On Tue, Apr 12, 2016 at 3:15 PM, Chris Murphy
<lists(a)colorremedies.com> wrote:
> On Tue, Apr 12, 2016 at 2:41 PM, CS DBA <cs_dba(a)consistentstate.com> wrote:
>> Hi all;
>>
>> Daily I see anywhere from 10 - 50 or more of these alerts via abrt. I'm
>> running Fedora 23 (KDE Spin) fully up to date on a Lenovo X1 carbon 3rd gen.
>> In all the alerts I see that the trip temp was exceeded, and then dropped
>> below the trip temp within 1 second.
>> Below is a sample of the output.
>>
>> Should I be concerned? If not can I disable the popup alert? If so,
>> recommendations? should I consider cleaning & replacing the cpu thermal
>> compound?
>>
>> Thanks in advance
>>
>>
>>
>>
>> The kernel log indicates that hardware errors were detected.
>> System log may have more information.
>> The last 20 mcelog lines of system log are:
>> ==========================================
>> Apr 12 14:33:19 F23-host mcelog: Please check your system cooling.
>> Performance will be impacted
>> Apr 12 14:33:19 F23-host mcelog: STATUS 8812080b MCGSTATUS 0
>> Apr 12 14:33:19 F23-host mcelog: MCGCAP 1000c07 APICID 1 SOCKETID 0
>> Apr 12 14:33:19 F23-host mcelog: CPUID Vendor Intel Family 6 Model 61
>> Apr 12 14:33:19 F23-host mcelog: Hardware event. This is not a software
>> error.
>> Apr 12 14:33:19 F23-host mcelog: MCE 0
>> Apr 12 14:33:19 F23-host mcelog: CPU 1 THERMAL EVENT TSC 134e63eb3c2f
>> Apr 12 14:33:19 F23-host mcelog: TIME 1460493199 Tue Apr 12 14:33:19 2016
>> Apr 12 14:33:19 F23-host mcelog: Processor 1 below trip temperature.
>> Throttling disabled
>> Apr 12 14:33:19 F23-host mcelog: STATUS 8813080a MCGSTATUS 0
>> Apr 12 14:33:19 F23-host mcelog: MCGCAP 1000c07 APICID 1 SOCKETID 0
>> Apr 12 14:33:19 F23-host mcelog: CPUID Vendor Intel Family 6 Model 61
>> Apr 12 14:33:19 F23-host mcelog: Hardware event. This is not a software
>> error.
>> Apr 12 14:33:19 F23-host mcelog: MCE 1
>> Apr 12 14:33:19 F23-host mcelog: CPU 0 THERMAL EVENT TSC 134e63eb83f6
>> Apr 12 14:33:19 F23-host mcelog: TIME 1460493199 Tue Apr 12 14:33:19 2016
>> Apr 12 14:33:19 F23-host mcelog: Processor 0 below trip temperature.
>> Throttling disabled
>> Apr 12 14:33:19 F23-host mcelog: STATUS 8813080a MCGSTATUS 0
>> Apr 12 14:33:19 F23-host mcelog: MCGCAP 1000c07 APICID 0 SOCKETID 0
>> Apr 12 14:33:19 F23-host mcelog: CPUID Vendor Intel Family 6 Model 61
> I think any MCE event shoudn't be ignored. There's just no way to know
> if it's bogus or not. I get these on my machine periodically to no ill
> effect and there's an upstream kernel bug with no response for years.
> I thought it should be true that the hardware itself won't allow an
> overheat, either GPU or CPU, and yet these messages suggest otherwise.
>
> What I've been doing is using thermald which you can get from copr.
>
https://copr.fedorainfracloud.org/coprs/hadrons123/thermald/
>
> However, what I just realized is that it says that 1.5.3 version build
> has failed, and yet rpm -q shows 1.5.3 is on my system. And 'thermald
> --version' also shows it's 1.5.3 and it's running. But it doesn't
seem
> to be working or producing the same messages it used to, where it'd
> throttle the CPU automatically. Hmmm any wonder why it's seems hotter
> than usual. Crap!
Well it appears to be doing something still, if I relaunch it in debug
mode. I guess maybe the new build is just not as verbose as the
previous build, by default. What I still don't get is how the copr
build state is failed, and yet I have that same build installed.
Thanks for the feedback, I've installed thermald, added it to
/etc/rc.d/rc.local and so far have not seen a single cpu temp alert.