LinuxQuestions.org - How to diagnose a system freeze

- Linux - Newbie (https://www.linuxquestions.org/questions/linux-newbie-8/)

- - How to diagnose a system freeze (https://www.linuxquestions.org/questions/linux-newbie-8/how-to-diagnose-a-system-freeze-4175432218/)

How to diagnose a system freeze

Hello, i am running Opensuse 12.2
I have had a couple system freezes where i can't ctrl alt f2, or ctrl alt delete or do anything. Sometimes i've had clementine playing music and it plays one second of part of the song over and over.
I have to hold the power off button and reboot.
Although it does this rarely, it really bugs me when computers do this and i'd like to find out what is causing it so i can help remove this issue.

What should i do to find out what's causing this?

Thankyou

What do the logs say ? /var/log/syslog & kern.log should be time and date stamped.

If nothing, boot up memtest86 to check the RAM sticks.

-Tristan

What are the sytem specs?
Machine model #,ram, processor, etc.If running older hardware you may need to go with a light-weight distro.

Just some alternatives to the power button.

<alt><sysreq>reisub can do a more gentle reboot

<alt><sysreq>k will only restart the graphical environment; this used to be <ctrl><alt><backspace> and that might still work on your system.

Note that <sysreq> often shares its key with another functionality like <printscreen>; you need to use <ctrl> in that case; so the combination becomes <alt><ctrl><sysreq>reisub
And on laptops, it might involve the <Fn> key.

references:
http://kember.net/articles/reisub-th...linux-restart/
http://en.wikipedia.org/wiki/Magic_SysRq_key

Quote:

Originally Posted by ceyx (Post 4805775)

What do the logs say ? /var/log/syslog & kern.log should be time and date stamped.

Thanks, but I have just checked this, i do not have them files. Here are the files in /var/log

Code:

acpid            kdm.log              messages-20121007.xz  vbox-install.log

alternatives.log  krb5                  messages-20121014.xz  warn

boot.log          lastlog              NetworkManager        wpa_supplicant.log

btmp              localmessages        news                  wtmp

ConsoleKit        mail                  ntp                  Xorg.0.log

cups              mail.err              pk_backend_zypp      Xorg.0.log.old

faillog          mail.info            pm-powersave.log      YaST2

firewall          mail.warn            pm-suspend.log        zypp

hp                messages              samba                zypper.log

journal          messages-20120926.xz  snapper.log          zypper.log-20121014.xz

Quote:

Originally Posted by trist007 (Post 4805779)

If nothing, boot up memtest86 to check the RAM sticks.

-Tristan

Thanks, I checked this not to long ago, there were no errors.

Quote:

Originally Posted by EDDY1 (Post 4805797)

What are the sytem specs?
Machine model #,ram, processor, etc.If running older hardware you may need to go with a light-weight distro.

It is not an old computer. It is a:
Lenovo Thinkpad X220
6GB ram, 2 dim. (2gb) and (4gb)
Intel® Core™ i7-2620M (2.7GHz, 4MB L3 cache)
Intel HD Graphics 3000
Modified 9.5mm, intel 320 series 600gb SSD (modified to fit 7mm hard drive bay)

Quote:

Originally Posted by Wim Sturkenboom (Post 4805825)

Thanks for that, i wasn't aware of that key combo.

Hey guys i'm still having this issue. Recently i installed Slackware 14, and i'm encountering the exact same freeze problems in Slackware, too, so i'm starting to suspect it's a hardware problem.
It's just occured while i was running Slack, even that reisub key combo is paralyzed, and i know i'm doing it right because i practiced using it so i knew exactly what to do.
Luckyly, Slackware does have a '/etc/syslog', but it doesn't have a kern.log file.
Here are all my /var/log files from Slackware.

Code:

btmp        dmesg    kdm.log  packages          sa      setup    Xorg.0.log

ConsoleKit  faillog  lastlog  pm-powersave.log  samba    spooler  Xorg.0.log.old

cron        hp      maillog  pm-suspend.log    sbopkg  syslog

cups        httpd    messages  removed_packages  scripts  uucp

debug      iptraf  nfsd      removed_scripts  secure  wtmp

I've booted my Opensuse partition so no new info will be written and now i've mounted my Slack partition. Here is the output of:

Code:

tail /mnt/var/log/syslog

Code:

Oct 22 11:37:00 darkstar NetworkManager[1826]: nm_ip4_config_add_wins: assertion `wins > 0' failed

Oct 22 11:37:00 darkstar NetworkManager[1826]: nm_ip4_config_add_wins: assertion `wins > 0' failed

Oct 22 11:53:01 darkstar kernel: [ 3250.098422] CPU2: Package power limit notification (total events = 69)

Oct 22 11:53:01 darkstar kernel: [ 3250.098426] CPU1: Package power limit notification (total events = 69)

Oct 22 11:53:01 darkstar kernel: [ 3250.098429] CPU3: Package power limit notification (total events = 69)

Oct 22 11:53:01 darkstar kernel: [ 3250.098431] CPU0: Package power limit notification (total events = 69)

Oct 22 11:58:28 darkstar kernel: [ 3577.221223] CPU1: Package power limit notification (total events = 102)

Oct 22 11:58:28 darkstar kernel: [ 3577.221226] CPU0: Package power limit notification (total events = 102)

Oct 22 11:58:28 darkstar kernel: [ 3577.221230] CPU3: Package power limit notification (total events = 102)

Oct 22 11:58:28 darkstar kernel: [ 3577.221232] CPU2: Package power limit notification (total events = 102)

Quote:

Have checked to see if you're overheating? I ask because my machine was freezing up, before by dc adapter went out. So when I replaced it I found that the vent for the fan was clogged. Cleaned it & haven't had a problem since.

If you haven't already install lm-sensors.
run command

Quote:

sensors

Thanks for the suggestion EDDY1, i've just installed that package on my Debian partition. My cores are running at 52 degrees C at the moment apparently. 86 is regarded as high.
I just thought of something the other day that i thought may be causing this. I have experienced these freezes on Opensuse 12.2, Debian 7, and Slackware 14. I never got these freezes on previous versions. I was thinking it may be due changes in the kerenl.
Opensuse = linux-3.4.6
Debian = linux-3.2
Slackware = linux-3.2.29
My computer is a Thinkpad x220 with Intel HD Graphics 3000.
To enhance battery life unde Gnu/Linux i'd always been running the following kernel parameters.

Code:

pcie_aspm=force i915.i915_enable_rc6=1 i915.lvds_downclock=1 i915.i915_enable_fbc=1

To be honest, i don't know what they actually do, but i know they give me around two and half more hours of battery life on my hardware. Because of changes in the kernel, i was thinking that these parameters may be affecting my system in a way that they never used to. I have temporarily removed these parameters to see if the freezes continue. So far i've been running Debian for four days with no freezes, but i wouldn't say the freezes happen that often either, so i'm still testing. I just wanted to share this theory to you guys who know what your doing, to get an idea if these parameters may be the reason my systems freeze?

thank you