Page MenuHomeSolus

Occasional hanging / freezing
Closed, InvalidPublic

Description

I have heard and witnessed this occur on multiple PC's running Solus. I've personally witnessed it happen upon waking from suspend or sleep or even from shutdown, where any input would not be registered for at least a minute or so.

So far I've only noticed this problem on Nvidia GPU's [surprise!], and I have also found / been given some threads online about previous driver versions causing this issue (of course on a different distro and different hardware).

Some more general information:

  • My PC: i7 6700k, GTX 1070, Solus installed on an SSD, current latest stable kernel 4.12.8-13 (x86_64), driver version 384.59 I don't experience this issue commonly, but have noticed it occurring more often recently
  • Other PC: i7 5500, GTX 1060, SSD, linux-current, latest driver (all as above).

I personally have seen it more upon first boot, my other PC has mostly seen it from waking from suspend or sleep.

Both have taken the upgrade path to Solus 3, instead of installing it new / fresh from the new ISO.

Event Timeline

The person named "iranen" in IRC mentioned that he's having a similar thing happen on boot; where it'll occasionally freeze (apparently "on systemd-fsck"). His PC is the Asus X59GL with an Intel Core 2 Duo and a GeForce 8200M G. So again, an Intel CPU and Nvidia GPU.

Thought I'd mention this is still occurring now. It seems to be more likely to occur after a decent amount of uptime. I wouldn't say it is all of Solus freezing, but the mouse and keyboard stop working for either a good amount of time or indefinitely (until reboot). Not always both mouse and keyboard though, usually just one or the other and rarely both.

So after many updates, this issue is less common, but still appearing. I'm beginning to think it's something to do with the keyboard I have plugged in. It's a Corsair Strafe RGB mechanical keyboard.
My reasoning is that the mouse doesn't always freeze, but the keyboard always does.

Just to add to this on the chance it helps any. I have a basic steelseries mechanical keyboard, MX performance mouse on a 2011 I7 gigabyte UD7 board ... old/middle range nvidia graphics card - NVIDIA GF116 [GeForce GTS 450 Rev. 2] (the machine is for code not play)

Last time things were stable was kernel 4.12.5 since then chaos. Frequent freezes, leaving machine on overnight to process stuff ALWAYS freezes it up.

I changed to dedicated nvidia glx driver etc, which might have made things worse :P - hard to tell, a lot of other changes near the same point.

Just now have turned lock screen off and added acpu=off to kernel boot to see if can further rule stuff out.

I previously flipped to the LTS kernel, had no effect so am back on current (atm 4.13.16-35.current).

Pondering the possible update of the graphics card - could be time :)

Cheers

Followup - on the latest update for kernel system went plop (would not boot) -- so I took opportunity to reinstall solus boot on a newly formatted drive and all works now - not freezes nada (and yes, I have installed everything I had prior).

My guess, something along the way went off road and never came back. I also removed another boot (mint) I wasn't using anymore, AND shifted the partition scheme for the drive solus is on to gpt (it had an mbr msdos idiocy from a windows installed many many years back).

the clt-boot-manager has a much better run now and everything else is back to the solus I love :)

PS: separating home to own partition as soon as possible -- worth it every time.

I am also having this issue with Intel Video
glxinfo | grep OpenGL
OpenGL vendor string: Intel Open Source Technology Center
OpenGL renderer string: Mesa DRI Intel(R) HD Graphics 620 (Kaby Lake GT2)
OpenGL core profile version string: 4.5 (Core Profile) Mesa 17.3.1
OpenGL core profile shading language version string: 4.50
OpenGL core profile context flags: (none)
OpenGL core profile profile mask: core profile
OpenGL core profile extensions:
OpenGL version string: 3.0 Mesa 17.3.1
OpenGL shading language version string: 1.30
OpenGL context flags: (none)
OpenGL extensions:
OpenGL ES profile version string: OpenGL ES 3.2 Mesa 17.3.1
OpenGL ES profile shading language version string: OpenGL ES GLSL ES 3.20
OpenGL ES profile extensions:

I have also just experienced a freeze and I am using a regular Intel graphics chip on an X220. Not sure if it is of any help but the only programme I had running was Wire and I had to force shutdown with the power button.

