Could it be there is a max environment temp for the Pi to operate as it should?
I’ve witnessed some strange behavior the last few days, and having ruled out all possible configuration errors I am left to only suspect a too high max temp where the Pi is stored.
I ask because it was working fine in the morning, midday issues started to arise (becoming unresponsive, having enormous lags and many timing errors as a consequence), and later, beginning in the evening, the Pi/Hassio slowly came back. Having a threshold temp of about 34/3 degrees Celsius. Behaving quite immediate now, without having changed the configuration
Might be a silly question, but maybe other community members have related experiences.
I can also correlate high ambient temperatures to RPi becoming unresponsive. An ambient temperature of 26 Celsius appeared to be the limit for the RPi in my garage. I don’t know the equivalent CPU temperature - today all my RPIs have a CPU temp in the mid 60s and all working well.
For my garage RPI it was largely my fault as it was also sitting on an UPS which also generates heat. I’ve now changed the location and will likely add heatsinks as the garage can get very hot.
26°C is NOTHING. We get in the 40’s here in summer. I’ve never had an issue with heat and my Pi’s. I have 2 x Pi 1B’s and a Pi 3 and a NAS etc in a closed cabinet as well as a 24 port gigabit switch and a mini-router… never seen any issue at all.
My Pi usually runs in the low 50’s to 60. I thought OP was asking about the ambient temperature though, not the board running temp.
sensor:
# CPU Temperature
- platform: command_line
name: CPU Temperature
command: "cat /sys/class/thermal/thermal_zone0/temp"
unit_of_measurement: "°C"
value_template: '{{ value | multiply(0.001) | round(1) }}'
Yep, but the only way ambient temperature can cause that is by contributing to the CPU core getting too hot. And that depends on what the CPU is doing, it’s going to be better with a load average of 0.1 at a higher ambient temp than if the la is 4.0. So you need to compare core temp, load and ambient temp to find any relevant correlations.
will create the core temp sensor, and see what happens.
Might also be the other way round, that continuous timing errors cause the Pi to go up in activity and heat because of that. Maybe. I say maybe because the processor use % is mostly round 30 or so with all timing errors included. So that shouldn’t be too intensive to frustrate regular operation.
will report back if anything noteworthy comes out of the core temp sensor.
current state:
taken the lid of the Pi:
seems the board itself generates the heat more than the environment is of influence. What a difference a lid makes…
waving the lid for a minute took the temp down another 5 degrees, but that hardly seems a viable option need a fan.
next-up: checking out the heatsink @BrendanMoran suggested
need a fan.