You are not logged in.

#1 2016-07-09 11:30:49

HappyClown
Member
Registered: 2015-08-15
Posts: 161

[SOLVED] Mysterious kernel panics?

I don't know which section this question belongs in, but since my new installation of ArchLinux, I've been having near constant kernel  panics. I thought this was entirely due to the intel open-source drivers, and simply used the modesetting driver instead. After another panic, I disabled hardware acceleration, as that was the particular cause of crashes on the other system. After yet another crash, after resizing lxpanel, I decided to investigate rather than guessing. At the end of the journal log, lxpanel "dumps it's core", so I can only assume it's the culprit. I will post the journalctl output when I connect using the computer. I have no idea what do do.

Jul 09 04:49:01 Bonfilia systemd-coredump[4448]: Process 4078 (lxpanel) of user 1000 dumped core.
                                                 
                                                 Stack trace of thread 4078:
                                                 #0  0x00007f08bdb99560 __strftime_internal (libc.so.6)
                                                 #1  0x00007f08bdb9b616 __strftime_l (libc.so.6)
                                                 #2  0x000000000040f7bb n/a (lxpanel)
                                                 #3  0x00007f08be435703 n/a (libglib-2.0.so.0)
                                                 #4  0x00007f08be434c8a g_main_context_dispatch (libglib-2.0.so.0)
                                                 #5  0x00007f08be435040 n/a (libglib-2.0.so.0)
                                                 #6  0x00007f08be435362 g_main_loop_run (libglib-2.0.so.0)
                                                 #7  0x00007f08bf8154e7 gtk_main (libgtk-x11-2.0.so.0)
                                                 #8  0x000000000040cb00 main (lxpanel)
                                                 #9  0x00007f08bdb0b741 __libc_start_main (libc.so.6)
                                                 #10 0x000000000040ce09 _start (lxpanel)
                                                 
                                                 Stack trace of thread 4079:
                                                 #0  0x00007f08bdbc96cd poll (libc.so.6)
                                                 #1  0x00007f08be434fd6 n/a (libglib-2.0.so.0)
                                                 #2  0x00007f08be4350ec g_main_context_iteration (libglib-2.0.so.0)
                                                 #3  0x00007f08be435131 n/a (libglib-2.0.so.0)
                                                 #4  0x00007f08be45b2b5 n/a (libglib-2.0.so.0)
                                                 #5  0x00007f08bde93484 start_thread (libpthread.so.0)
                                                 #6  0x00007f08bdbd26dd __clone (libc.so.6)

Last edited by HappyClown (2016-07-13 02:58:18)

Offline

#2 2016-07-09 12:39:08

HappyClown
Member
Registered: 2015-08-15
Posts: 161

Re: [SOLVED] Mysterious kernel panics?

Another panic has occurred, even though lxpanel isn't running but there isn't anything about it in the journal.

Offline

#3 2016-07-09 12:58:28

WorMzy
Forum Moderator
From: Scotland
Registered: 2010-06-16
Posts: 11,787
Website

Re: [SOLVED] Mysterious kernel panics?

Are you applying the microcode updates at boot?

https://wiki.archlinux.org/index.php/Microcode


Sakura:-
Mobo: MSI MAG X570S TORPEDO MAX // Processor: AMD Ryzen 9 5950X @4.9GHz // GFX: AMD Radeon RX 5700 XT // RAM: 32GB (4x 8GB) Corsair DDR4 (@ 3000MHz) // Storage: 1x 3TB HDD, 6x 1TB SSD, 2x 120GB SSD, 1x 275GB M2 SSD

Making lemonade from lemons since 2015.

Offline

#4 2016-07-09 13:07:51

HappyClown
Member
Registered: 2015-08-15
Posts: 161

Re: [SOLVED] Mysterious kernel panics?

I will admit, I have not. On this hardware, last I checked, there are no microcode updates. I'll apply any now.

As expected, no updates.

[    1.502813] microcode: CPU0 sig=0x30678, pf=0x4, revision=0x809
[    1.502898] microcode: CPU1 sig=0x30678, pf=0x4, revision=0x809
[    1.503283] microcode: CPU2 sig=0x30678, pf=0x4, revision=0x809
[    1.503364] microcode: CPU3 sig=0x30678, pf=0x4, revision=0x809
[    1.503854] microcode: Microcode Update Driver: v2.01 <tigran@aivazian.fsnet.co.uk>, Peter Oruba

Last edited by HappyClown (2016-07-09 13:12:20)

Offline

#5 2016-07-09 13:36:01

WorMzy
Forum Moderator
From: Scotland
Registered: 2010-06-16
Posts: 11,787
Website

Re: [SOLVED] Mysterious kernel panics?

