You are not logged in.
My wife's computer (IntelPentium CPU N3700 @ 1.60GHz running LXDE) freezes randomly some time in a day, no matter what she is doing.
The cursor doesn't respond to mouse, the clock on desktop stops, the caps-lock and block-num leds don't change state on key press, access via ssh hangs even if it was alredy active.
The only possibility is the reset button.
The log file reports nothing.
The 8 GB RAM is fine since I ran eight hours of memtest without errors.
CPU temperature is quite cold.
I tried running Mint21.1 xfce live and it froze too.
I tried kernel downgrading with no success.
Ssd passed health test with smartmontools
The kernel version is:
$ uname -a
Linux archlinux-usb 6.2.10-arch1-1 #1 SMP PREEMPT_DYNAMIC Fri, 07 Apr 2023 02:10:43 +0000 x86_64 GNU/Linuxthe graphic controller is:
grep -A 3 "8086:22b1"
00:02.0 VGA compatible controller [0300]: Intel Corporation Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Integrated Graphics Controller [8086:22b1] (rev 21)
Subsystem: ASRock Incorporation Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Integrated Graphics Controller [1849:22b1]
Kernel driver in use: i915
Kernel modules: i915To exclude that the cause of the failure is the video driver I uninstalled xf86-video-intel (X.org Intel i810/i830/i915/945G/G965+ video drivers) despite this, the kernel still loads the i915 module, even if I blacklisted it:
# cat /etc/modprobe.d/blacklist.conf
blacklist i915If i try to disable i915 with kernel boot parameter module_blacklist=i915 , boot hangs.
So I need some help!
Last edited by ziomarco (2023-07-14 17:03:51)
Offline
Enable Sysrq and check whether rebooting with the REISUB sequence works (pause a couple of seconds between each press) and post a journal from the boot after the failing condition, i.e.
sudo journalctl -b-1for the previous boot https://wiki.archlinux.org/title/List_o … n_services
Some more general notes, why is the kernel that old and are you literally running said install from the USB - or is that the output of a rescue attempt with the live system? And the xorg ddx is completely independent of the kernel module so it still getting loaded is not further surprising. The "general" way to try to limit the gpu to a minimal feature set not allowing more than a console prompt is adding nomodeset to the kernel parameters (and potentially systemd.unit=multi-user.target to boot to a terminal from the get go)
Offline
- The kernel is so old because i supposed the problem was related to some new kernel feature, so i attempted to resolve downgrading it. Obviously i can do a pacman -Syu
- Regarding REISUB : the keyboard is completely dead when system freezes, so i didn't try it, I enabled it just now:
# cat /proc/sys/kernel/sysrq
1- The system is normally installed on SSD, the "archlinux-usb" in the bash prompt is only a confusing setting of my hostname command....i'll change it with something less misleading...like archlinux2
here one complete session log
i can see many recurring message like:
lug 10 14:21:28 archlinux-usb rtkit-daemon[457]: Supervising 2 threads of 1 processes of 1 users.maybe is it the problem?
Last edited by ziomarco (2023-07-10 21:35:07)
Offline
Those are mostly created by firefox creating differently prioritized threads for media on webpages and the like and should not be inherently the cause, but FWIW are you always having firefox open when the issue happens? Can you test chromium or so?
Don't see anything obviously off here, that is from a crashing session right?
Last edited by V1del (2023-07-10 22:44:36)
Offline
Yes, that log is a whole crashed session.
I'll try to use midori instead of firefox waiting the next freeze...
The most of time that computer is used for browsing
Offline
Midori is really too slow, practically unusable...I fall back to chrome although it's quite similar to firefox.
meanwhile i updated, so i'm also using the newest kernel, because there was no reason to keep an old one if the kernel is not directly responsible of the issue.
Today i plan to "stress test" the N3700 with an high intensity browsing session, and we'll see what happens
Offline
Using chrome the issue seems to be solved. No more crash today....
Edit 2023-07-14: The problem hasn't occurred in the last two days, so i mark it as solved.
Last edited by ziomarco (2023-07-14 17:10:27)
Offline