You are not logged in.
Hello all,
Since the last update I did, my computer freezes out of nothing. As I cannot run any command or perform any action other than turning the computer off and resetting it again, I am not able to locate processes or issues that cause this to happen.
Such an issue had not happened previous to the installation. Therefore, I suspect it has to do with the last update I did on 02.01.2021. One thing to consider is that this happened 4 times whilst actively browsing on firefox, once opening a program to build diagrams dia and once while passively listening to a YouTube video. The Ram usage on the last two events where 1.9G and 1.1G respectively, since I pushed the computer to higher RAMs I discarded problems regarding overuse of RAM. I also have been monitoring the use of CPU and I do not encounter any anomaly either.
I have to say I am very confused since this problem is extremely annoying and I cannot escape the freeze invoking killcomands or similar, basically because nothing responds.
Find also below the list of upgraded packages:
[2021-01-02T18:41:27+0100] [ALPM] upgraded tzdata (2020e-1 -> 2020f-1)
[2021-01-02T18:41:27+0100] [ALPM] upgraded gmp (6.2.0-1 -> 6.2.1-1)
[2021-01-02T18:41:27+0100] [ALPM] upgraded autoconf (2.69-7 -> 2.70-1)
[2021-01-02T18:41:27+0100] [ALPM] upgraded git (2.29.2-1 -> 2.30.0-1)
[2021-01-02T18:41:27+0100] [ALPM] upgraded harfbuzz (2.7.3-1 -> 2.7.4-1)
[2021-01-02T18:41:27+0100] [ALPM] upgraded iproute2 (5.10.0-1 -> 5.10.0-2)
[2021-01-02T18:41:27+0100] [ALPM] upgraded python-setuptools (1:51.0.0-1 -> 1:51.1.1-1)
[2021-01-02T18:41:27+0100] [ALPM] upgraded jupyter-nbconvert (5.6.1-3 -> 6.0.7-1)
[2021-01-02T18:41:27+0100] [ALPM] upgraded libarchive (3.5.0-1 -> 3.5.1-1)
[2021-01-02T18:41:27+0100] [ALPM] upgraded libmicrohttpd (0.9.71-1 -> 0.9.72-1)
[2021-01-02T18:41:27+0100] [ALPM] upgraded libmpc (1.1.0-2 -> 1.2.1-1)
[2021-01-02T18:41:27+0100] [ALPM] upgraded libproxy (0.4.16-1 -> 0.4.17-1)
[2021-01-02T18:41:28+0100] [ALPM] upgraded linux (5.9.14.arch1-1 -> 5.10.3.arch1-1)
[2021-01-02T18:41:29+0100] [ALPM] upgraded linux-headers (5.9.14.arch1-1 -> 5.10.3.arch1-1)
[2021-01-02T18:41:29+0100] [ALPM] upgraded mesa (20.3.1-1 -> 20.3.2-2)
[2021-01-02T18:41:30+0100] [ALPM] upgraded openjpeg2 (2.3.1-3 -> 2.4.0-1)
[2021-01-02T18:41:30+0100] [ALPM] upgraded psmisc (23.3-2 -> 23.3-4)
[2021-01-02T18:41:30+0100] [ALPM] upgraded python-resolvelib (0.5.3-1 -> 0.5.4-1)
[2021-01-02T18:41:30+0100] [ALPM] upgraded python-urllib3 (1.25.11-1 -> 1.26.1-1)
[2021-01-02T18:41:30+0100] [ALPM] upgraded python-pip (20.2.3-1 -> 20.2.4-1)
[2021-01-02T18:41:30+0100] [ALPM] upgraded xf86-video-intel (1:2.99.917+913+g9236c582-1 -> 1:2.99.917+914+ga511f22c-1)
Anybody with similar issues ? Unfortunately I haven't seen solutions on the Mailing Lists or anything related. So it is a peculiarity of my system.
Are there any solutions or workarounds ? Thank you in advance.
Best wishes.
