topic Re: mcelog in BladeSystem - General

mcelog

wali_chabi — Wed, 04 Oct 2017 14:06:39 GMT

Hi all,

someone knows what the following errors are related to?

mcelog: Processor 35 heated above trip temperature. Throttling enabled.
mcelog: Please check your system cooling. Performance will be impacted
mcelog: Running trigger `unknown-error-trigger'
mcelog: Too many trigger children running already
mcelog: Processor 4 heated above trip temperature. Throttling enabled.
mcelog: Please check your system cooling. Performance will be impacted
mcelog: Running trigger `unknown-error-trigger'
mcelog: Too many trigger children running already
mcelog: Processor 26 heated above trip temperature. Throttling enabled.
mcelog: Please check your system cooling. Performance will be impacted
mcelog: Running trigger `unknown-error-trigger'
mcelog: Too many trigger children running already
mcelog: Please check your system cooling. Performance will be impacted
mcelog: Running trigger `unknown-error-trigger'
mcelog: Too many trigger children running already
1 mcelog: Processor 33 heated above trip temperature. Throttling enabled.
mcelog: Please check your system cooling. Performance will be impacted
mcelog: Running trigger `unknown-error-trigger'
mcelog: Too many trigger children running already
mcelog: Processor 24 heated above trip temperature. Throttling enabled.
mcelog: Please check your system cooling. Performance will be impacted

Red Hat Enterprise Linux Server release 6.7 (Santiago)

Re: mcelog

Jimmy Vance — Wed, 04 Oct 2017 16:54:36 GMT

The internal CPU PROCHOT sensor is seeing a temperature above a preset threshold and is throttling the CPU back to help lower the temp. Check the system cooling, and/or the CPU heat sink. The heat sink could be installed improperly, or possibly no enough thermal grease was applied to the CPU.

If the hardware is OK, you may need to change the system cooling setting to increased or maximum cooling. I've seen a few cases where optimal cooling and just the right workload would trip CPU throttling as iLO wasn't ramping up the fans quite fast enough

Re: mcelog

wali_chabi — Thu, 05 Oct 2017 09:24:05 GMT

Thank you Jimmy. I will apply the thermal greace and monitor the server behavior