How do I interpret mcelog output?

[linuxdork]

I have a 3 month old Lenovo U430 Touch. I keep getting the following errors in /var/log/messages and the abrt keeps annoying the crap out of me about it. The error seems intermittent. I've also encountered the error when using a SSD. All the searching I've done eludes to an actual hardware error, likely with the CPU or RAM. I found the same message when running mcelog on an Ubuntu live CD. I then ran the Intel Diagnostic tool under Windows 8 and that passed on all tests (including memory and CPU). I also ran the Windows Memory tests without error. What am I missing here? Could this in fact be a software issue or false possitive?

    Jun 18 20:51:58 -hostname- mcelog: Hardware event. This is not a software error.
    Jun 18 20:51:58 -hostname- mcelog: MCE 0
    Jun 18 20:51:58 -hostname- mcelog: CPU 0 BANK 5
    Jun 18 20:51:58 -hostname- mcelog: MISC 138a0000086 ADDR ff887540
    Jun 18 20:51:58 -hostname- mcelog: TIME 1403124697 Wed Jun 18 16:51:37 2014
    Jun 18 20:51:58 -hostname- mcelog: MCG status:
    Jun 18 20:51:58 -hostname- mcelog: MCi status:
    Jun 18 20:51:58 -hostname- mcelog: Uncorrected error
    Jun 18 20:51:58 -hostname- mcelog: MCi_MISC register valid
    Jun 18 20:51:58 -hostname- mcelog: MCi_ADDR register valid
    Jun 18 20:51:58 -hostname- mcelog: Processor context corrupt
    Jun 18 20:51:58 -hostname- mcelog: MCA: corrected filtering (some unreported errors in same region)
    Jun 18 20:51:58 -hostname- mcelog: Generic CACHE Level-2 Generic Error
    Jun 18 20:51:58 -hostname- mcelog: STATUS ae0000000040110a MCGSTATUS 0
    Jun 18 20:51:58 -hostname- mcelog: MCGCAP c07 APICID 0 SOCKETID 0
    Jun 18 20:51:58 -hostname- mcelog: CPUID Vendor Intel Family 6 Model 69
    Jun 18 20:51:58 -hostname- mcelog: Hardware event. This is not a software error.
    Jun 18 20:51:58 -hostname- mcelog: MCE 1
    Jun 18 20:51:58 -hostname- mcelog: CPU 0 BANK 6
    Jun 18 20:51:58 -hostname- mcelog: MISC 1b8a0000086 ADDR ff887480
    Jun 18 20:51:58 -hostname- mcelog: TIME 1403124697 Wed Jun 18 16:51:37 2014
    Jun 18 20:51:58 -hostname- mcelog: MCG status:
    Jun 18 20:51:58 -hostname- mcelog: MCi status:
    Jun 18 20:51:58 -hostname- mcelog: Error overflow
    Jun 18 20:51:58 -hostname- mcelog: Uncorrected error
    Jun 18 20:51:58 -hostname- mcelog: MCi_MISC register valid
    Jun 18 20:51:58 -hostname- mcelog: MCi_ADDR register valid
    Jun 18 20:51:58 -hostname- mcelog: Processor context corrupt
    Jun 18 20:51:58 -hostname- mcelog: MCA: corrected filtering (some unreported errors in same region)
    Jun 18 20:51:58 -hostname- mcelog: Generic CACHE Level-2 Generic Error
    Jun 18 20:51:58 -hostname- mcelog: STATUS ee0000000040110a MCGSTATUS 0
    Jun 18 20:51:58 -hostname- mcelog: MCGCAP c07 APICID 0 SOCKETID 0
    Jun 18 20:51:58 -hostname- mcelog: CPUID Vendor Intel Family 6 Model 69

[dobbi]

DESCRIPTION
X86 CPUs report errors detected by the CPU as machine check events (MCEs). These can be data corruption detected in the CPU
caches, in main memory by an integrated memory controller, data transfer errors on the front side bus or CPU interconnect or
other internal errors. Possible causes can be cosmic radiation, instable power supplies, cooling problems, broken hardware,
or bad luck.

Most errors can be corrected by the CPU by internal error correction mechanisms. Uncorrected errors cause machine check
exceptions which may panic the machine.

When a corrected error happens the x86 kernel writes a record describing the MCE into a internal ring buffer available
through the /dev/mcelog device mcelog retrieves errors from /dev/mcelog, decodes them into a human readable format and
prints them on the standard output or optionally into the system log.

Optionally it can also take more options like keeping statistics or triggering shell scripts on specific events.

source: man mcelog

This also may interest you:

You might like to try replacing the DIMM in question (CPU 0, SOCKET 8) and seeing whether the MCE messages continue to be generated.

The mcelog package comes configured with some default thresholds for various MCE events that occur over time. Check out /etc/mcelog/mcelog.conf for details. For memory page errors the threshold is 10 events over 24 hours. (I'm not really sure where this number comes from but it's probably a reasonable reference point). Your post mentions 77 correctable events over 24 hours against a whole bunch of pages, so it's pretty likely that the DIMM has developed a problem which may or may not turn into something more serious.

I wouldn't be too upset about receiving inconsistent information from different sources. In general I have found that anything at the firmware level is pretty platform specific (i.e particular to that particular hardware model). My rule of thumb for firmware-related problems is that the vendor tools are usually the most accurate, but the least usable. The more generic open source tools are easier to work with, but may not provide enough information to show exactly what's going on.

source: http://serverfault.com/questions/447...t-mce-messages

Good advice in there.

This is a hardware problem, those messages come directly from the processor not the software part, change it, find a firmware, contact the manufacturer.

To find out what CPU0 bank 5 you can use dmidecode it has a lot of options.
ps: But everytime I see the word bank I am sure it is the memory sticks.

Resources:
/etc/mcelog
/etc/cron.hourly/mcelog.cron
/etc/mcelog/mcelog.conf
/etc/mcelog/mcelog.setup
/etc/mcelog/triggers
/etc/mcelog/triggers/cache-error-trigger
/etc/mcelog/triggers/dimm-error-trigger
/etc/mcelog/triggers/page-error-trigger
/etc/mcelog/triggers/socket-memory-error-trigger
/etc/systemd/system/multi-user.target.wants/mcelog.service
/usr/include/asm/mce.h
/usr/share/man/man8/mcelog.8.gz (man page)

[linuxdork]

Sound advice. Thank you for your reply. I don't have a memory bank 5, at least not as far as dmidecode is concerned. It only goes up from 0 to 3 (interesting, since this is a laptop with only 1 slot/bank available on the board). I've yet to experience any hangs or crashes as a result of these errors. Any other ideas on how to determine what piece of hardware BANK 5 and BANK 6 refer to? Hoping desperately, I don't have to send this thing in for repair. I really am having the worst luck with laptops this year.

[dobbi]

Sorry, does this shows anything?

# mcelog --client

source: http://www.mcelog.org/memory.html

According to the man pages a certain amount of errors is to be expected only when a lot of them happen in a short time there should be need to worry about.

Do you live near a power line?
Coronas discharge in power lines can be a source of grief.
You can even check for those with a camera they show up as bright spots in any camera image because the high energy particles overwhelm the image sensors.

http://www.powerlinepatrol.com/images/uvcorona1.jpg

Else you can configure the triggers so it doesn't report those constantly.

Resources:
https://en.wikipedia.org/wiki/Memory_bank
https://en.wikipedia.org/wiki/Memory_address
https://en.wikipedia.org/wiki/Memory_geometry
http://halobates.de/lk10-mcelog.pdf (PDF file)

[linuxdork]

All the power around here seems to be underground. I also see the issue at work. I'm going to just ride it out until there is an issue I can see, such as a system lockup. Thank you for all your help!