Last edited by vonNeumannStability (2021-01-05 08:39:15)
Offline
I managed to freeze it again and happened while streaming a video on YouTube on Firefox. which consumes about 50% of my CPU. By opening many tabs I saw a reading of 120% of CPU on htop, which is strange. Does anybody know why this has not happened before and is a problem manifested only after the 2021.01.02 upgrade ?
Offline
Potentially a bug with the intel driver and the new kernel, major kernel updates are always a good candidate for regressions. After rebooting from such a crash do you see any logs in that direction in
sudo journalctl -b-1
?
Offline
Potentially a bug with the intel driver and the new kernel, major kernel updates are always a good candidate for regressions. After rebooting from such a crash do you see any logs in that direction in
sudo journalctl -b-1
?
Thanks for you answer V1del. I do not see any relevant messages on journalctl that might be helpful. Only messages of actions I performed, but no notice of an intel driver failing or anything anomalous. Please find below a list of recent activities:
Jan 04 17:54:55 mymachine systemd-logind[300]: Lid closed.
Jan 04 18:05:34 mymachine systemd-logind[300]: Lid opened.
Jan 04 18:05:39 mymachine kernel: usb 3-1: USB disconnect, device number 2
Jan 04 18:05:39 mymachine kernel: usb 3-1.4: USB disconnect, device number 4
Jan 04 18:05:39 mymachine kernel: usb 4-1: USB disconnect, device number 2
Jan 04 18:05:54 mymachine systemd-logind[300]: Power key pressed.
Jan 04 18:05:54 mymachine systemd-logind[300]: Powering Off...
Jan 04 18:05:54 mymachine systemd-logind[300]: System is powering down.
Those were activities happening around the crash. The system crashed at around 18:05.
Would you recommend in this case that I downgrade the kernel and its headers, as well as listing it on ignorepkg so that it doesn't update following the directions of Arch Linux - Downgrading_the_kernel ?
Offline
It's definitely worth a try, however if there aren't any issues on the kernel side and the power off goes through normally, it might also be just user space, xf86-video-intel is in general a popular candidate for weird issues, maybe try removing that (and xorg.conf referencing it if present) maybe also the mesa update, though as that was a stabilisation increment I'd hope it to not introduce freezing issues (... but there's never a guarantee of course)
Offline
If you install the linux-lts kernel in addition to the linux kernel can you reproduce the issue using linux-lts?
Offline
It's definitely worth a try, however if there aren't any issues on the kernel side and the power off goes through normally, it might also be just user space, xf86-video-intel is in general a popular candidate for weird issues, maybe try removing that (and xorg.conf referencing it if present) maybe also the mesa update, though as that was a stabilisation increment I'd hope it to not introduce freezing issues (... but there's never a guarantee of course)
I am trying to downgrade linux, linux-headers, xf86-video-intel and mesa just in case. Unfortunately when I run the command on the wiki:
sudo pacman -U linux-5.9.14.arch1-1-x86_64.pkg.tar.zst linux-headers-5.9.14.arch1-1-x86_64.pkg.tar.zst mesa-20.3.1-1-x86_64.pkg.tar.zst xf86-video-intel-1:2.99.917+913+g9236c582-1
I get
loading packages...
error: 'linux-5.9.14.arch1-1-x86_64.pkg.tar.zst': could not find or read package
error: 'linux-headers-5.9.14.arch1-1-x86_64.pkg.tar.zst': could not find or read package
error: 'mesa-20.3.1-1-x86_64.pkg.tar.zst': could not find or read package
error: 'xf86-video-intel-1:2.99.917+913+g9236c582-1': could not find or read package
I only found answers telling me that the database was not refreshed. Although everything seems to be there on the cache.
Offline
you need to write out the path (... or be cd'ed into the cache directory)
Last edited by V1del (2021-01-04 18:01:46)
Offline
you need to write out the path (... or be cd'ed into the cache directory)
Perfect, I should have thought about that.
