Skip to contentSkip to conversation

SSD overheating during scrub in laptop

DISCUSSION Martin Steigerwald 2 participants 8 messages Sep 14, 2025 – Nov 14 Btrfs file system

lore
Share & tools
lore
#0 lore raw

Hi!

Just to share an experience and ask whether others have experienced this.

Already during summer it happened, that the Samsung 990 Pro 4 TB SSD in this ThinkPad T14 AMD Gen 5 said goodbye during scrubbing a 2 TB BTRFS filesystem with almost 2 TB of data in big files running at about 3,5 to 4,2 GiB/s. I concluded this being due to excessive heat.

However it still succeeded with the 500 GiB /home with about 300-350 GiB of data in it back then. I worked around it, by scrubbing two minutes, then canceling, waiting to cool down, resuming for two minutes until the 7-9 minutes of scrubbing were complete. I tried to work around with lowered speed, but even then the temperature slowly rose till the SSD said goodbye. I think I went down to 1 GiB/s of speed maximum, maybe even below that. But then the scrubbing takes longer so more time for SSD to heat up. However… I would have thought it would not heat that much with a slower speed. Maybe it would have worked with 50 or 100 MiB/s. But this takes long.

Now I had these SSD goodbyes during regular use in times of heavy I/O and in the end it could not even scrub that /home partition anymore in one go. Linux hung and only way to recover was to forcefully power off the laptop.

I opened the laptop and removed dust with high pressure air can while holding the fan still so it would not generate current. And with disabled laptop battery.

This fixed the SSDs goodbye issue and I could even scrub that 2 TiB filesystem again. However, according to sensors command it still had about 80,8 °C composite temperature and even 100,8 °C for sensor 2 for the NVME SSD at "nvme-pci-0300" shortly before the end of the scrub, with critical for composite at 84,8 °C, but no critical set for sensor 2. That is still quite high. Granted, it took about 7-8 minutes of scrubbing at 3,5 to 4,2 GiB/s in one go for it to heat up like this. But on the other hand it is not summer anymore and room is not as hot as in summer.

Anyone had similar experiences?

My solution to this will be to remove the dust inside laptop about every half year. However… I was a bit surprised that the NVME SSD would not throttle itself before saying goodbye. Or maybe it tried and it was not enough or to late? The laptop is a bit less than 15 months old. So I conclude that it takes less than a year for the cooling system to become quite a bit less effective due to dust. Good old ThinkPad T520 took way longer for that. But it is way larger on the other hand with more space to distribute heat.

I bet it is out of scope for btrfs scrub command to monitor NVME SSD temperature and pause when to high. And I conclude the NVME layer of the kernel is not throttling to hot NVME flash either.

I checked my data filesystems, two BTRFS and one BCacheFS by scrubbing and fsck'ing without changing anything. They are okay.

Best,

Signature · 1 line
Martin
#1 lore raw
Quoted 20 lines · Martin Steigerwald

Hi!

Just to share an experience and ask whether others have experienced this.

Already during summer it happened, that the Samsung 990 Pro 4 TB SSD in this ThinkPad T14 AMD Gen 5 said goodbye during scrubbing a 2 TB BTRFS filesystem with almost 2 TB of data in big files running at about 3,5 to 4,2 GiB/s. I concluded this being due to excessive heat.

However it still succeeded with the 500 GiB /home with about 300-350 GiB of data in it back then. I worked around it, by scrubbing two minutes, then canceling, waiting to cool down, resuming for two minutes until the 7-9 minutes of scrubbing were complete. I tried to work around with lowered speed, but even then the temperature slowly rose till the SSD said goodbye. I think I went down to 1 GiB/s of speed maximum, maybe even below that. But then the scrubbing takes longer so more time for SSD to heat up. However… I would have thought it would not heat that much with a slower speed. Maybe it would have worked with 50 or 100 MiB/s. But this takes long.

Quoted 3 lines

Now I had these SSD goodbyes during regular use in times of heavy I/O and in the end it could not even scrub that /home partition anymore in one go. Linux hung and only way to recover was to forcefully power off the laptop.

Can you setup netconsole and catch the dying message?

I doubt if it's really the drive dying.

Quoted 5 lines

I opened the laptop and removed dust with high pressure air can while holding the fan still so it would not generate current. And with disabled laptop battery.

Quoted 8 lines

This fixed the SSDs goodbye issue and I could even scrub that 2 TiB filesystem again. However, according to sensors command it still had about 80,8 °C composite temperature and even 100,8 °C for sensor 2 for the NVME SSD at "nvme-pci-0300" shortly before the end of the scrub, with critical for composite at 84,8 °C, but no critical set for sensor 2. That is still quite high. Granted, it took about 7-8 minutes of scrubbing at 3,5 to 4,2 GiB/s in one go for it to heat up like this. But on the other hand it is not summer anymore and room is not as hot as in summer.

I have hit similar situation, but the symptom is very different, the death come silently at boot, the drive is no longer recognized by the BIOS thus no longer bootable, and Linux kernel from liveUSB will not recognize it either.

That's why I'm asking if it's really dying caused by the heat.

Quoted 11 lines

Anyone had similar experiences?

My solution to this will be to remove the dust inside laptop about every half year. However… I was a bit surprised that the NVME SSD would not throttle itself before saying goodbye. Or maybe it tried and it was not enough or to late? The laptop is a bit less than 15 months old. So I conclude that it takes less than a year for the cooling system to become quite a bit less effective due to dust. Good old ThinkPad T520 took way longer for that. But it is way larger on the other hand with more space to distribute heat.

You can refer to man page of btrfs-scrub, it provides the way to limit the bandwidth of scrub using cgroup or even the btrfs sysfs interface.

Thanks,
Qu

Quoted 9 lines

I bet it is out of scope for btrfs scrub command to monitor NVME SSD temperature and pause when to high. And I conclude the NVME layer of the kernel is not throttling to hot NVME flash either.

I checked my data filesystems, two BTRFS and one BCacheFS by scrubbing and fsck'ing without changing anything. They are okay.

Best,

#2 lore raw
Quoted 63 lines · Qu Wenruo
Quoted 20 lines · Martin Steigerwald

Hi!

Just to share an experience and ask whether others have experienced this.

Already during summer it happened, that the Samsung 990 Pro 4 TB SSD in this ThinkPad T14 AMD Gen 5 said goodbye during scrubbing a 2 TB BTRFS filesystem with almost 2 TB of data in big files running at about 3,5 to 4,2 GiB/s. I concluded this being due to excessive heat.

However it still succeeded with the 500 GiB /home with about 300-350 GiB of data in it back then. I worked around it, by scrubbing two minutes, then canceling, waiting to cool down, resuming for two minutes until the 7-9 minutes of scrubbing were complete. I tried to work around with lowered speed, but even then the temperature slowly rose till the SSD said goodbye. I think I went down to 1 GiB/s of speed maximum, maybe even below that. But then the scrubbing takes longer so more time for SSD to heat up. However… I would have thought it would not heat that much with a slower speed. Maybe it would have worked with 50 or 100 MiB/s. But this takes long.

Quoted 3 lines

Now I had these SSD goodbyes during regular use in times of heavy I/O and in the end it could not even scrub that /home partition anymore in one go. Linux hung and only way to recover was to forcefully power off the laptop.

Can you setup netconsole and catch the dying message?

I doubt if it's really the drive dying.

Quoted 5 lines

I opened the laptop and removed dust with high pressure air can while holding the fan still so it would not generate current. And with disabled laptop battery.

Quoted 8 lines

This fixed the SSDs goodbye issue and I could even scrub that 2 TiB filesystem again. However, according to sensors command it still had about 80,8 °C composite temperature and even 100,8 °C for sensor 2 for the NVME SSD at "nvme-pci-0300" shortly before the end of the scrub, with critical for composite at 84,8 °C, but no critical set for sensor 2. That is still quite high. Granted, it took about 7-8 minutes of scrubbing at 3,5 to 4,2 GiB/s in one go for it to heat up like this. But on the other hand it is not summer anymore and room is not as hot as in summer.

