New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Is this normal ? i am having high cpu temperature with hetzner ?
gemini_geek
Member
in Help
hi,
here are my few logs, system crashed few minutes ago, already opened a ticket, how to approach with them regarding this , any one have experience what to expect in this situation ?
coretemp-isa-0000
Package id 0: +88.0°C (high = +80.0°C, crit = +100.0°C)
Core 0: +63.0°C (high = +80.0°C, crit = +100.0°C)
Core 4: +59.0°C (high = +80.0°C, crit = +100.0°C)
Core 8: +88.0°C (high = +80.0°C, crit = +100.0°C)
Core 12: +60.0°C (high = +80.0°C, crit = +100.0°C)
Core 16: +69.0°C (high = +80.0°C, crit = +100.0°C)
Core 20: +58.0°C (high = +80.0°C, crit = +100.0°C)
Core 24: +59.0°C (high = +80.0°C, crit = +100.0°C)
Core 28: +56.0°C (high = +80.0°C, crit = +100.0°C)
Core 32: +56.0°C (high = +80.0°C, crit = +100.0°C)
Core 33: +56.0°C (high = +80.0°C, crit = +100.0°C)
Core 34: +56.0°C (high = +80.0°C, crit = +100.0°C)
Core 35: +56.0°C (high = +80.0°C, crit = +100.0°C)
Core 36: +55.0°C (high = +80.0°C, crit = +100.0°C)
Core 37: +55.0°C (high = +80.0°C, crit = +100.0°C)
Core 38: +55.0°C (high = +80.0°C, crit = +100.0°C)
Core 39: +55.0°C (high = +80.0°C, crit = +100.0°C)
---
03:31:44 up 1:07, 2 users, load average: 5.24, 5.01, 5.20
---
Thanked by 1dev077
Comments
Hetzner offer one of the cheapest dedis for a reason
Run a cpu stress test (install 'stress-ng' then run 'stress-ng --cpu X' where X is your core count) and copy the temps under load. If it hits 90c then ask Hetzner to replace the CPU cooling. Simple as that.
done all that sent them, the issue is with cooling for sure , they are investigating it, but i am not sure i should keep this even after its fixed.
it happened as i was in between moving another service and was about to point DNS
i have some thing like this from them , so its just an hardware issue
Load average is lower here though
its a vps, its 4:30 am here
Hi there @gemini_geek - Please do your best to continue to document the issue as best as possible and continue to communicate with our team via the support ticket that you have already opened. Our team will do their best to help you. --Katie
Those temperatures aren’t normal for a Hetzner dedicated server, and a crash confirms a cooling or hardware issue. Just tell Hetzner the CPU package is hitting around 88°C under moderate load and the system is shutting down, and they’ll check the heatsink, fans, or thermal paste and repair or replace the hardware.
they confirmed and changed fan + put new paste, but still the temperature is high , i ran stress test for 2 min it reached 100.
i am backing up my data , i will let it go may be buy a new one , this time will stress test before i deploy
What CPU is this? Those temps would be acceptable on my 13900k lol
interesting this is 12th Gen Intel(R) Core(TM) i9-12900K
For the i9-12900K, normal temperatures under load are typically in the 70-85°C range, though it's common for it to hit 90-95°C, and even briefly 100°C during intense tasks like benchmarking.
This is normal. :-)
Check the load in my first post, the thermal throttle number was 10000+ in 10 hrs for almost all cores
its like this now
i9-12900K is a power hog. My bet is the chassis can't handle the heat.
My 12900 with Hetz, load around 20%
Package id 0: +74.0°C (high = +80.0°C, crit = +100.0°C)
Core 0: +74.0°C (high = +80.0°C, crit = +100.0°C)
Core 4: +64.0°C (high = +80.0°C, crit = +100.0°C)
Core 8: +70.0°C (high = +80.0°C, crit = +100.0°C)
Core 12: +66.0°C (high = +80.0°C, crit = +100.0°C)
Core 16: +73.0°C (high = +80.0°C, crit = +100.0°C)
Core 20: +73.0°C (high = +80.0°C, crit = +100.0°C)
Core 24: +73.0°C (high = +80.0°C, crit = +100.0°C)
Core 28: +74.0°C (high = +80.0°C, crit = +100.0°C)
Core 32: +66.0°C (high = +80.0°C, crit = +100.0°C)
Core 33: +66.0°C (high = +80.0°C, crit = +100.0°C)
Core 34: +66.0°C (high = +80.0°C, crit = +100.0°C)
Core 35: +66.0°C (high = +80.0°C, crit = +100.0°C)
Core 36: +64.0°C (high = +80.0°C, crit = +100.0°C)
Core 37: +64.0°C (high = +80.0°C, crit = +100.0°C)
Core 38: +64.0°C (high = +80.0°C, crit = +100.0°C)
Core 39: +64.0°C (high = +80.0°C, crit = +100.0°C)
If you have not done so already, please respond directly to your last support ticket and explain that you are still seeing high temperatures. --Katie
i have critical data, i will reply when i am done backing it up and point my deployed service to new server, i already lost my large redis cache because of crash , i do not want to loose more stuff.
Just cause i found it weird earlier and i still find it weird now. No modern system will crash due to high cpu temps. At all, thats a thing of the ancient past. They all throttle to death but the system will not crash just because its toasty. That is a completely different issue.
What CPU/server is this ?
Had this also a longer time ago and they replaced the cooler and temperature went to normal level again. Maybe thermal paste is pretty old or their cooler is almost dying.
the social coldness in germany means that district heating operators now have to provide more heat to prevent the population from rebelling!
A common approach is to request a full hardware replacement when basic cooling fixes like fan and paste don't resolve persistent overheating issues. Clearly document the new stress test results showing it still hits 100°C. If they can't provide a stable unit, moving to a new server is a sensible option.