I downgraded linux, linux-headers, xf86-video-intel and mesa. Shall I leave this issue open for at least 24h ? I will try to crash it again and if I cannot reproduce the behavior I will mark this issue as solved.
Many thanks again for your time and help.
Last edited by vonNeumannStability (2021-01-04 18:13:26)
Offline
you need to write out the path (... or be cd'ed into the cache directory)
Unfortunately this happened again even after the downgrade to the previous state. What I see as pattern is that I was using firefox and waterfox at all times, which is recent. I had been previously using chromium and I never had such a crash.
Could it be something with the browser ?
Thanks.
Offline
Did you reboot between that? I'd still reccommend you try without xf86-video-intel in any case, it's somewhat notorious for weird visual bugs, and if what you posted is indeed the point of the crash then you don't have a crash but just a graphics bug (... can you switch your VT when the "crash" happens?)
Offline
Did you reboot between that? I'd still reccommend you try without xf86-video-intel in any case, it's somewhat notorious for weird visual bugs, and if what you posted is indeed the point of the crash then you don't have a crash but just a graphics bug (... can you switch your VT when the "crash" happens?)
When the crash/Graphics Bug happens I cannot do anything. My mouse just freezes and keyboard inputs do not work besides power-off.
I probably misunderstood the wiki in this part xf86-video-intel. But to run Xorg on Intel Graphics one needs this package right (xf86-video-intel) ? If I remove them, will I even be able to run X ?
Offline
No you don't need it as mentioned in the part you link to, a lot of distributions opt for not setting up xf86-video-intel on newer chipsets as it hasn't seen many significant updates in ~8 years or so. If you do that xorg will opt for a built in modesetting driver which uses 3D acceleration to do the things xf86-video-intel was used for.
Last edited by V1del (2021-01-04 21:10:43)
Offline
No you don't need it as mentioned in the part you link to, a lot of distributions opt for not setting up xf86-video-intel on newer chipsets as it hasn't seen many significant updates in ~8 years or so. If you do that xorg will opt for a built in modesetting driver which uses 3D acceleration to do the things xf86-video-intel was used for.
Thanks. I have just uninstalled xf86-video-intel and rebooted. I report again if it crashes.
Offline
V1del wrote:No you don't need it as mentioned in the part you link to, a lot of distributions opt for not setting up xf86-video-intel on newer chipsets as it hasn't seen many significant updates in ~8 years or so. If you do that xorg will opt for a built in modesetting driver which uses 3D acceleration to do the things xf86-video-intel was used for.
Thanks. I have just uninstalled xf86-video-intel and rebooted. I report again if it crashes.
I had a similar issue after the last update and now I can confirm (after a lot of hours w/ my laptop up && running) that downgrading xf86-video-intel from "2.99.917+914+ga511f22c" to "2.99.917+913+g9236c582" avoid this crash.
```
> downgrade xf86-video-intel
Available packages:
1) xf86-video-intel 1 2.99.917+858+gc37c7ee0 1 x86_64 (remote)
- 2) xf86-video-intel 1 2.99.917+859+g33ee0c3b 1 x86_64 (remote)
- 3) xf86-video-intel 1 2.99.917+860+g3a2dec17 1 x86_64 (remote)