Popped back to say - whatever whoever did, they done did it well. I've not seen a freeze for a few weeks now.

Simply 'the' singular Linux OS :)

I am not having that success at all. Daily basis system will not wake and have to do a forced power down.

maybe it is worth adding this is a touch screen as well and it continues to not respond from a suspend state, the computer is heard coming on but there is no screen response.

DataDrake triaged this task as Needs More Info priority.Aug 26 2018, 7:28 PM
DataDrake added a subscriber: DataDrake.

Is this still an issue on the new kernel (4.18.5)?

Closing due to lack of response in over 30 days.

Update on the issue, from a new angle.
I have apparently had this issue, though almost never, over the the past few months, on my desktop (only once so far), an Intel DH55H, with an i3-540 , my laptop (about 15 minutes ago was my 5th time so far), the Lenovo IdeaPad 300-15ISK, with an i5-6200U, and my sister's Chromebook (Twice so far), the Acer CB3-131, with a Celeron N2840.

Neither laptop had a dGPU, and the desktop had only experienced them when using the iGPU, though I had not used a dGPU long enough to verify anything.

The one thing that was common amongst the cases, was that a video was played, either in VLC, or on YouTube (regardless of the browser).
I do not have logs, but that is because I cannot seem to find the logs, and using journalctl gave me absolutely no logs from the timeframe I was interested in.

Well, I had forgotten to post this, but it had happened again on thursday, both on my, and my sister's, laptops.

Mine was after a few hours of running firefox (no YouTube this time around, I was reading articles), gimp, vlc (for music, I don't always use Lolllypop), and gedit at the same time, so I cannot point out what thread could have caused the hang.
While on my sister's laptop, it was a few minutes after boot, while running gnome-mpv (this was the first time a hang was with gnome-mpv rather than with vlc).
The kernel on my laptop was 4.18.16, and while I do not remember the version on my sister's laptop, I'm guessing it to also be 4.18.16, since it was the latest version at the time (albeit, on the stable branch, not unstable like my laptop).

Update: I was mistaken as to the kernel version on my siter's laptop, as I had not had the time to update the system there, so it was 4.18.15.
Now it had happened again on her laptop, when I was about to test the touchpad gesture latency there, after updating the system, so now it was on kernel 4.18.16, and this time the OS hung a few seconds after logging in, when trying to open the app (start for Windows users) menu.

I am starting to have suspicions that we may actually have 2 separate issues at hand here, freezes on NVidia hardware, which may or may not be due to their drivers, and which may have already been resolved, and freezes on Intel hardware, which might not be due to graphics after all.

Since the only Intel hardware here, other than my desktop, which had an Intel wireless card by means of an M.2 to PCIe adapter, are laptops, I have a suspicion that the Intel hangs may actually be due to either iwlwifi or it's accompanying firmwares, and as such may be connected to T7182.

If so, and I'll attempt to check this (I have Intel, Realtek, and Qualcomm M.2 and mPCIe cards, as well as both Qualcomm and Ralink based PCI cards (I had given my Broadcom based PCIe card to my grandmother) (the Qualcomm cards are failing to work on my desktop for some odd reason, though, only on my backup desktop do they work)) , even though it is extremely inconsistent, and if it is thus, I'll post a new task about possible kernel lockups with Intel Wireless products.

Update: Now that I had gotten another Fenvi FV-102 M.2 to PCIe WiFi/BT adapter, I can test this alongside (preGCN) AMD graphics,

Update 2: I had not had stability issues with AMD+Ralink.
Nothing to report yet on other, either untested or currently being tested, combinations.

Update 3: I have not had any stability issues (barring networking) for the past few weeks, can someone add their experiences so that we may or may not be able to close this issue?

GPU+WiFi Combinations tested:
Intel+Intel
Intel+Qualcomm
Intel+Realtek
AMD+Intel
AMD+Qualcomm
AMD+Realtek

At least this appears to rule out WiFi being connected, however we need to see whether this is still the situation for NVidia (I had given my 2nd NVidia card (a GT-9400) to my grandmother, and my friend's PC still needs a new hard drive (at least an external one, they don't have enough space for all their pictures) and a system reinstallation (currently they are running 32-bit on 64-bit capable hardware, we had already upgraded their RAM to 4GB), before I can receive my 1st one back (a GT-8200).