Skip to content

Find out how much resources your Kubernetes Pods actually use and visualise CPU throttling.

Notifications You must be signed in to change notification settings

wuestkamp/k8s-example-resource-monitoring

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

13 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Kubernetes Resource Monitoring by example

Medium article: https://medium.com/@wuestkamp/k8s-monitor-pod-cpu-and-memory-usage-with-prometheus-28eec6d84729?source=friends_link&sk=b498011bceb730596ee93d56869a2f5c

NOTICE K8s > 1.16

If you’re using Kubernetes >=1.16 you’ll have to use pod instead of pod_name and container instead of container_name.

https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG/CHANGELOG-1.16.md#removed-metrics

run yourself

You need a cluster, like with Gcloud:

gcloud container clusters create resources --num-nodes 3 --zone europe-west3-b --machine-type n1-standard-2 --node-version=1.15 --cluster-version=1.15

Install helm repos:

Helmfile is used for managing helm repos.

kubectl create ns grafana
kubectl create ns prometheus

# if you need to install metrics-server create the namespace and uncomment code in i/helm/helmfile.yaml
#kubectl create ns metrics-server

cd i/helm
helmfile diff
helmfile sync

Install k8s test app:

kubectl apply -f i/k8s

monitor

Run Grafana

./run/grafana.sh

Then head to http://localhost:3000/login

User: admin

Password: printed in the terminal output.

install Dashboard

Go to http://localhost:3000/dashboard/import and import the dashboard from i/grafana/dashboard_k8s_1.15.json or i/grafana/dashboard_k8s_1.16.json.

There are different dashboards because metric names have changes from k8s 1.16 on.

Use resources

kubectl run curl --image=curlimages/curl --rm --restart=Never -it sh

curl --data "millicores=400&durationSec=300" compute:8080/ConsumeCPU
curl --data "megabytes=400&durationSec=300" compute:8080/ConsumeMem

Prometheus queries K8s till 1.15

cpu

# container usage
rate(container_cpu_usage_seconds_total{pod=~"compute-.*", image!="", container_name!="POD"}[5m])

# container requests
avg(kube_pod_container_resource_requests_cpu_cores{pod=~"compute-.*"})

# container limits
avg(kube_pod_container_resource_limits_cpu_cores{pod=~"compute-.*"})

# throttling
rate(container_cpu_cfs_throttled_seconds_total{pod=~"compute-.*", container_name!="POD", image!=""}[5m])

memory

# container usage
container_memory_working_set_bytes{pod_name=~"compute-.*", image!="", container_name!="POD"}

# container requests
avg(kube_pod_container_resource_requests_memory_bytes{pod=~"compute-.*"})

# container limits
avg(kube_pod_container_resource_limits_memory_bytes{pod=~"compute-.*"})

Prometheus queries K8s from 1.16

cpu

# container usage
rate(container_cpu_usage_seconds_total{pod=~"compute-.*", image!="", container!="POD"}[5m])

# container requests
avg(kube_pod_container_resource_requests_cpu_cores{pod=~"compute-.*"})

# container limits
avg(kube_pod_container_resource_limits_cpu_cores{pod=~"compute-.*"})

# throttling
rate(container_cpu_cfs_throttled_seconds_total{pod=~"compute-.*", container!="POD", image!=""}[5m])

memory

# container usage
container_memory_working_set_bytes{pod=~"compute-.*", image!="", container!="POD"}

# container requests
avg(kube_pod_container_resource_requests_memory_bytes{pod=~"compute-.*"})

# container limits
avg(kube_pod_container_resource_limits_memory_bytes{pod=~"compute-.*"})

About

Find out how much resources your Kubernetes Pods actually use and visualise CPU throttling.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages