full system hang, not responsive

Devotedfollower · 2011-01-22 22:44:08

Hello all,

on my LG p510 laptop I have regular hangs where the system goes into a full freeze. I can't exit kde or access the laptop via ssh after this happens. It used to happen a lot with the old kernel and hadn't happened at all with the newest kernel update...this all changed in the last couple of days when it has started up again. I used to think that it had something to do with the intel wifi card I have in this system since I'd always see wifi activity after the full freeze would happen but in the last two instances I'd be using a local app and there wouldn't be a blinking wifi activity.

Please let me know what I can do so that this machine can be more stable than my XP machine....

thanks,
DF

lagagnon · 2011-01-23 00:45:56

IMHO I find that full system freezes are infrequently related to the operating system/software but rather to faulty hardware. I suggest you run a Live Linux CD on that laptop (or from a USB stick) and stress the machine with the LiveCD. If you still get freezes then your problem is undoubtedly hardware related (overheating, bad RAM, faluty hard drive, faulty power supply).

R00KIE · 2011-01-23 01:08:15

From my experience, if X crashes hard then it takes the system down with it, as in not even sysreq working and having to shut down the system by force.

Might be worth to check if the graphics drivers have been updated lately and if you can find the app that triggers the crash (should be quite obvious after you get a few crashes in a row ).

Devotedfollower · 2011-02-17 12:34:57

I had never heard the sysreq thing so I've enabled it after every reboot. Please tell me what key combo ctrl + alt + prints + ? to hit to find out useful information to trouble shoot this bug. It frequently but not always happens during youtube videos (flash) and when it freezes the sound will loop the last second or so.

I do system updates pretty much everyday so I don't think the NVIDIA drivers are the issue.

I will do some stress testing, but need to backup my HD first.

As I said XP had been working fine, but perhaps there has been problems that have developed with the hardware since then.

Thanks again for all your guys help.
DF

Devotedfollower · 2011-02-19 12:47:57

ok, so I've followed the https://wiki.archlinux.org/index.php/Stress_Test and each test has come back as fine. I've also edit the /etc/sysctl.conf file so that I can use sysrq keys.

I think that Rookie may be right in it being the graphics driver as most freezes happen when an application an action has just executed (starting, ending an application or minimizing/maximizing). Yesterday the system froze just as I was about to reboot the system.

What should I be hitting to get a good error log that will help resolve this problem?

Thanks,
DF

Last edited by Devotedfollower (2011-02-19 12:49:14)

Devotedfollower · 2011-02-19 13:11:45

ok, just crashed again and this time I tried the sysrq keys and they don't work... how do I check if this is a graphics or harddrive issue? Any other ideas on what could be broken with hardware if a full crash 'isn't' possible in linux these days?

Awebb · 2011-02-19 14:52:47

Sysreq: Ctrl+Alt+Print Raising Elephants Is So Utterly Boring

Devotedfollower · 2011-02-20 00:17:00

???

ssri · 2011-02-20 02:53:44

Devotedfollower wrote:

???

archwiki is your friend: https://wiki.archlinux.org/index.php/Li … _Shortcuts

Devotedfollower · 2011-02-20 22:28:45

ahh thanks for that, but the more complete list for Sysreq combos is here: http://en.wikipedia.org/wiki/Magic_SysRq_key

So linking to the wiki is that what Raising Elephants Is So Utterly Boring means?

In any case, my last full post says 'ok, just crashed again and this time I tried the sysrq keys and they don't work...' and to further add to this: If I hit the key combo for rebooting or shutdown now (normal system state), it works. If I do it when the computer freezes, it doesn't.

Where do I go from here? Please read post #6.

Thanks,
DF

Last edited by Devotedfollower (2011-02-20 22:30:49)

dedanna1029 · 2011-02-21 01:04:30

Tried the Ctrl+Alt+Backspace trick? You'd have some editing to do in xorg.conf to enable this.

Devotedfollower · 2011-02-21 12:03:22

I will try this dedanna, in the meantime though I am trying to get http://aur.archlinux.org/packages.php?ID=35901 working. I want to do a video stresstest as I think this is a video problem. Last night while watching a video, my screen started to display boxes of colours (box being 5x5 pixels or so). As I moved the mouse the boxes would change colour...weird things happening...
After a reboot everything worked again just fine.

and the error I'm having with unigine heaven is:
-> Compressing package...
bsdtar: Write error
==> ERROR: Failed to create package file.
==> ERROR: Makepkg was unable to build unigine-heaven.
==> Restart building unigine-heaven ? [y/N]
==> ---------------------------------------
==>

edit://why would the ctrl + alt + backspace (for x.org) work, when the kernel level commands don't work? Shouldn't kernel level stuff still work if x.org can receive commands?