I have hit similar situation, but the symptom is very different, the death come silently at boot, the drive is no longer recognized by the BIOS thus no longer bootable, and Linux kernel from liveUSB will not recognize it either.

That's why I'm asking if it's really dying caused by the heat.

Quoted 11 lines

Anyone had similar experiences?

My solution to this will be to remove the dust inside laptop about every half year. However… I was a bit surprised that the NVME SSD would not throttle itself before saying goodbye. Or maybe it tried and it was not enough or to late? The laptop is a bit less than 15 months old. So I conclude that it takes less than a year for the cooling system to become quite a bit less effective due to dust. Good old ThinkPad T520 took way longer for that. But it is way larger on the other hand with more space to distribute heat.

Quoted 1 line

You can refer to man page of btrfs-scrub, it provides the way to limit the bandwidth of scrub using cgroup or even the btrfs sysfs interface.

And I forgot my physical solution, with a thick thermal pad, you can connect the NVME to the back plate of the laptop to dissipate heat.

If there is enough space (I doubt for laptop though), you can add some thin heatsink for the drive.

It works fine for one of my laptop, and the drive keeps a healthy temperature since then.

Thanks,
Qu

Quoted 15 lines

Thanks,
Qu

Quoted 9 lines

I bet it is out of scope for btrfs scrub command to monitor NVME SSD temperature and pause when to high. And I conclude the NVME layer of the kernel is not throttling to hot NVME flash either.

I checked my data filesystems, two BTRFS and one BCacheFS by scrubbing and fsck'ing without changing anything. They are okay.

Best,

#3 lore raw

Thanks.

Qu Wenruo - 14.09.25, 23:33:01 CEST:

Quoted 2 lines · Qu Wenruo
Quoted 1 line · Martin Steigerwald

[…]

Quoted 7 lines
Quoted 7 lines
Quoted 5 lines

Now I had these SSD goodbyes during regular use in times of heavy I/O and in the end it could not even scrub that /home partition anymore in one go.
Linux hung and only way to recover was to forcefully power off the laptop.

Can you setup netconsole and catch the dying message?

I suppose I could do that as I have several laptops at hand. However… after cleansing out at least some of the dust it did not happen anymore. And I do not quite feel like putting some dust back or otherwise provoke the issue. I do not quite feel like trying to overheat the SSD again. :) I looked at it more closely, it looks fine enough. Nothing melted apparently, but what do I know? And I think it does not add to the lifetime of the SSD to overheat it on purpose.

I might have made a photo of some earlier time, but I am not sure. I will have a look and see whether I can find any. I remember the kernel wrote a lot about NVME opcodes. However I do not recall the details. Unfortunately at this recent occasion I did not make a photo.

In case it happens again by itself, I am at going for a netconsole log – in case I can reproduce it then. Otherwise I would need to run netconsole permanently to be sure to catch it on the first occurrence already. Not sure whether that is a good idea.

Quoted 1 line
Quoted 1 line

I doubt if it's really the drive dying.

Why? The issue went away after removing at least some of the dust. Of course that is a correlation and not necessarily a causation, but what makes you think that something different is happening?

I checked the drive smart status. SMART status is passed. So everything okay. There is 2% of spare area used, but that is still a very good value.

An indication that there is something to your suspicion is:

Media and Data Integrity Errors:    0
Error Information Log Entries:      0
Warning  Comp. Temperature Time:    0
Critical Comp. Temperature Time:    0

So the drive itself recorded no critical composite temperature times. And if it has fields for that, I suppose it could still record it on overheat condition.

Oh by the way, it is a Samsung 990 Pro with a firmware version above that firmware version that was known to be broken regarding SMART status reporting. In the first I bought I made sure of that myself, but this one came with a newer firmware version already.

Quoted 14 lines
Quoted 14 lines
Quoted 13 lines

I opened the laptop and removed dust with high pressure air can while holding the fan still so it would not generate current. And with disabled laptop battery.

This fixed the SSDs goodbye issue and I could even scrub that 2 TiB filesystem again. However, according to sensors command it still had about 80,8 °C composite temperature and even 100,8 °C for sensor 2 for the NVME SSD at "nvme-pci-0300" shortly before the end of the scrub, with critical for composite at 84,8 °C, but no critical set for sensor 2. That is still quite high. Granted, it took about 7-8 minutes of scrubbing at 3,5 to 4,2 GiB/s in one go for it to heat up like this. But on the other hand it is not summer anymore and room is not as hot as in summer.

Quoted 4 lines
Quoted 4 lines

I have hit similar situation, but the symptom is very different, the death come silently at boot, the drive is no longer recognized by the BIOS thus no longer bootable, and Linux kernel from liveUSB will not recognize it either.

Hmm, did you capture any logs?

Quoted 1 line
Quoted 1 line

That's why I'm asking if it's really dying caused by the heat.

Ah okay… You mentioned your physical solution being a thermal pad. So did this symptom go away with it? Or was your physical solution a general solution to reduce temperature, a solution unrelated to that symptom you described?

Quoted 9 lines
Quoted 9 lines
Quoted 8 lines

My solution to this will be to remove the dust inside laptop about every half year. However… I was a bit surprised that the NVME SSD would not throttle itself before saying goodbye. Or maybe it tried and it was not enough or to late? The laptop is a bit less than 15 months old. So I conclude that it takes less than a year for the cooling system to become quite a bit less effective due to dust. Good old ThinkPad T520 took way longer for that. But it is way larger on the other hand with more space to distribute heat.

Quoted 5 lines
Quoted 2 lines

You can refer to man page of btrfs-scrub, it provides the way to limit the bandwidth of scrub using cgroup or even the btrfs sysfs interface.

And I forgot my physical solution, with a thick thermal pad, you can connect the NVME to the back plate of the laptop to dissipate heat.

Well it has some violet colored pad below the NVME SSD. So a pad that is between NVME SSD and motherboard. Not sure whether it is a cooling pad. It might be more about isolating the motherboard from the heat of the SSD? It has been there for the initial SSD Lenovo put into the laptop that I replaced with the larger Samsung 990 Pro one.

But yeah, I may order a cooling plate to connection to the back plate. Good idea.

Quoted 2 lines

If there is enough space (I doubt for laptop though), you can add some thin heatsink for the drive.

Ha ha, no way. I would like to put a 8 TB SSD in there, but it would be double layered and I doubt it would be a good idea to put a double layered SSD into the laptop.

Best,

Signature · 1 line
Martin
#4 lore raw
Quoted 71 lines · Martin Steigerwald

Thanks.

Qu Wenruo - 14.09.25, 23:33:01 CEST:

Quoted 2 lines · Qu Wenruo
Quoted 1 line · Martin Steigerwald

[…]

Quoted 7 lines
Quoted 7 lines
Quoted 5 lines

Now I had these SSD goodbyes during regular use in times of heavy I/O and in the end it could not even scrub that /home partition anymore in one go.
Linux hung and only way to recover was to forcefully power off the laptop.

Can you setup netconsole and catch the dying message?

I suppose I could do that as I have several laptops at hand. However… after cleansing out at least some of the dust it did not happen anymore. And I do not quite feel like putting some dust back or otherwise provoke the issue. I do not quite feel like trying to overheat the SSD again. :) I looked at it more closely, it looks fine enough. Nothing melted apparently, but what do I know? And I think it does not add to the lifetime of the SSD to overheat it on purpose.

I might have made a photo of some earlier time, but I am not sure. I will have a look and see whether I can find any. I remember the kernel wrote a lot about NVME opcodes. However I do not recall the details. Unfortunately at this recent occasion I did not make a photo.

In case it happens again by itself, I am at going for a netconsole log – in case I can reproduce it then. Otherwise I would need to run netconsole permanently to be sure to catch it on the first occurrence already. Not sure whether that is a good idea.

Quoted 1 line
Quoted 1 line

I doubt if it's really the drive dying.

Why? The issue went away after removing at least some of the dust. Of course that is a correlation and not necessarily a causation, but what makes you think that something different is happening?

I checked the drive smart status. SMART status is passed. So everything okay. There is 2% of spare area used, but that is still a very good value.

An indication that there is something to your suspicion is:

Media and Data Integrity Errors:    0
Error Information Log Entries:      0
Warning  Comp. Temperature Time:    0
Critical Comp. Temperature Time:    0

So the drive itself recorded no critical composite temperature times. And if it has fields for that, I suppose it could still record it on overheat condition.

Oh by the way, it is a Samsung 990 Pro with a firmware version above that firmware version that was known to be broken regarding SMART status reporting. In the first I bought I made sure of that myself, but this one came with a newer firmware version already.

Quoted 14 lines
Quoted 14 lines
Quoted 13 lines

I opened the laptop and removed dust with high pressure air can while holding the fan still so it would not generate current. And with disabled laptop battery.

This fixed the SSDs goodbye issue and I could even scrub that 2 TiB filesystem again. However, according to sensors command it still had about 80,8 °C composite temperature and even 100,8 °C for sensor 2 for the NVME SSD at "nvme-pci-0300" shortly before the end of the scrub, with critical for composite at 84,8 °C, but no critical set for sensor 2. That is still quite high. Granted, it took about 7-8 minutes of scrubbing at 3,5 to 4,2 GiB/s in one go for it to heat up like this. But on the other hand it is not summer anymore and room is not as hot as in summer.

Quoted 6 lines
Quoted 4 lines
Quoted 4 lines

I have hit similar situation, but the symptom is very different, the death come silently at boot, the drive is no longer recognized by the BIOS thus no longer bootable, and Linux kernel from liveUSB will not recognize it either.

Hmm, did you capture any logs?

Nothing suspicious at all, except no PCIE device detected at all.

So in my case it's really toasted so that it can not even be detected anymore.

Quoted 5 lines
Quoted 2 lines
Quoted 2 lines

That's why I'm asking if it's really dying caused by the heat.

Ah okay… You mentioned your physical solution being a thermal pad. So did this symptom go away with it?

The nvme drive went away literally, and never came back.

Quoted 3 lines

Or was your physical solution a general
solution to reduce temperature, a solution unrelated to that symptom you described?

Unrelated to my long-gone drive, but a new drive never suffered critical high temperature anymore.

Thanks,
Qu

Quoted 33 lines
Quoted 10 lines
Quoted 10 lines
Quoted 9 lines

My solution to this will be to remove the dust inside laptop about every half year. However… I was a bit surprised that the NVME SSD would not throttle itself before saying goodbye. Or maybe it tried and it was not enough or to late? The laptop is a bit less than 15 months old. So I conclude that it takes less than a year for the cooling system to become quite a bit less effective due to dust. Good old ThinkPad T520 took way longer for that. But it is way larger on the other hand with more space to distribute heat.

Quoted 5 lines
Quoted 2 lines

You can refer to man page of btrfs-scrub, it provides the way to limit the bandwidth of scrub using cgroup or even the btrfs sysfs interface.

And I forgot my physical solution, with a thick thermal pad, you can connect the NVME to the back plate of the laptop to dissipate heat.

Well it has some violet colored pad below the NVME SSD. So a pad that is between NVME SSD and motherboard. Not sure whether it is a cooling pad. It might be more about isolating the motherboard from the heat of the SSD? It has been there for the initial SSD Lenovo put into the laptop that I replaced with the larger Samsung 990 Pro one.

But yeah, I may order a cooling plate to connection to the back plate. Good idea.

Quoted 2 lines

If there is enough space (I doubt for laptop though), you can add some thin heatsink for the drive.

Ha ha, no way. I would like to put a 8 TB SSD in there, but it would be double layered and I doubt it would be a good idea to put a double layered SSD into the laptop.

