0
714
ViewsHow can I summarize all the resource limits and requests in Kubernetes with kubectl?
I started using Lens and noticed that it gives you some warnings when the pods inside the nodes have limits higher than the actual capacity.
So I tried to get this information with kubectl but I'm new to jsonpath and I just managed to get the raw info using something like this:
kubectl get pods -o=jsonpath='{.items..resources.limits}' -A
That produces something like this:
{"cpu":"200m","memory":"1Gi"} {"cpu":"200m","memory":"1Gi"} {"cpu":"200m","memory":"512Mi"} {"cpu":"500m","memory":"250Mi"} {"memory":"170Mi"} {"memory":"170Mi"} {"cpu":"2","memory":"2Gi"} {"cpu":"2","memory":"2Gi"} {"cpu":"2","memory":"2Gi"} {"cpu":"1","memory":"1Gi"} {"cpu":"1","memory":"1Gi"} {"cpu":"2","memory":"2Gi"} {"cpu":"100m","memory":"128Mi"} {"cpu":"100m","memory":"128Mi"} {"cpu":"500m","memory":"600Mi"} {"cpu":"1","memory":"1Gi"} {"cpu":"100m","memory":"25Mi"} {"cpu":"100m","memory":"25Mi"}
So, my questions are, how can I sum all these values? Will these values be accurate or am I missing any other query? I've checked using LimitRange and the values I got seem to be correct, the results include the limits set by the LimitRange configuration.
3 answers
Answer question0
It is not possible with just kubectl commands. However, you can use the output of kubectl and write a basic shell script to compute the total values.
The following shell script will output total CPU limits(in m units) of all the pods in all namespaces.
res=$(kubectl get pods -o=jsonpath='{.items[*]..resources.limits.cpu}' -A)
let tot=0
for i in $res
do
if [[ $i =~ "m" ]]; then
i=$(echo $i | sed 's/[^0-9]*//g')
tot=$(( tot + i ))
else
tot=$(( tot + i*1000 ))
fi
done
echo $tot
You can extend the above to compute the CPU requests and the memory requests and limits values as well.
0
Unfortunately, it is not possible to do so with the use of the kubectl solely.
However, you could for example consider using:
Returns the arithmetic sum of an array of numbers.
We’ll walk through steps necessary to monitor how much resources (CPU or memory) a Kubernetes pod is using. Hence we’ll look at:
CPU requests / limits / actual usage / throttling
Memory requests / limits / actual usage / termination
or alternatively monitor Kubernetes Node CPU and Memory Requests/Limits:
Node CPU requests/limits are a sum of the CPU requests/limits for all pods running on that node. Similarly, node memory requests/limits are a sum of memory requests/limits of all pods
kube-state-metrics is a simple service that listens to the Kubernetes API server and generates metrics about the state of the objects. (See examples in the Metrics section below.) It is not focused on the health of the individual Kubernetes components, but rather on the health of the various objects inside, such as deployments, nodes and pods.
0
you can use a kubectl plugin to list/sort pods with cpu limits:
kubectl resource-capacity --sort cpu.limit --util --pods