alex.dupuy at mac.com
Wed Aug 4 14:33:48 CDT 2010
Rahul Nabar wrote:
> I monitor temperatures via lm_sensors. Again in-band. I try to keep my
> monitoring in-band unless there is a compelling reason to use ipmi.
> Maybe some sensors are not available to lm_sensors.
My experience with Dell server systems that have BMC or iDRAC cards
standard (9th/10th/11th gen, at least) is that lm_sensors doesn't have
any usable sensors to monitor, as Dell have wired them all up to the BMC
instead (AMD CPU temperature sensors builtin to the CPU itself perhaps
being one exception - but all my Dell servers are Intel).
You can still do in-band management, but you end up running the
monitoring via local IPMI anyhow, e.g.
# ipmitool sdr type Temperature
Temp | 01h | ok | 3.1 | -53 degrees C
Temp | 02h | ok | 3.2 | -52 degrees C
Temp | 05h | ok | 10.1 | 16 degrees C
Ambient Temp | 07h | ok | 10.1 | 87 degrees C
Temp | 06h | ok | 10.2 | 42 degrees C
Ambient Temp | 08h | ok | 10.2 | 19 degrees C
Ambient Temp | 0Eh | ok | 7.1 | 23 degrees C
Planar Temp | 0Fh | ok | 7.1 | 40 degrees C
IOH THERMTRIP | 5Dh | ns | 7.1 | Disabled
CPU Temp Interf | 76h | ns | 7.1 | Disabled
Temp | 0Ah | ok | 8.1 | 26 degrees C
Temp | 0Bh | ok | 8.1 | 28 degrees C
Temp | 0Ch | unc | 8.1 | 42 degrees C
(those wacky -53 degree readings are relative to CPU melting point or
something - not sure what the 87 degree ambient temperature is about -
but basically I just look for ok/unc/ucr for nominal/warning/error state
and ignore the specific temps anyhow).
mailto:alex.dupuy at mac.com
More information about the Linux-PowerEdge