Fair enough. That's just the first thing* I think to check if someone is experiencing random segfaults/panics/crashes. The next thing I would do is check the machine's RAM by running a memtest from a liveCD. If that doesn't find any problems, try the -lts kernel. If that's fine, then try previous versions of the regular kernel and see if you can identify when the problem started.


* At least, the first thing after checking the forums and bug tracker for similar issues, which I assume you did before posting.


Sakura:-
Mobo: MSI MAG X570S TORPEDO MAX // Processor: AMD Ryzen 9 5950X @4.9GHz // GFX: AMD Radeon RX 5700 XT // RAM: 32GB (4x 8GB) Corsair DDR4 (@ 3000MHz) // Storage: 1x 3TB HDD, 6x 1TB SSD, 2x 120GB SSD, 1x 275GB M2 SSD

Making lemonade from lemons since 2015.

Offline

#6 2016-07-09 13:43:30

HappyClown
Member
Registered: 2015-08-15
Posts: 161

Re: [SOLVED] Mysterious kernel panics?

Of course, I found nothing. That's how I learned of the modesetting driver in the first place. Now that I think about it, though I doubt it, it could be that I switched the driver (Intel Graphics, by the way). I've done memtesting before and I've had no issues. This is pretty new hardware and memtesting is tedious. Before this installation, I found that there were differences with the LTS kernel but I forgot what they were. I'll try booting with the other kernel and see what happens. For the sake of providing information, here's my 20-intel.conf:

Section "Device"
   Identifier  "Intel Graphics"
   Driver      "modesetting"
  Option "NoAccel" "True"
EndSection

Offline

#7 2016-07-09 14:22:14

HappyClown
Member
Registered: 2015-08-15
Posts: 161

Re: [SOLVED] Mysterious kernel panics?

It panicked again with the LTS kernel. There's nothing in the journal.

Offline

#8 2016-07-09 20:18:24

HappyClown
Member
Registered: 2015-08-15
Posts: 161

Re: [SOLVED] Mysterious kernel panics?

Two passes and no issues found.  I should have  mentioned that issues appear exclusively when running X . It hasn't crashed while in the terminal.

Offline

#9 2016-07-09 20:44:58

JohnBobSmith
Member
From: Canada
Registered: 2014-11-29
Posts: 804

Re: [SOLVED] Mysterious kernel panics?

If that is the case, then chances are very likely you are loading a bad video driver. Perhaps this bad driver only takes down the system when certain actions (re-sizing your lxpanel for example) are carried out. Thus the crashes appear to be random. I will say that It is highly unlikely that lxpanel by itself could take down the kernel. Thats why I suspect the video driver. Can you post...
-Detailed hardware information (inxi -SCG should suffice)
-latest xorg.0.log and journal log (use pastebin services if they are large)
-The last 20 lines of dmesg (dmesg | tail -n20)
-Anything that may have changed before and after the issue appeared
-Anything else that you think would help solve the problem.

Kernel panics are absolutely annoying/brutal. I would hate to see you leave Arch as a result of this issue.

Good luck!


I am diagnosed with bipolar disorder. As it turns out, what I thought was my greatest weakness is now my greatest strength.

Everyday, I make a conscious choice to overcome my challenges and my problems. It's not easy, but its better than the alternative...

Offline

#10 2016-07-09 21:22:02

HappyClown
Member
Registered: 2015-08-15
Posts: 161

Re: [SOLVED] Mysterious kernel panics?

Thanks, I'd hate to leave Arch. I had Arch running on this system before without issues. I remember having kernel panics when I first installed it. Adding the NoAccel True to the 20-intel.conf fixed it. I was running the intel driver before but now I'm using the modesetting driver.

inxi output:

[1;34mSystem:   [0;37m [1;34mHost:[0;37m Bonfilia [1;34mKernel:[0;37m 4.4.14-1-lts x86_64 (64 bit) [1;34mConsole:[0;37m tty 1 [1;34mDistro:[0;37m Arch Linux[0;37m
[1;34mCPU:      [0;37m [1;34mQuad core[0;37m Intel Pentium J2900 (-MCP-)[0;37m [1;34mcache:[0;37m 1024 KB[0;37m 
[1;34m          [0;37m [1;34mclock speeds:[0;37m [1;34mmax:[0;37m 2665 MHz [1;34m1:[0;37m 2087 MHz [1;34m2:[0;37m 1332 MHz [1;34m3:[0;37m 1891 MHz [1;34m4:[0;37m 2319 MHz[0;37m
[1;34mGraphics: [0;37m [1;34mCard:[0;37m Intel Atom Processor Z36xxx/Z37xxx Series Graphics & Display[0;37m
[1;34m          [0;37m [1;34mDisplay Server:[0;37m X.org 1.18.3 [1;34mdriver:[0;37m N/A [1;34mtty size:[0;37m 160x64 [1;34mAdvanced Data:[0;37m N/A out of X[0;37m
[0m

I'll be honest, I removed today's xorg.0.conf in an attempt to get it to regenerate it, but now I know it doesn't work that way. Here's xorg.0.conf.old (from about 15 hours ago at least):
http://pastebin.com/PEpQ65Du

The last 20 lines of dmesg are:

[   10.717249] snd_hda_codec_realtek hdaudioC0D0:      Front Mic=0x19
[   10.717251] snd_hda_codec_realtek hdaudioC0D0:      Rear Mic=0x18
[   10.717253] snd_hda_codec_realtek hdaudioC0D0:      Line=0x1a
[   10.772425] input: HDA Intel PCH Front Mic as /devices/pci0000:00/0000:00:1b.0/sound/card0/input12
[   10.772536] input: HDA Intel PCH Rear Mic as /devices/pci0000:00/0000:00:1b.0/sound/card0/input13
[   10.772636] input: HDA Intel PCH Line as /devices/pci0000:00/0000:00:1b.0/sound/card0/input14
[   10.772739] input: HDA Intel PCH Line Out as /devices/pci0000:00/0000:00:1b.0/sound/card0/input15
[   10.773418] input: HDA Intel PCH Front Headphone as /devices/pci0000:00/0000:00:1b.0/sound/card0/input16
[   10.773549] input: HDA Intel PCH HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:1b.0/sound/card0/input17
[   10.803618] Console: switching to colour frame buffer device 160x64
[   10.810258] i915 0000:00:02.0: fb0: inteldrmfb frame buffer device
[  106.954410] IPv6: ADDRCONF(NETDEV_UP): wlp1s0: link is not ready
[  112.378376] IPv6: ADDRCONF(NETDEV_UP): wlp1s0: link is not ready
[  116.022565] wlp1s0: authenticate with 6c:19:8f:f0:bb:13
[  116.036161] wlp1s0: send auth to 6c:19:8f:f0:bb:13 (try 1/3)
[  116.044837] wlp1s0: authenticated
[  116.051963] wlp1s0: associate with 6c:19:8f:f0:bb:13 (try 1/3)
[  116.059751] wlp1s0: RX AssocResp from 6c:19:8f:f0:bb:13 (capab=0x411 status=0 aid=5)
[  116.059888] wlp1s0: associated
[  116.059939] IPv6: ADDRCONF(NETDEV_CHANGE): wlp1s0: link becomes ready

Journal is actually too big for pastebin.com, download here: https://drive.google.com/open?id=0B2wj9 … FhrekNHSWc

Also, it panicked again, and says lxpanel dumped it's core, but lxpanel wasn't running (as far as I know).

Last edited by HappyClown (2016-07-09 21:35:30)

Offline

#11 2016-07-09 21:45:09

JohnBobSmith
Member
From: Canada
Registered: 2014-11-29
Posts: 804

Re: [SOLVED] Mysterious kernel panics?

The only way to eliminate lxpanel as a culprit would be to remove (paman -Rnsu) it, any files in ~/.config, and any files in your home folder relating to it. That would certainly narrow it down. However, I'm not convinced. There was this in your xorg log:

[ 13224.285] (II) LoadModule: "fbdev"
[ 13224.285] (WW) Warning, couldn't open module fbdev
[ 13224.285] (II) UnloadModule: "fbdev"
[ 13224.285] (II) Unloading fbdev
[ 13224.285] (EE) Failed to load module "fbdev" (module does not exist, 0)
[ 13224.285] (II) LoadModule: "vesa"
[ 13224.286] (WW) Warning, couldn't open module vesa
[ 13224.286] (II) UnloadModule: "vesa"
[ 13224.286] (II) Unloading vesa
[ 13224.286] (EE) Failed to load module "vesa" (module does not exist, 0)

Note the errors (EE). This is important because in your dmesg:

[   10.803618] Console: switching to colour frame buffer device 160x64
[   10.810258] i915 0000:00:02.0: fb0: inteldrmfb frame buffer device

Curious, no? You appear to be loading a different (conflicting, maybe??) frame buffer(s). I would look up anything relating to inteldrmfb and kernel panics. Also, have you changed any of the frame buffer type stuff? Could an old configuration file still be loading conflicting framebuffers?? This has a very high potential to be the cause of your kernel panic, as far as I'm aware. I'll keep looking...


