Temperature monitoring and stress testing

Install

helm upgrade --install monitor ./helm

Temperature

kubectl exec -it $(kubectl get pod --selector=app=monitor --output=jsonpath={.items..metadata.name}) -c glances -- watch -n1 sensors

CPU Throttling

kubectl exec -it $(kubectl get pod --selector=app=monitor --output=jsonpath={.items..metadata.name}) -c glances -- watch -n1 lscpu

Stress

kubectl exec -it $(kubectl get pod --selector=app=monitor --output=jsonpath={.items..metadata.name}) -c stress -- stress -c 48

Overclocking

https://www.techpowerup.com/review/amd-ryzen-threadripper-3000-overclocking-deep-dive-asus-rog-zenith-ii-extreme/3.html

kubectl exec -it $(kubectl get pod --selector=app=monitor --output=jsonpath={.items..metadata.name}) -c stress -- bash time $(i=0; while (( i < 9999999 )); do (( i ++ )); done)

Stock

real 0m15.979s user 0m15.956s sys 0m0.000s

real 0m15.804s user 0m15.803s sys 0m0.000s

4.1@1.15v

real 0m16.364s user 0m16.363s sys 0m0.000s

real 0m16.441s user 0m16.439s sys 0m0.000s

Disk

Test command:

dd if=/dev/urandom of=$DISK/output bs=1M count=1024; dd if=$DISK/output of=/dev/null; rm -f $DISK/output
dd if=/dev/zero of=$DISK/output bs=1M count=1024; dd if=$DISK/output of=/dev/null; rm -f $DISK/output

kubectl exec -it $(kubectl get pod --selector=app=monitor --output=jsonpath={.items..metadata.name}) -c disk -- bash
export DISK=/tmp/hdd/
export DISK=/tmp/usbssd/
export DISK=/tmp/nvme/

Mainframe

export DISK=/nvme

Freenas

export DISK=/mnt/enc0/kube
export DISK=/mnt/USB_SSD/kube

1.7 KiB Raw Blame History