Second time my laptop is burned by neon

Help
, ,
1

Dear all,

I usually don’t rant on forums, but this time I think this is very much justified. :frowning: :enraged_face:

I use Framework laptops since their debut, and that is the second time the motherboard is fried by a bad power management behavior of KDE neon.

Of course, this can be due to my doing, but I think there are some fundamental mechanisms that should prevent the laptop from not going into sleep, or randomly rebooting and getting stuck in SDDM with full power.

It is very bad and destroyed a lot of parts on the motherboard of my laptop, in just enough time for a daily commute (~1h). It happened at least 5 times, and 2 were long enough to fly the electronics. This cost me more than 800€ so far. :frowning:

The problem is the same, I close the lid, put the laptop into a sleeve, put the sleeve in the backpack, somehow it does not go to sleep, or even reboot (probably because of overheating).

Then it is so hot (the sleeve is concentrating the heat of course), that the chips like the USB-c controller or the CPU are fried. Why? When I close the lid, it is not supposed to go to sleep/low power mode?

I use it everyday, and 99.9% of the time it works fine, but when it does not, the problems happen.

Never have a screen connected in such situation, no power cable connected. The laptop just does not go to sleep. Does this happen to anyone else?

To me this is a major problem that makes me reconsider using neon.

2

This is user error so you and you alone did the damage. This is two times in two days I’ve seen someone blame Linux for bad hardware.

3

I think that is one of the options.

Currently not on my laptop but it will go into sleep if the lid is closed while on battery but not do that while connected to power (in which case it only locks the screen and powers the display off).

Since this has worked super reliably since 2018 I am wondering if you might have some process running that inhibits sleep?

3 Likes
4

I am 99% convinced it is a crypttab/drakut problem, otherwise I wouldn’t allow me to rant, honestly. It would also mean Neon isn’t to blame. The lid sensor should just force shutdown, since the kernel isn’t loaded.

But neon should not reboot randomly either! And that is across the ubuntu handling of kernel/sysd. So my apologies, it is a matter beyond just neon.

The mitigation I have to use is to disable TPM PIN request on boot, removing a layer of security to machine access. I wish I don’t have to expect the worst again.

5

Running a bunch of test at the moment. Might update if there is something new.

1 Like
6

Don’t know about drakut but on my laptop I’ve been using crypttab to automatically unlock a second internal drive during startup (after manually unlocking the primary disk at boot).

In my case the computer is a Tuxedo InfinityBook 13 (version 2 if I remember correctly), so there could be hardware differences in play as well.

7

Dracut is a fairly recent “TPMed-replacment” of initramfs. But I cannot test the problem in such case. One would need to have a / partition with LUKS.

8

Interesting, I will have to look into this.

Using LUKS for root partition on both laptop and workstation, with the former being an older setup with separated /boot

9

That CSM and Secure Boot are ALL designed for one thing and one thing only, and that is to try to lock your hardware to a specific copy of one’s OS. This is because Microsoft knows their OS is sh_t and desperate to stop people from using other OS’s.

10

Your description of your problem sounds hardware related, specifically your bios. The power management functions are managed by the bios firmware, as are the thermal protection sensors that should be preventing the system from overheating. You are having multiple failures at once it seems. Given you have the same problem with the same brand and model suggests there is either a defect in the chip or a bug in the firmware. Have you looked to see if there is a bios update for your system?

Also, if the CPU in the laptop is a 13th or 14th gen Intel CPU then it may be suffering from the well known degradation problem.

1 Like
11

Some hardware doesn’t power off correctly when told to hibernate. My 2019 Dell laptop is one. If you’ve ruled out processes preventing it hibernating, you might need to use the trick specified in the Arch wiki here:

12

Thanks all for your comments. One of the major problem is the unpredictability and the frequency of this events. It makes debugging a pain.

1 Like
13

i am not sure this software issue or only issue
the first thing is KDE neon isn’t supported by framework officially so you wouldn’t be able to get officiall suport on that

secound thing is the laptop firmware is responsible for protecting the laptop not the OS so even if the OD gone rough the Firmware should stop it

my advice is use one of the officially supported distro and work with framework team directly there is something seriously wrong here

14

i want to add my putting the laptop in sleep and out it in the bag is bad practice in general although in theory it shouldn’t go out of sleep but life happens i would advice putting it in hibernate

because there is no fresh air to cool the system so you gamble every thing on the laptop not going out of sleep

1 Like
15

My money would be shutting it down for travel, no sleep or hibernate. Personally I can wait the minute for it to boot if it means protecting it from overheating during travel.

2 Likes
16

hibernate is shutdown too the difference it saves the session on disk before shutting down

17

So is Save Session. Smarter move that allows for complete shutdown. And no hibernation is not a full shutdown.

18

no you’re completely wrong i am not planing to enter a long debate i will just Quote arch wiki and leave a link if anyone want to understand more

Blockquote
Suspend to disk (aka hibernate)
The S4 sleeping state as defined by ACPI. Saves the machine’s state into swap space and completely powers off the machine. When the machine is powered on, the state is restored. Until then, there is zero power consumption.

19

I wonder why you don’t add a heater in the backpack as well. [/sarcasm off]

I never use sleep mode , I always switch the laptop off completely and I never have problems with an overheated laptop, plus the battery is not drained in vain. Try it, see how well it works.

20

Yes, why not! :slight_smile:

The sleep funtion on these AMD chips is relarcable, it does not heat-up at all, sleep time is over a week… Hence I believe my way of usage is valid.

I had macbooks for many years, and it was never a problem to do this with them.