Last edited by Devotedfollower (2011-02-21 12:06:02)

combuster · 2011-02-21 12:18:52

Could you look into logs when you reboot after the crash, /var/log/errors.log (everything.log, kernel.log, Xorg.0.log.old upon reboot) could contain valuable information on system freezing. If you see something like:

[   12.237202] ------------[ cut here ]------------
[   12.237208] WARNING: at kernel/printk.c:288 do_syslog+0x88/0x444()
[   12.237210] Hardware name: H67MA-UD2H
[   12.237211] Attempt to access syslog with CAP_SYS_ADMIN but no CAP_SYSLOG (deprecated).
[   12.237213] Modules linked in: fuse joydev snd_hda_codec_realtek snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device snd_pcm_oss snd_mixer_oss snd_hda_intel sg snd_hda_codec snd_hwdep snd_pcm snd_timer iTCO_wdt snd r8169 coretemp iTCO_vendor_support xhci_hcd cpufreq_ondemand soundcore snd_page_alloc mii ehci_hcd evdev cpufreq_powersave ext4 mbcache jbd2 crc16 sd_mod sr_mod ahci libahci libata scsi_mod
[   12.237237] Pid: 995, comm: syslog-ng Not tainted 2.6.38-rc5-RC #1
[   12.237239] Call Trace:
[   12.237243]  [<ffffffff81041c34>] ? warn_slowpath_common+0x80/0x98
[   12.237246]  [<ffffffff81119098>] ? kmsg_release+0x0/0x18
[   12.237249]  [<ffffffff81041ce0>] ? warn_slowpath_fmt+0x41/0x43
[   12.237252]  [<ffffffff81134c79>] ? security_capable+0x24/0x26
[   12.237254]  [<ffffffff810421b6>] ? do_syslog+0x88/0x444
[   12.237257]  [<ffffffff8103b8ff>] ? get_parent_ip+0x11/0x42
[   12.237259]  [<ffffffff8103b8ff>] ? get_parent_ip+0x11/0x42
[   12.237261]  [<ffffffff81119098>] ? kmsg_release+0x0/0x18
[   12.237264]  [<ffffffff811190c7>] ? kmsg_open+0x17/0x19
[   12.237266]  [<ffffffff811102fb>] ? proc_reg_open+0xa3/0x144
[   12.237268]  [<ffffffff811190b0>] ? kmsg_open+0x0/0x19
[   12.237271]  [<ffffffff81110258>] ? proc_reg_open+0x0/0x144
[   12.237273]  [<ffffffff810cd4b8>] ? __dentry_open+0x16b/0x28e
[   12.237277]  [<ffffffff81312f36>] ? _raw_spin_unlock+0x28/0x33
[   12.237280]  [<ffffffff810d6d65>] ? dget+0x26/0x2e
[   12.237282]  [<ffffffff810ce227>] ? nameidata_to_filp+0x5b/0x62
[   12.237285]  [<ffffffff810d9f49>] ? finish_open+0x9c/0x15a
[   12.237287]  [<ffffffff810d92ae>] ? do_path_lookup+0xed/0x118
[   12.237289]  [<ffffffff810da485>] ? do_filp_open+0x199/0x6b8
[   12.237291]  [<ffffffff8103b8ff>] ? get_parent_ip+0x11/0x42
[   12.237294]  [<ffffffff810d22d7>] ? cp_new_stat+0xe9/0x101
[   12.237296]  [<ffffffff8103b8ff>] ? get_parent_ip+0x11/0x42
[   12.237298]  [<ffffffff8103b8ff>] ? get_parent_ip+0x11/0x42
[   12.237301]  [<ffffffff81315a71>] ? sub_preempt_count+0x92/0xa6
[   12.237303]  [<ffffffff81312f36>] ? _raw_spin_unlock+0x28/0x33
[   12.237306]  [<ffffffff810e46e5>] ? alloc_fd+0x106/0x118
[   12.237309]  [<ffffffff8116443d>] ? strncpy_from_user+0x34/0x37
[   12.237312]  [<ffffffff810ce289>] ? do_sys_open+0x5b/0xed
[   12.237314]  [<ffffffff810ce336>] ? sys_open+0x1b/0x1d
[   12.237317]  [<ffffffff8100297b>] ? system_call_fastpath+0x16/0x1b
[   12.237319] ---[ end trace 154d92f19b14dca3 ]---
[   12.853406] r8169 0000:03:00.0: eth0: link down
[   12.853439] r8169 0000:03:00.0: eth0: link down
[   14.009731] hub 2-0:1.0: hub_suspend
[   14.009742] usb usb2: bus auto-suspend
[   14.009746] ehci_hcd 0000:00:1d.0: suspend root hub
[   14.669706] r8169 0000:03:00.0: eth0: link up
[  350.079863] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[  350.079874] [drm:kick_ring] *ERROR* Kicking stuck semaphore on blt ring
[  738.012459] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[  738.012950] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -11 (awaiting 917041 at 917034, next 917042)
[  738.013210] [drm:init_ring_common] *ERROR* render ring initialization failed ctl 00000000 head 00000000 tail 00000000 start 00000000
[  744.154389] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[  744.154400] [drm:kick_ring] *ERROR* Kicking stuck semaphore on blt ring
[  745.654954] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[  745.654965] [drm:kick_ring] *ERROR* Kicking stuck semaphore on blt ring
[  747.155609] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[  747.155664] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -11 (awaiting 917059 at 917034, next 917060)

you could paste it here. I gave those errors as an example on what to look for - not the exact same thing you could find.

Devotedfollower · 2011-02-21 15:23:24

looked through the logs and haven't found anything similar or alarming

if the system is fully unresponsive and I'm assuming harddisk access also stops...I will wait for another crash and post the logs here.

I've also updated to testing repo in the meantime to see if that helps matters as this started happening a couple of months ago and I just dealt with it (but wasn't happening during the summer of 2010)

Devotedfollower · 2011-02-22 08:36:31

ok, I have the logs copied, is there a way to attach them within this thread?

Devotedfollower · 2011-02-22 13:31:30

ok...since it doesn't look like I can attach things in the forum, I've just upload them to my vps.

http://dumpod.org/arch/Xorg.0.log.old
http://dumpod.org/arch/error.log
http://dumpod.org/arch/everything.log
http://dumpod.org/arch/kernel.log

I've also since installed nvidia-beta drivers as I feel as though there really isn't anything else to try.

lagagnon · 2011-02-22 16:15:28

I don't really know if these are the root causes but I just went through your everything log searching for "error". The 3 problems you have in that regard are: bluetooth (bluez), pulseaudio and KDM is missing a picture for your login. Try turning off pulseaudio and bluetooth and see if that fixes anything.

Devotedfollower · 2011-02-26 10:27:05

still haven't touch any of the bluetooth, pulseaudio or KDM issues, but I fixed the problem.

NVIDIA's drivers are to blame for the 130M gefroce go. I updated to the beta drivers and since then there has been no more crashes! hope this helps someone!

Last edited by Devotedfollower (2011-02-26 10:27:29)

Arch Linux

#1 2011-01-22 22:44:08

full system hang, not responsive

#2 2011-01-23 00:45:56

Re: full system hang, not responsive

#3 2011-01-23 01:08:15

Re: full system hang, not responsive

#4 2011-02-17 12:34:57

Re: full system hang, not responsive

#5 2011-02-19 12:47:57

Re: full system hang, not responsive

#6 2011-02-19 13:11:45

Re: full system hang, not responsive

#7 2011-02-19 14:52:47

Re: full system hang, not responsive

#8 2011-02-20 00:17:00

Re: full system hang, not responsive

#9 2011-02-20 02:53:44

Re: full system hang, not responsive

#10 2011-02-20 22:28:45

Re: full system hang, not responsive

#11 2011-02-21 01:04:30

Re: full system hang, not responsive

#12 2011-02-21 12:03:22

Re: full system hang, not responsive

#13 2011-02-21 12:18:52

Re: full system hang, not responsive

#14 2011-02-21 15:23:24

Re: full system hang, not responsive

#15 2011-02-22 08:36:31

Re: full system hang, not responsive

#16 2011-02-22 13:31:30

Re: full system hang, not responsive

#17 2011-02-22 16:15:28

Re: full system hang, not responsive

#18 2011-02-26 10:27:05

Re: full system hang, not responsive

Board footer