- 4) xf86-video-intel 1 2.99.917+863+g6afed33b 1 x86_64 (remote)
- 5) xf86-video-intel 1 2.99.917+865+g60022507 1 x86_64 (remote)
- 6) xf86-video-intel 1 2.99.917+870+g6f4972d5 1 x86_64 (remote)
7) xf86-video-intel 1 2.99.917+892+gc6cb1b19 1 x86_64 (remote)
- 8) xf86-video-intel 1 2.99.917+893+gbff5eca4 1 x86_64 (remote)
- 9) xf86-video-intel 1 2.99.917+895+gcb6bff95 1 x86_64 (remote)
- 10) xf86-video-intel 1 2.99.917+897+g0867eea6 1 x86_64 (remote)
- 11) xf86-video-intel 1 2.99.917+899+gf66d3954 1 x86_64 (remote)
12) xf86-video-intel 1 2.99.917+901+gf2a54e25 1 x86_64 (remote)
13) xf86-video-intel 1 2.99.917+904+gf2853658 1 x86_64 (remote)
- 14) xf86-video-intel 1 2.99.917+906+g846b53da 1 x86_64 (remote)
- 15) xf86-video-intel 1 2.99.917+908+g7181c5a4 1 x86_64 (remote)
16) xf86-video-intel 1 2.99.917+910+g67f15b36 1 x86_64 (remote)
- 17) xf86-video-intel 1 2.99.917+912+gad5540f6 1 x86_64 (remote)
+ 18) xf86-video-intel 1 2.99.917+913+g9236c582 1 x86_64 (remote)
+ 19) xf86-video-intel 1 2.99.917+913+g9236c582 1 x86_64 (local)
- 20) xf86-video-intel 1 2.99.917+914+ga511f22c 1 x86_64 (remote)
select a package by number: 19
```
Offline
vonNeumannStability wrote:V1del wrote:No you don't need it as mentioned in the part you link to, a lot of distributions opt for not setting up xf86-video-intel on newer chipsets as it hasn't seen many significant updates in ~8 years or so. If you do that xorg will opt for a built in modesetting driver which uses 3D acceleration to do the things xf86-video-intel was used for.
Thanks. I have just uninstalled xf86-video-intel and rebooted. I report again if it crashes.
I had a similar issue after the last update and now I can confirm (after a lot of hours w/ my laptop up && running) that downgrading xf86-video-intel from "2.99.917+914+ga511f22c" to "2.99.917+913+g9236c582" avoid this crash.
```
> downgrade xf86-video-intel
Available packages:1) xf86-video-intel 1 2.99.917+858+gc37c7ee0 1 x86_64 (remote)
- 2) xf86-video-intel 1 2.99.917+859+g33ee0c3b 1 x86_64 (remote)
- 3) xf86-video-intel 1 2.99.917+860+g3a2dec17 1 x86_64 (remote)
- 4) xf86-video-intel 1 2.99.917+863+g6afed33b 1 x86_64 (remote)
- 5) xf86-video-intel 1 2.99.917+865+g60022507 1 x86_64 (remote)
- 6) xf86-video-intel 1 2.99.917+870+g6f4972d5 1 x86_64 (remote)
7) xf86-video-intel 1 2.99.917+892+gc6cb1b19 1 x86_64 (remote)
- 8) xf86-video-intel 1 2.99.917+893+gbff5eca4 1 x86_64 (remote)
- 9) xf86-video-intel 1 2.99.917+895+gcb6bff95 1 x86_64 (remote)
- 10) xf86-video-intel 1 2.99.917+897+g0867eea6 1 x86_64 (remote)
- 11) xf86-video-intel 1 2.99.917+899+gf66d3954 1 x86_64 (remote)
12) xf86-video-intel 1 2.99.917+901+gf2a54e25 1 x86_64 (remote)
13) xf86-video-intel 1 2.99.917+904+gf2853658 1 x86_64 (remote)
- 14) xf86-video-intel 1 2.99.917+906+g846b53da 1 x86_64 (remote)
- 15) xf86-video-intel 1 2.99.917+908+g7181c5a4 1 x86_64 (remote)
16) xf86-video-intel 1 2.99.917+910+g67f15b36 1 x86_64 (remote)
- 17) xf86-video-intel 1 2.99.917+912+gad5540f6 1 x86_64 (remote)
+ 18) xf86-video-intel 1 2.99.917+913+g9236c582 1 x86_64 (remote)
+ 19) xf86-video-intel 1 2.99.917+913+g9236c582 1 x86_64 (local)
- 20) xf86-video-intel 1 2.99.917+914+ga511f22c 1 x86_64 (remote)select a package by number: 19
```
Thanks for commenting. I confirm this too for users that come to this question: I had no more frozen screens since I deleted this driver and ran modesetting.
Offline