Best,

#5 lore raw

Re: SSD overheating during scrub in laptop (SOLVED: no overheating but firmware issue)

Greetings Qu, greetings.

Qu Wenruo - 16.09.25, 23:17:18 CET:

Quoted 6 lines
Quoted 6 lines
Quoted 5 lines
Quoted 5 lines

I have hit similar situation, but the symptom is very different, the death come silently at boot, the drive is no longer recognized by the
BIOS thus no longer bootable, and Linux kernel from liveUSB will not recognize it either.

Quoted 6 lines
Quoted 1 line

Hmm, did you capture any logs?

Nothing suspicious at all, except no PCIE device detected at all.

So in my case it's really toasted so that it can not even be detected anymore.

I think I finally got this sorted.

You have been right, Qu: It was no overheating issue.

Just completed a scrub of 2 TiB data. Twice in a row, first on TTY without loaded desktop with about 3 GiB/s, then on desktop with power-profiles- daemon set to performance mode with about 3,8 GiB/s.

Sensor 2 of SSD temp went up to 101 degree Celsius and it was no issue.

What did I change? I upgraded firmware from 4B2QJXD7 to 8B2QJXD7 after in a dialog with AI I was hinted about issues that users lost their device at least temporarily after unsafe shutdown. Or even needed to send them in for repair. Luckily I never had one of these.

However they also wrote something interesting about read-operation stability:

-----------------------------------------------------------------------

* (8B2QJXD7) To improve read-operation stability. (Release : December 2025)

* (7B2QJXD7) To address the intermittent non-recognition and blue screen issue. (Release: September 2025)

* (4B2QJXD7) To address reports of high temperatures logged on Samsung Magician. (Release: December 2024)

*990 PRO I 990 PRO with Heatsink will be manufactured using a mixed production between the V7 and V8 process starting September 2023.

-----------------------------------------------------------------------

https://semiconductor.samsung.com/consumer-storage/support/tools/

What can I say?

With 4B2QJXD7 a scrub of more than 1 TiB data was always interrupted with SSD being gone until forced restart of the machine, unless I paused and resumed it myself quite an amount of times.

With 8B2QJXD7 two scrubs of 2 TiB in a row completed without any issue.

Happy to have this sorted. Thanks for firmware update to Samsung!

So make a backup and update your firmware!

What a mess, really: Just took some years until I finally seem to have a fully working product. And it has not been the first firmware issue I have seen. Not even the first issue regarding this SSD model. Also I still remember "no operating system found" with Intel SSD 320 after unsafe shutdown with a firmware that should even fix this issue and despite this SSD even having some sort of Power Loss Protection (PLP) unlike the Samsung 990 Pro.

Another note to Samsung: Pretty please update the way you create your firmware ISOs. Kernel 4.5.3 and it does not even boot on my laptop? You gotta be kidding me. Sure I can extract the update tool from initrd and run it manually, but really? How about just going with Linux Vendor Firmware Service via fwupd? Then I would have seen this update much earlier.

Best,

Signature · 1 line
Martin
#6 lore raw

Re: SSD overheating during scrub in laptop (SOLVED: no overheating but firmware issue)

Martin Steigerwald - 14.11.26, 11:29:30 CET:

Quoted 3 lines

Just completed a scrub of 2 TiB data. Twice in a row, first on TTY without loaded desktop with about 3 GiB/s, then on desktop with power-profiles- daemon set to performance mode with about 3,8 GiB/s.

A third successful attempt:

Rate: 4.22GiB/s

So it really seems read stability has improved.

With an BCacheFS filesystem with 180 GiB I have even seen rates of up to 5,5 GiB/s during scrub.

Best,

Signature · 1 line
Martin
#7 lore raw

Re: SSD overheating during scrub in laptop (SOLVED: no overheating but firmware issue)

Martin Steigerwald - 14.11.26, 12:05:24 CET:

Quoted 2 lines

With an BCacheFS filesystem with 180 GiB I have even seen rates of up to 5,5 GiB/s during scrub.

Make that 6.1 GiB/s.

Signature · 1 line
Martin