I am diagnosed with bipolar disorder. As it turns out, what I thought was my greatest weakness is now my greatest strength.

Everyday, I make a conscious choice to overcome my challenges and my problems. It's not easy, but its better than the alternative...

Offline

#12 2016-07-09 21:52:43

HappyClown
Member
Registered: 2015-08-15
Posts: 161

Re: [SOLVED] Mysterious kernel panics?

Thank you so much for your help, this is a whole new revelation. Keep in mind, this is a near-fresh install, only one day old. Absolutely curious, I haven't touched anything with even the description of frame buffer since the previous install, and that was a program which loaded images from the terminal. I'll absolutely look into it.

EDIT: I took a small break from troubleshooting, as I've been doing it for hours. I'm testing out the intel drivers over the modesetting drivers to see how it works out.

Last edited by HappyClown (2016-07-09 22:33:33)

Offline

#13 2016-07-09 22:10:28

JohnBobSmith
Member
From: Canada
Registered: 2014-11-29
Posts: 804

Re: [SOLVED] Mysterious kernel panics?

You're very welcome smile keep me posted.

EDIT: Here's something. In your journal...

Jul 09 04:49:01 Bonfilia systemd-coredump[4448]: Process 4078 (lxpanel) of user 1000 dumped core.
                                                 
                                                 Stack trace of thread 4078:
                                                 #0  0x00007f08bdb99560 __strftime_internal (libc.so.6)
                                                 #1  0x00007f08bdb9b616 __strftime_l (libc.so.6)
                                                 #2  0x000000000040f7bb n/a (lxpanel)
                                                 #3  0x00007f08be435703 n/a (libglib-2.0.so.0)
                                                 #4  0x00007f08be434c8a g_main_context_dispatch (libglib-2.0.so.0)
                                                 #5  0x00007f08be435040 n/a (libglib-2.0.so.0)
                                                 #6  0x00007f08be435362 g_main_loop_run (libglib-2.0.so.0)
                                                 #7  0x00007f08bf8154e7 gtk_main (libgtk-x11-2.0.so.0)
                                                 #8  0x000000000040cb00 main (lxpanel)
                                                 #9  0x00007f08bdb0b741 __libc_start_main (libc.so.6)
                                                 #10 0x000000000040ce09 _start (lxpanel)
                                                 
                                                 Stack trace of thread 4079:
                                                 #0  0x00007f08bdbc96cd poll (libc.so.6)
                                                 #1  0x00007f08be434fd6 n/a (libglib-2.0.so.0)
                                                 #2  0x00007f08be4350ec g_main_context_iteration (libglib-2.0.so.0)
                                                 #3  0x00007f08be435131 n/a (libglib-2.0.so.0)
                                                 #4  0x00007f08be45b2b5 n/a (libglib-2.0.so.0)
                                                 #5  0x00007f08bde93484 start_thread (libpthread.so.0)
                                                 #6  0x00007f08bdbd26dd __clone (libc.so.6)
-- Reboot --

...We find a stack trace with references to many core libraries: libc, libglib-2.0, etc. I find it unlikely, but totally possible, that there is a bug in the libraries or lxpanel. However, since these crashes always seem to revolve around lxpanel in some way, I can now more firmly recommend removing lxpanel. We could then see if the issue stops. But this begs the question: How can lxpanel take down the system??? I'm at a loss there...

More info might help. Can you post the *exact* text you are presented with when you get the kernel panic? A picture would work if all else fails.

Last edited by JohnBobSmith (2016-07-09 22:24:01)


I am diagnosed with bipolar disorder. As it turns out, what I thought was my greatest weakness is now my greatest strength.

Everyday, I make a conscious choice to overcome my challenges and my problems. It's not easy, but its better than the alternative...

Offline

#14 2016-07-09 22:42:28

HappyClown
Member
Registered: 2015-08-15
Posts: 161

Re: [SOLVED] Mysterious kernel panics?

It's odd, I know. I'm testing out what happens if I switch video drivers (It crashed while I was typing this). I took a small break from troubleshooting as I've been it for hours. I have no idea how lxpanel can take down the system (NOTE: Uninstalling lxpanel did nothing). I don't know it's a kernel panic from the dialog, I know it's a panic from the symptoms. No dialogs are displayed.

