Total Time	: 0
Thermal Management T2 Total Time	: 0

nvme error-log: https://pastebin.com/tNaGCd0S

Last edited by Jhin (2025-04-16 16:10:52)

seth · 2025-04-16 07:18:25

only fix is to shutdown completely and turn back on, if I press the reset button it wont detect the ssd.

This is only after ~3h uptime, though? No S3/S4 cycle inbetween?
Is there a parallel OS (windows)?
Notebook? Battery or external power supply?

cat /proc/cmdline

confirms the presence of those parameters?
What else is on that bus?

lspci -tvnn

Jhin · 2025-04-16 15:31:53

seth wrote:

This is only after ~3h uptime, though? No S3/S4 cycle in between?

Well, I haven't detected anything related to uptime, It can be like 1h from boot or like 18h from it. My pc has no sleep/hibernate function active afaik, if I leave the pc for 5h it will be there just waiting turned on, I'm on Hyperland but It used to happen on gnome too.
The only thing I found a slight relationship with the problem was when it was being heavily used, like when downloading a steam game or something, not 100% of the time that I'm downloading something it occurs nor it occurs only during heavy usage, but many of the times was on high read/write.

Is there a parallel OS (windows)?

Yes, but on a separate SATA HD, I never use it and its not even on grub list, I installed windows 100% on the hd, and arch 100% on the ssd, they don't even share the boot partition, to access windows I go into the bios and boot from the HD there.

Notebook? Battery or external power supply?

Desktop pc, no nobreak or enything like that.

cat /proc/cmdline
confirms the presence of those parameters?

❯ cat /proc/cmdline
BOOT_IMAGE=/vmlinuz-linux root=UUID=1ac550ba-f349-409b-b72e-55f6149de7a0 rw zswap.enabled=0 rootfstype=ext4 loglevel=3 quiet splash nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off

What else is on that bus?
lspci -tvnn

❯ lspci -tvnn
-[0000:00]-+-00.0  Advanced Micro Devices, Inc. [AMD] Starship/Matisse Root Complex [1022:1480]
           +-00.2  Advanced Micro Devices, Inc. [AMD] Starship/Matisse IOMMU [1022:1481]
           +-01.0  Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Host Bridge [1022:1482]
           +-01.1-[01]----00.0  ADATA Technology Co., Ltd. ATOM 50, LEGEND 840 NVMe SSD (DRAM-less) [1cc1:613a]
           +-01.2-[02-07]--+-00.0  Advanced Micro Devices, Inc. [AMD] Device [1022:43ec]
           |               +-00.1  Advanced Micro Devices, Inc. [AMD] 500 Series Chipset SATA Controller [1022:43eb]
           |               \-00.2-[03-07]--+-00.0-[04]----00.0  Realtek Semiconductor Co., Ltd. RTL8111/8168/8211/8411 PCI Express Gigabit Ethernet Controller [10ec:8168]
           |                               +-01.0-[05]--
           |                               +-02.0-[06]--
           |                               \-03.0-[07]--
           +-02.0  Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Host Bridge [1022:1482]
           +-03.0  Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Host Bridge [1022:1482]
           +-03.1-[08-0a]----00.0-[09-0a]----00.0-[0a]--+-00.0  Advanced Micro Devices, Inc. [AMD/ATI] Navi 33 [Radeon RX 7600/7600 XT/7600M XT/7600S/7700S / PRO W7600] [1002:7480]
           |                                            \-00.1  Advanced Micro Devices, Inc. [AMD/ATI] Navi 31 HDMI/DP Audio [1002:ab30]
           +-04.0  Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Host Bridge [1022:1482]
           +-05.0  Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Host Bridge [1022:1482]
           +-07.0  Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Host Bridge [1022:1482]
           +-07.1-[0b]----00.0  Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Function [1022:148a]
           +-08.0  Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Host Bridge [1022:1482]
           +-08.1-[0c]--+-00.0  Advanced Micro Devices, Inc. [AMD] Starship/Matisse Reserved SPP [1022:1485]
           |            +-00.1  Advanced Micro Devices, Inc. [AMD] Starship/Matisse Cryptographic Coprocessor PSPCPP [1022:1486]
           |            +-00.3  Advanced Micro Devices, Inc. [AMD] Matisse USB 3.0 Host Controller [1022:149c]
           |            \-00.4  Advanced Micro Devices, Inc. [AMD] Starship/Matisse HD Audio Controller [1022:1487]
           +-14.0  Advanced Micro Devices, Inc. [AMD] FCH SMBus Controller [1022:790b]
           +-14.3  Advanced Micro Devices, Inc. [AMD] FCH LPC Bridge [1022:790e]
           +-18.0  Advanced Micro Devices, Inc. [AMD] Matisse/Vermeer Data Fabric: Device 18h; Function 0 [1022:1440]
           +-18.1  Advanced Micro Devices, Inc. [AMD] Matisse/Vermeer Data Fabric: Device 18h; Function 1 [1022:1441]
           +-18.2  Advanced Micro Devices, Inc. [AMD] Matisse/Vermeer Data Fabric: Device 18h; Function 2 [1022:1442]
           +-18.3  Advanced Micro Devices, Inc. [AMD] Matisse/Vermeer Data Fabric: Device 18h; Function 3 [1022:1443]
           +-18.4  Advanced Micro Devices, Inc. [AMD] Matisse/Vermeer Data Fabric: Device 18h; Function 4 [1022:1444]
           +-18.5  Advanced Micro Devices, Inc. [AMD] Matisse/Vermeer Data Fabric: Device 18h; Function 5 [1022:1445]
           +-18.6  Advanced Micro Devices, Inc. [AMD] Matisse/Vermeer Data Fabric: Device 18h; Function 6 [1022:1446]
           \-18.7  Advanced Micro Devices, Inc. [AMD] Matisse/Vermeer Data Fabric: Device 18h; Function 7 [1022:1447]

Last edited by Jhin (2025-04-16 16:10:28)

seth · 2025-04-16 16:05:07

but on a separate SATA HD, I never use it and its not even on grub list, I installed windows 100% on the hd, and arch 100% on the ssd, they don't even share the boot partition, to access windows I go into the bios and boot from the HD there.

And none of that matters at all.

3rd link below. Mandatory.
Disable it (it's NOT the BIOS setting!) and reboot windows and linux twice for voodo reasons.

like when downloading a steam game or something

The ethernet controller is on the same bus and your internet speed is likely not enough to stress your nvme at all?
https://wiki.archlinux.org/title/Stress … image_file

But see the windows situation first.

Edit: Please use [code][/code] tags. Edit your post in this regard.

Last edited by seth (2025-04-16 16:05:30)

Jhin · 2025-04-17 01:29:02

3rd link below. Mandatory.
Disable it (it's NOT the BIOS setting!) and reboot windows and linux twice for voodo reasons.

I disabled it and did the voodoo thing, I didn't know that another drive could mess up with the ssd, I can even remove the HD if its necessary, really don't use it.

I'm going to try the stress test, It's something I can't reproduce so it's hard to know if it worked.

Edit: Please use [code][/code] tags. Edit your post in this regard.

First time posting here, MB

About the links, the first one about upload text I cant do because when it happens I cant access any command, not even opening a terminal, or doing cat,ls...
The boot without GUI and xinitrc I didn't understand the reason.

seth · 2025-04-17 06:53:44

I didn't know that another drive could mess up with the ssd

This has nothing to do w/ the drives, the UEFI/ACPI gets "confused" about the current state, the hardware shows up in an undefined state, cats are fucking dogs… you're running two OS at the same time and all sorts of stuff can happen now. The network being on the same bus as the nvme is an extra-problem since network (though more specifically wifi/radio) is one of the most commonly affected subsystems.

The links on the bottom of my posts are just my signature, stressing some common problems…

Jhin · 2025-04-18 00:43:51

Just happened again while listening to spotify even after the windows fast startup and hibernation change, do u have any other recommendation? Like maybe remove the HD or anything like that, my motherboard only has 1 nvme slot so the ethernet bus problem I can't fix it.

I really don't know what's happening, I downloaded a stress test and nothing happened, then during 0 load it just disconnects.

Can it be PCIe version? My last ssd was 3.0 and worked great, this is 4.0 on a 3.0 limited mobo

Last edited by Jhin (2025-04-18 00:47:08)

seth · 2025-04-18 07:34:28

so the ethernet bus problem I can't fix it

First step would be to figure whether it's coming from there…

downloaded a stress test and nothing happened, then during 0 load it just disconnects … while listening to spotify

suggests as much…

https://aur.archlinux.org/packages/r8168-dkms - or do you maybe have an ethernet/wifi dongle you could temporarily use (and ideally deactivate the onboard chip in the UEFI/BIOS settings)?

PCIe is supposed to be backward and forward compatible and since it's not related to high load (but either power saving or the bus noise?) it's unlikely insufficient power supply because of the lower standard and flawd compatibility implementation either.

Did you try to disable the iommu, https://wiki.archlinux.org/title/Solid_ … nd_support ?
(w/ the lspci you'd likely eventually get away w/ fullflush, but try "iommu=soft" first.

Arch Linux

#1 2025-04-15 05:36:10

SSD Disconnects randomly

#2 2025-04-15 16:01:58

Re: SSD Disconnects randomly

#3 2025-04-15 16:46:43

Re: SSD Disconnects randomly

#4 2025-04-16 01:04:43

Re: SSD Disconnects randomly

#5 2025-04-16 07:18:25

Re: SSD Disconnects randomly

#6 2025-04-16 15:31:53

Re: SSD Disconnects randomly

#7 2025-04-16 16:05:07

Re: SSD Disconnects randomly

#8 2025-04-17 01:29:02

Re: SSD Disconnects randomly

#9 2025-04-17 06:53:44

Re: SSD Disconnects randomly

#10 2025-04-18 00:43:51

Re: SSD Disconnects randomly

#11 2025-04-18 07:34:28

Re: SSD Disconnects randomly

Board footer