question about reported temperatures

[William A. Mahaffey III]

On 08/04/15 11:40, Quartz wrote:
>> During the subsequent
>> reboots, I got output from healthd saying one of my cores was a bit warm
>> @ 52 (presumably C). However, sysctl reports wildly different temps:
>
>> Why the difference ? Which values are thought to be right or more right
>> ? TIA & have a good one. BTW:
>
> Back in June I asked a couple questions about hardware monitoring 
> packages for reporting temps and fans speed and such. Towards the end 
> of the thread you responded offhand that temp monitoring was quirky on 
> your system:
>
> >> To expand a bit, mbmon & healthd seem to agree on (chipset ?) temp, 
> as distinct from amdtemp. However both only report for 2 CPU's (cores 
> ?), while amdtemp reports (wrongly, IMHO) for all 4 (in my case) .... 
> If healthd &/or mbmon would report (accurately) for all 4 cores, I'd 
> love them more :-) .... <<<
>
> Are you saying it's worse now, or something else changed...?
> (I obviously have no worthwhile input on this topic since I didn't 
> even know half these programs existed two months ago, I'm just asking 
> for details since I think you might've left out some important info 
> specific to your setup).


No, not worse or better, just recently noticable since I just had an 
unexpected reboot :-). My various linux boxen seem to get this right (or 
at least more credible numbers), have for years, in the lm_sensors 
package. This is obviously not a huge problem, but it would be sweet to 
be able to monitor stuff like system temps reliably & accurately, 
presumably more so on possibly distant servers. I *think* all this stuff 
is tied up in the basic system, not pkgs, but I could be wrong. If more 
info is needed, just ask. Have a good one.

-- 

	William A. Mahaffey III

  ----------------------------------------------------------------------

	"The M1 Garand is without doubt the finest implement of war
	 ever devised by man."
                            -- Gen. George S. Patton Jr.