Hi all,

My system (Slackware64 13.37 on AMD Phenom 2.6GHz with 16 GB of memory) started freezing up on me and I'm wondering if you could help me debug it.

This freeze happens randomly and there seems to be no association with a particular software (First I thought VirtualBox was causing it but it also happened without it running). When it happens, everything becomes non-responsive and key-sequences for SysRq do not work. All I could do is hard-reboot by pressing down the power button.

I ran memtest86+ but nothing seems to be wrong with memory.

It already froze up twice in the last hour...

Could anyone help me debug this problem? Any suggestions would be appreciated.

Thanks,
gibb

Are you able to see or getting any error messages? Can you share the errors with us?

Hi Satyaveer Arya,

Thanks for your response.

No, I don't have any error messages. I checked /var/log/messages but didn't see any suspicious messages. Should I look for somewhere else?

Is it possible to set up memory dump (or something similar) at the time of freeze-up? Since I can't trigger sysrq manually, I have no idea to get anything.

Thanks in advance,
gibb

I don't know exactly how to set up memory dump. So, let any guru or gurus see your thread, maybe they can help you somehow.
Checkout kernel logs and boot logs.
Files are -
/var/log/kern.log
/var/log/boot.log

I had a similar problem with my system. My issue was that the memory was overheating and causing instability. How I debugged this issue was removed all of the memory sticks. Inserted one stick and booted. Repeat with all sticks of memory. If problem doesn't occur then you most likely have a similar issue. Insert 2 sticks and test for problem. With my system the problem occurred with 2 or more. If this turns out to be the problem double check latency for memory is compatible with mb and if it is you might want to contact the memory manufacturer to see if could be a defect or buy a cooler for your memory.

I should mention that I was running a windows 7 OS and the error code from the constant bluescreen of death was not helpful because it's error code was saying that it was my video card. I also couldn't find any error codes because the system would just freeze and it wouldn't be caused by an error.

Hi Satyaveer Arya,

I don't seem to have kern.log or boot.log in /var/log...


Hi bilyboy65,

Oh I don't really hope it's a memory issue, but I'll give your suggestion a try; testing each and some combination of modules. It'll take quite a while to report back.

Thanks!
gibb

What logging facility do you use (e.g. syslog, rsyslog, syslog-ng)? It might be that you simply have kern.log disabled (commonly a default unless debugging).

See logging config files (depending on what your system is using):
/etc/rsyslog.conf
/etc/syslog.conf
/etc/syslog-ng/syslog-ng.conf

If there is a - at the beginning of the log file path
then the logging is disabled (also if the line is commented out). Enable it like so
By default on my system, kern logs to /dev/console and I had to change it to /var/log/kern.log. For ryslog there could be custom facility rules in /etc/rsyslog.d/ directory.
SAM

Hi sag47,

Thanks for your post.

I took bilyboy65's advice and swapped around memory modules. What do you know... it was really a problem with memory. I took old two modules that originally came with the system and let it run for a few days. Freeze-up never happened anymore. After swapping with old modules, it froze-up in an hour. I guess I got bad memory modules.

Thank you all for your inputs!
gibb