I have an AMD system with an AMD Epyc 7302P (16C/32T over 4 CCDs; 4 cores per CCD) and I'd like to monitor CPU temperature using the graphite server reporting feature in TrueNAS 13.0 U4.
If I enable the graphite reporting feature, I get 32 temperature readings in database with names like:
If I check with
So it seems those values come from there. Is there any way to read the temperatures from the individual cores?
For instance,
At least there are individual readings from the individual CCDs, but I have no clue why this is repeated 4 times?!
Anyway, those
The dashboard isn't working correctly either. All CPU threads have the same temperature:
If I enable the graphite reporting feature, I get 32 temperature readings in database with names like:
servers.<server_name>.cputemp.<core#>.temperature
, but they all report the exact same value.If I check with
sysctl
, I get something like this:Code:
dev.cpu.31.temperature: 51.7C dev.cpu.30.temperature: 51.7C dev.cpu.29.temperature: 51.7C dev.cpu.28.temperature: 51.7C dev.cpu.27.temperature: 51.7C dev.cpu.26.temperature: 51.7C dev.cpu.25.temperature: 51.7C dev.cpu.24.temperature: 51.7C dev.cpu.23.temperature: 51.7C dev.cpu.22.temperature: 51.7C dev.cpu.21.temperature: 51.7C dev.cpu.20.temperature: 51.7C dev.cpu.19.temperature: 51.7C dev.cpu.18.temperature: 51.7C dev.cpu.17.temperature: 51.7C dev.cpu.16.temperature: 51.7C dev.cpu.15.temperature: 51.7C dev.cpu.14.temperature: 51.7C dev.cpu.13.temperature: 51.7C dev.cpu.12.temperature: 51.7C dev.cpu.11.temperature: 51.7C dev.cpu.10.temperature: 51.7C dev.cpu.9.temperature: 51.7C dev.cpu.8.temperature: 51.7C dev.cpu.7.temperature: 51.7C dev.cpu.6.temperature: 51.7C dev.cpu.5.temperature: 51.7C dev.cpu.4.temperature: 51.7C dev.cpu.3.temperature: 51.7C dev.cpu.2.temperature: 51.7C dev.cpu.1.temperature: 51.7C dev.cpu.0.temperature: 51.7C
So it seems those values come from there. Is there any way to read the temperatures from the individual cores?
For instance,
sysctl
has some entries from amdtemp
as well. Something likle this:Code:
dev.amdtemp.3.ccd6: 51.5C dev.amdtemp.3.ccd4: 49.6C dev.amdtemp.3.ccd2: 54.6C dev.amdtemp.3.ccd0: 47.5C dev.amdtemp.3.core0.sensor0: 51.7C dev.amdtemp.3.sensor_offset: 0 dev.amdtemp.3.%parent: hostb32 dev.amdtemp.3.%pnpinfo: dev.amdtemp.3.%location: dev.amdtemp.3.%driver: amdtemp dev.amdtemp.3.%desc: AMD CPU On-Die Thermal Sensors dev.amdtemp.2.ccd6: 51.5C dev.amdtemp.2.ccd4: 49.6C dev.amdtemp.2.ccd2: 54.6C dev.amdtemp.2.ccd0: 47.5C dev.amdtemp.2.core0.sensor0: 51.7C dev.amdtemp.2.sensor_offset: 0 dev.amdtemp.2.%parent: hostb16 dev.amdtemp.2.%pnpinfo: dev.amdtemp.2.%location: dev.amdtemp.2.%driver: amdtemp dev.amdtemp.2.%desc: AMD CPU On-Die Thermal Sensors dev.amdtemp.1.ccd6: 51.5C dev.amdtemp.1.ccd4: 49.6C dev.amdtemp.1.ccd2: 54.6C dev.amdtemp.1.ccd0: 47.5C dev.amdtemp.1.core0.sensor0: 51.7C dev.amdtemp.1.sensor_offset: 0 dev.amdtemp.1.%parent: hostb8 dev.amdtemp.1.%pnpinfo: dev.amdtemp.1.%location: dev.amdtemp.1.%driver: amdtemp dev.amdtemp.1.%desc: AMD CPU On-Die Thermal Sensors dev.amdtemp.0.ccd6: 51.5C dev.amdtemp.0.ccd4: 49.6C dev.amdtemp.0.ccd2: 53.5C dev.amdtemp.0.ccd0: 47.5C dev.amdtemp.0.core0.sensor0: 51.7C dev.amdtemp.0.sensor_offset: 0 dev.amdtemp.0.%parent: hostb0 dev.amdtemp.0.%pnpinfo: dev.amdtemp.0.%location: dev.amdtemp.0.%driver: amdtemp dev.amdtemp.0.%desc: AMD CPU On-Die Thermal Sensors dev.amdtemp.%parent:
At least there are individual readings from the individual CCDs, but I have no clue why this is repeated 4 times?!
Anyway, those
amdtemp
readings do not show up in the graphite server. Is there any way to get those amdtemp
values to the graphite server without scripting this myself?The dashboard isn't working correctly either. All CPU threads have the same temperature:
Last edited: