Arch Linux

GOLDER

Member

Registered: 2018-11-07

Posts: 2

[SOLVED] Random freezes every ~15min

Hello, this is my first post on this forum, because there is an issue I can't find the answer anywhere.
So basically, I have 2 different setups for my computer, one week its connected to a TV and ethernet, and the other week it's connected to a standard monitor and wifi or usb ether (from iphone)
And on the first weeks I never get any issues, but on the second week, I get freezes on my system every 10-15 minutes or so, and this issue has been consistent on several kernels, and even several OS's (had this problem on Manjaro too)
During the freeze, I can't do anything, no mouse, keyboard (can't access other TTY) and any sound that was playing before the freeze just repeats forever until I press reset on my case.
This is my setup:
SW:
Latest arch
Gnome 3
Zen kernel (also had issue on 4.14, 4.18)
Latest nvidia driver (4.10.66)

HW:
i5 6400 (latest microcode)
8gb ddr4 (memtested)
sandisk ssd (no active swap partitions in case that was the problem)
gtx 750 ti
and a pci wifi card
monitor: samsung syncmaster display (potential issue ?)

and this is the error I get every time I get this crash (/var/log/Xorg.0.log.old):

[   861.922] (--) NVIDIA(GPU-0): Samsung SyncMaster (CRT-0): connected
[   861.922] (--) NVIDIA(GPU-0): Samsung SyncMaster (CRT-0): 400.0 MHz maximum pixel clock
[   861.922] (--) NVIDIA(GPU-0): 
[   861.922] (--) NVIDIA(GPU-0): DFP-0: disconnected
[   861.922] (--) NVIDIA(GPU-0): DFP-0: Internal TMDS
[   861.922] (--) NVIDIA(GPU-0): DFP-0: 330.0 MHz maximum pixel clock
[   861.922] (--) NVIDIA(GPU-0): 
[   861.922] (--) NVIDIA(GPU-0): DFP-1: disconnected
[   861.922] (--) NVIDIA(GPU-0): DFP-1: Internal TMDS
[   861.922] (--) NVIDIA(GPU-0): DFP-1: 165.0 MHz maximum pixel clock
[   861.922] (--) NVIDIA(GPU-0): 
[   909.049] (**) Option "fd" "32"
[   909.049] (II) event2  - Power Button: device removed
[   909.050] (**) Option "fd" "35"
[   909.050] (II) event1  - Power Button: device removed
[   909.050] (**) Option "fd" "36"
[   909.050] (II) event0  - Sleep Button: device removed
[   909.051] (**) Option "fd" "37"
[   909.051] (II) event3  - SteelSeries Rival Gaming Mouse: device removed
[   909.051] (**) Option "fd" "38"
[   909.052] (II) event4  - SteelSeries Rival Gaming Mouse: device removed
[   909.052] (**) Option "fd" "39"
[   909.052] (**) Option "fd" "40"
[   909.053] (II) event6  - SONiX USB DEVICE: device removed
[   909.053] (**) Option "fd" "41"
[   909.053] (II) event7  - SONiX USB DEVICE Keyboard: device removed
[   909.053] (**) Option "fd" "42"
[   909.054] (II) event8  - SONiX USB DEVICE System Control: device removed
[   909.054] (**) Option "fd" "43"
[   909.054] (**) Option "fd" "44"
[   909.054] (II) event10 - C-Media USB Headphone Set  : device removed
[   909.055] (**) Option "fd" "45"
[   909.055] (II) event12 - Eee PC WMI hotkeys: device removed
[   909.055] (**) Option "fd" "39"
[   909.055] (II) event5  - SteelSeries Rival Gaming Mouse: device removed
[   909.055] (**) Option "fd" "43"
[   909.055] (II) event9  - SONiX USB DEVICE Consumer Control: device removed
[   909.122] (II) systemd-logind: got pause for 13:74
[   909.122] (II) systemd-logind: got pause for 13:65
[   909.122] (II) systemd-logind: got pause for 13:69
[   909.122] (II) systemd-logind: got pause for 13:64
[   909.122] (II) systemd-logind: got pause for 13:72
[   909.122] (II) systemd-logind: got pause for 13:66
[   909.122] (II) systemd-logind: got pause for 13:71
[   909.122] (II) systemd-logind: got pause for 13:68
[   909.122] (II) systemd-logind: got pause for 13:76
[   909.122] (II) systemd-logind: got pause for 13:67
[   909.122] (II) systemd-logind: got pause for 13:73
[   909.122] (II) systemd-logind: got pause for 13:70

Any help would be appreciated ! (I can provide more info if need be!)

Last edited by GOLDER (2018-11-08 21:24:08)

twelveeighty

Member

Registered: 2011-09-04

Posts: 1,456

Re: [SOLVED] Random freezes every ~15min

There a quite a few blogs and bug reports when you Google "systemd-logind: got pause for", for example https://chrisdown.name/2016/10/18/x-fre … pause.html. Have you tried eliminating either the monitor, wifi or USB tether as a factor? Start with the working setup with the TV/ethernet and then introduce the monitor, wifi and USB tether one at a time to see when you start seeing the freezes?

theburritoeater

Member

Registered: 2018-03-21

Posts: 9

Re: [SOLVED] Random freezes every ~15min

It seems that the first events that are triggering these freezes are NVIDIA problems, which then start closing "fd" aka file descriptors for several devices connected to your computer.  Any device that is connected to your computer needs a driver, and all drivers in linux are simply files, which get written to by file descriptors.  So what is most likely happening is the NVIDIA driver is crashing causing an edge level interrupt, which can cause several other drivers (thus linked to "fd"s) to be closed, which can result in a full system hang.  When it comes to NVIDIA and laptops there are a slew of problems.  Some approaches I suggest trying out are:
  - make sure the correct version of the NVIDIA driver installed.  Perhaps the latest version is not suitable for you 9xx series card
  - make sure you have the correct software installed for switching your graphics on and off, perhaps try out bumblebee and intel-virtual-output if you haven't
  - make sure you have a kernel that supports the correct types of interrupts, check out this issue report if you want to learn more about kernal interrupt type errors.
  - if all else fails, make sure you do not have conflicting video drivers working against each other to cause this interrupt issue.  When I first tried to get my 1060Ti working on my laptop I unknowingly installed 2 different graphics drivers on my machine and this caused several issues.

GOLDER

Member

Registered: 2018-11-07

Posts: 2

Re: [SOLVED] Random freezes every ~15min

Thanks guys, I actually updated my packages via pacman (also did it a few days ago but nothing happened) and everything seems to be stable now, I will investigate though