EDIT: Are you certain that two framebuffer devices are conflicting each other and that the two mentioned devices aren't the same one? Also, which frame buffer is expected to run? Vesa and fbdev are a part of two other video drivers, but, on this page: https://bbs.archlinux.org/viewtopic.php?id=70063, "WHY do you want to use fbdev as your video driver?! It's really REALLY slow! Use vesa if you don't want any other driver, other than that just install and use a driver for your specific card" (Note: I'm currently using the intel driver)

EDIT2: It seems as though X is scanning for drivers, though I specified the driver in 20-intel.conf. It's generally safe to ignore the errors, but they shouldn't appear if the drivers are specified. (They don't appear anymore.)

EDIT3: It seems as though the crashes only occur when the machine is active, which means a particular event is triggering it.

Last edited by HappyClown (2016-07-10 00:28:10)

Offline

#15 2016-07-10 01:54:59

HappyClown
Member
Registered: 2015-08-15
Posts: 161

Re: [SOLVED] Mysterious kernel panics?

So I found something pretty interesting at the end of my journal after yet another panic:

Jul 09 19:35:39 Bonfilia pulseaudio[666]: E: [alsa-sink-USB Audio] alsa-sink.c: ALSA woke us up to write new data to the device, but there was actually nothing to write.
Jul 09 19:35:39 Bonfilia pulseaudio[666]: E: [alsa-sink-USB Audio] alsa-sink.c: Most likely this is a bug in the ALSA driver 'snd_usb_audio'. Please report this issue to the ALSA developers.
Jul 09 19:35:39 Bonfilia pulseaudio[666]: E: [alsa-sink-USB Audio] alsa-sink.c: We were woken up with POLLOUT set -- however a subsequent snd_pcm_avail() returned 0 or another value < min_avail.

Offline

#16 2016-07-11 12:23:47

JohnBobSmith
Member
From: Canada
Registered: 2014-11-29
Posts: 804

Re: [SOLVED] Mysterious kernel panics?

I'm beginning to feel somewhat lost with this, sad. What are the contents of /etc/modrope.d/modprobe.conf? are there any other conf's in that directory? I'm wondering if you're loading/blacklisting something you shouldn't be.

EDIT: Re - frame buffers: No, I am not certain that the frame buffers are an issue. Some of your output's regarding frame buffers just struck me as being a bit odd, and possibly worth investigating.

Hopefully we can find a solution soon. smile

Last edited by JohnBobSmith (2016-07-11 12:26:27)


I am diagnosed with bipolar disorder. As it turns out, what I thought was my greatest weakness is now my greatest strength.

Everyday, I make a conscious choice to overcome my challenges and my problems. It's not easy, but its better than the alternative...

Offline

#17 2016-07-12 02:16:08

HappyClown
Member
Registered: 2015-08-15
Posts: 161

Re: [SOLVED] Mysterious kernel panics?

Thanks for your patience, I'm as lost as you are. The modprobe.d directory is actually empty. I suspect that the issue is with my graphics hardware, as it's an Intel product which is only about a year old. I'm reading about issues that other people have had, and I'm thinking that maybe an alternative driver is needed, other than intel and modesetting, but I have gotten this working before using the intel driver, so this is confusing.

Aha! I may have found the solution, and it seems it was my 20-intel.conf after all. I'm not sure if I found it, I'll let you guys know if it works. See: https://bugs.launchpad.net/ubuntu/+sour … mments/147

Last edited by HappyClown (2016-07-12 02:35:43)

Offline

#18 2016-07-13 02:57:31

HappyClown
Member
Registered: 2015-08-15
Posts: 161

Re: [SOLVED] Mysterious kernel panics?

It seems silly, and I don't remember adding that line to the conf the first time, but it works. I could've sworn I didn't add it in the last installation on this hardware. Sorry for wasting time and thanks for helping out. The system isn't crashing anymore (so far).

Offline

#19 2016-07-13 06:20:54

JohnBobSmith
Member
From: Canada
Registered: 2014-11-29
Posts: 804

Re: [SOLVED] Mysterious kernel panics?

You're welcome and I hope you continue to have a stable system! big_smile


I am diagnosed with bipolar disorder. As it turns out, what I thought was my greatest weakness is now my greatest strength.

Everyday, I make a conscious choice to overcome my challenges and my problems. It's not easy, but its better than the alternative...

Offline

#20 2016-07-14 01:25:03

HappyClown
Member
Registered: 2015-08-15
Posts: 161

Re: [SOLVED] Mysterious kernel panics?

Thanks, you're awesome!

Offline

Board footer

Powered by FluxBB