Rather frightenng dmesg errors
I have been getting a log of these. Depends on ambient and CPU load, i.e.
killing boinc number-crunchers reduces the occurance, but it happens also when
the ambient is not as high as it is now. I am running the most recent 2.6.39-
PAE 32-bit kernel from Sid on a P4 that looks like two cores which is why they
are in pairs. I do not always get the "Hardware Error" but that is the most
scarey. There are no log entries other than the one-liner.
5292.082495] CPU1: Core temperature above threshold, cpu clock throttled
(total events = 43517)
[ 5292.082525] CPU0: Core temperature above threshold, cpu clock throttled
(total events = 43518)
[ 5292.083623] CPU1: Core temperature/speed normal
[ 5292.083629] CPU0: Core temperature/speed normal
[ 5399.988063] [Hardware Error]: Machine check events logged
LM-Sensors is not correctly measuring temperatures for the MB so yields no
information.
What to do about this besides wait for an end to global warming :-)
Reply to: