Kubernetes part9

Kubernetes part9

Class 94 Kubernetes Part8 August 22nd

Resource Quota: Resource quota is on of the rich features of Kubernetes that helps to manage and distribute resources according to the requirements.

Just assume situation We have 2 teams( team-A,team-B) which are working on single Kubernetes cluster

Team -A need more CPUS & memory because of heavy workload tasks. Team-B pod are not created with resource quota,

whenever team-A application resource high usage trying create one more pod, if space is less delete the team-B pod and occupied the space, because team-B  space not allocated to resource quota.for the reason every pod must resource quota is important

There are tow type of restrictions that need to be mentioned while using resource quota

Limit: Represent the maximum amount of CPUS or memory the pod can use 

(limit: 4 cpus & 4 gb ram)

Request: Represent the minimum amount of CPU or memory that the pod is guaranteed to have

(Pod creation minimum requirement 1cpu & 1 gb ram)

Practical:

Step1: minikube already setup done ec2-instance amazon Linux 

[root@ip-10-0-0-29 ~]# systemctl start docker

[root@ip-10-0-0-29 ~]# minikube start --driver=docker --force

[root@ip-10-0-0-29 ~]# minikube status

minikube

type: Control Plane

host: Running

kubelet: Running

apiserver: Running

kubeconfig: Configured

Step2:Create namespace and going to inside the namespace 

[root@ip-10-0-0-29 ~]# kubectl create ns dev

namespace/dev created

[root@ip-10-0-0-29 ~]# kubectl config  set-context --current --namespace=dev

Context "minikube" modified.

Step3: As you see here  Create resource 

[root@ip-10-0-0-29 ~]# vi resourcequote.yaml

[root@ip-10-0-0-29 ~]# cat resourcequote.yaml

apiVersion: v1

kind: Pod

metadata:

  name: team-a

spec:

  containers:

    - name: container1

      image: nginx

      ports:

        - containerPort: 80

      resources:

        requests:

          memory: "10Mi"

          cpu: "100m"

        limits:

          memory: "20Mi"

          cpu: "200m"

[root@ip-10-0-0-29 ~]# kubectl create -f resourcequote.yaml

pod/team-a created

Step3: As see below namespace is dev and for the pod  created with 100mi memory 10 mi

[root@ip-10-0-0-29 ~]# kubectl describe po team-a

Name:             team-a

Namespace:        dev

Priority:         0

Service Account:  default

Node:             minikube/192.168.49.2

Start Time:       Mon, 25 Aug 2025 15:22:56 +0000

Labels:           <none>

Annotations:      <none>

Status:           Running

IP:               10.244.0.38

 Ready:          True

    Restart Count:  0

    Limits:

      cpu:     200m

      memory:  20Mi

    Requests:

      cpu:        100m

      memory:     10Mi

    Environment:  <none>

Step4: Now trying to create new pod without request options ,give limit  observer what limit you have given automatically it was taken request also same size.

 

[root@ip-10-0-0-29 ~]# cat resourcequote.yaml

apiVersion: v1

kind: Pod

metadata:

  name: team-b

spec:

  containers:

    - name: container1

      image: nginx

      ports:

        - containerPort: 80

      resources:

        limits:

          memory: "20Mi"

          cpu: "200m"

[root@ip-10-0-0-29 ~]# kubectl create -f resourcequote.yaml

pod/team-b created

[root@ip-10-0-0-29 ~]# kubectl describe po team-b

Name:             team-b

Namespace:        dev

Priority:         0

Service Account:  default

Node:             minikube/192.168.49.2

    Limits:

      cpu:     200m

      memory:  20Mi

    Requests:

      cpu:        200m

      memory:     20Mi

Step5: Now we plan to putting request and removed limits 

[root@ip-10-0-0-29 ~]# cat resourcequote.yaml

apiVersion: v1

kind: Pod

metadata:

  name: team-c

spec:

  containers:

    - name: container1

      image: nginx

      ports:

        - containerPort: 80

      resources:

         requests:

          memory: "10Mi"

          cpu: "100m"

[root@ip-10-0-0-29 ~]# kubectl create -f resourcequote.yaml

pod/team-c created

As see below request given it will take request only , no limit that mean it is unlimited , it is not good practice 

[root@ip-10-0-0-29 ~]# kubectl describe po team-c

Name:             team-c

Namespace:        dev

Priority:         0

Service Account:  default

Node:             minikube/192.168.49.2

 Requests:

      cpu:        100m

      memory:     10Mi

    Environment:  <none>

    Mounts:

Note: Three case above 

1. Both request and limit --it as follow limit

2. Limit  limit=request

3. Request request=unlimited

[root@ip-10-0-0-29 ~]# kubectl delete po --all

pod "team-a" deleted

pod "team-b" deleted

pod "team-c" deleted

Realtime 

Step1:

[root@ip-10-0-0-29 ~]# cat realtime.yaml

---

apiVersion: v1

kind: ResourceQuota

metadata:

  name: myreq

spec:

  hard:

    requests.cpu: "258m"

    requests.memory: "500Mi"

    limits.cpu: "1000m"

    limits.memory: "1500Mi"

[root@ip-10-0-0-29 ~]# kubectl create -f realtime.yaml

resourcequota/myreq created

We have added the resource to dev tablespace 

[root@ip-10-0-0-29 ~]# kubectl get ns

NAME              STATUS   AGE

default           Active   4d

dev               Active   3h58m

As see below showing how many cpu 1, you able create memory 

[root@ip-10-0-0-29 ~]# kubectl describe ns dev

Name:         dev

Labels:       kubernetes.io/metadata.name=dev

Annotations:  <none>

Status:       Active

Resource Quotas

  Name:            myreq

  Resource         Used  Hard

  --------         ---   ---

  limits.cpu       0     1

  limits.memory    0     1500Mi

  requests.cpu     0     258m

  requests.memory  0     500Mi

No LimitRange resource.

Step6:Lets create the pod in dev namspace , you must specified the limit other wise pod will not create 

[root@ip-10-0-0-29 ~]# cat pod.yml

apiVersion: v1

kind: Pod

metadata:

  name: team-a

spec:

  containers:

    - name: container1

      image: nginx

      ports:

        - containerPort: 80

      resources:

        requests:

          memory: "10Mi"

          cpu: "100m"

        limits:

          memory: "20Mi"

          cpu: "200m"

--Now namespace Resource quota used dev namespace

Resource Quotas

  Name:            myreq

  Resource         Used  Hard

  --------         ---   ---

  limits.cpu       200m  1

  limits.memory    20Mi  1500Mi

  requests.cpu     100m  258m

  requests.memory  10Mi  500Mi

Step7:now trying to create without resource, see getting must be specified request and limit need to give

[root@ip-10-0-0-29 ~]# kubectl create -f pod.yml

Error from server (Forbidden): error when creating "pod.yml": pods "team-b" is forbidden: failed quota: myreq: must specify limits.cpu for: container1; limits.memory for: container1; requests.cpu for: container1; requests.memory for: container1

[root@ip-10-0-0-29 ~]#

Step8: Created one more pod  team-b CPU 58m left , let us create one more pod 

[root@ip-10-0-0-29 ~]# vi pod.yml

[root@ip-10-0-0-29 ~]# kubectl create -f pod.yml

pod/team-b created

[root@ip-10-0-0-29 ~]# kubectl get pod

NAME     READY   STATUS    RESTARTS   AGE

team-a   1/1     Running   0          7m52s

team-b   1/1     Running   0          15s

[root@ip-10-0-0-29 ~]# kubectl describe ns dev

Name:         dev

Labels:       kubernetes.io/metadata.name=dev

Annotations:  <none>

Status:       Active

Resource Quotas

  Name:            myreq

  Resource         Used  Hard

  --------         ---   ---

  limits.cpu       400m  1

  limits.memory    40Mi  1500Mi

  requests.cpu     200m  258m

  requests.memory  20Mi  500Mi

Step9: Creating more pod ,getting quora exceeded, now increase the request quota

[root@ip-10-0-0-29 ~]# vi pod.yml

[root@ip-10-0-0-29 ~]# kubectl create -f pod.yml

Error from server (Forbidden): error when creating "pod.yml": pods "team-c" is forbidden: exceeded quota: myreq, requested: requests.cpu=100m, used: requests.cpu=200m, limited: requests.cpu=258m

[root@ip-10-0-0-29 ~]#

                                                    Increase the request quota

Step1:increasing cpu 258 to 400m

[root@ip-10-0-0-29 ~]# cat realtime.yaml

---

apiVersion: v1

kind: ResourceQuota

metadata:

  name: myreq

spec:

  hard:

    requests.cpu: "400m"

    requests.memory: "500Mi"

    limits.cpu: "1000m"

    limits.memory: "1500Mi"

[root@ip-10-0-0-29 ~]# kubectl apply -f realtime.yaml

Warning: resource resourcequotas/myreq is missing the kubectl.kubernetes.io/last-applied-configuration annotation which is required by kubectl apply. kubectl apply should only be used on resources created declaratively by either kubectl create --save-config or kubectl apply. The missing annotation will be patched automatically.

resourcequota/myreq configured

--See below resource quota increased

[root@ip-10-0-0-29 ~]# kubectl describe ns dev

Name:         dev

Labels:       kubernetes.io/metadata.name=dev

Annotations:  <none>

Status:       Active

Resource Quotas

  Name:            myreq

  Resource         Used  Hard

  --------         ---   ---

  limits.cpu       400m  1

  limits.memory    40Mi  1500Mi

  requests.cpu     200m  400m

  requests.memory  20Mi  500Mi

[root@ip-10-0-0-29 ~]# kubectl create -f pod.yml

pod/team-c created

[root@ip-10-0-0-29 ~]# kubectl get po

NAME     READY   STATUS    RESTARTS   AGE

team-a   1/1     Running   0          19m

team-b   1/1     Running   0          11m

team-c   1/1     Running   0          18s

--Deleted all pod Resourcequota name dev quota released 

[root@ip-10-0-0-29 ~]# kubectl delete pod --all

pod "team-a" deleted

pod "team-b" deleted

pod "team-c" deleted

[root@ip-10-0-0-29 ~]# kubectl describe ns dev

Name:         dev

Labels:       kubernetes.io/metadata.name=dev

Annotations:  <none>

Status:       Active

Resource Quotas

  Name:            myreq

  Resource         Used  Hard

  --------         ---   ---

  limits.cpu       0     1

  limits.memory    0     1500Mi

  requests.cpu     0     400m

  requests.memory  0     500Mi

--This is my resource quota 

[root@ip-10-0-0-29 ~]# kubectl get ResourceQuota

NAME    REQUEST                                          LIMIT                                      AGE

myreq   requests.cpu: 0/400m, requests.memory: 0/500Mi   limits.cpu: 0/1, limits.memory: 0/1500Mi   30m

PROBES 

Container and container health checks purpose we are using probs 

PROBES: used to determine the health and readiness of containers running within pods. Probes

are 3 types:

1.Readiness probes are used to indicate when a container is ready to receive traffic.

2.Liveness probes are used to determine whether a container is still running and responding to

requests.

3.Startup Probe are used to determines whether the application within the container has

started successfully. It's used to delay the liveness and readiness probes until the application

is ready to handle traffic

WHY PROBES?

• In K8s, its common to scale up and scale down the pods. But when the pod is created newly it
• will take some time to start the container and run the applications.
• If a pod is not ready to receive traffic, it may receive requests that it cannot handle, that
• causes downtime for our application.
• Similarly, if a container is not running correctly, it may not be able to respond to requests,
• resulting in the pod being terminated and replaced with a new one.
• To overcome this issues, we are using Probes.

TYPES OF PROBES:    

There are several types of probes that can be used in Kubernetes which includes HTTP, TCP, and

command probes

Practical:

Real time need to check application running or not using curl  5 sec wait for 30 sec, it application still not responding, container will recreate.

Step1:

[root@ip-10-0-0-29 ~]# cat pod.yml

apiVersion: v1

kind: Pod

metadata:

  name: team-a

spec:

  containers:

    - name: container1

      image: nginx

      ports:

        - containerPort: 80

      args:

        - /bin/bash

        - -c

        - touch /tmp/ccit; sleep 10000

      livenessProbe:

        exec:

          command:

            - cat

            - /tmp/ccit

        initialDelaySeconds: 5

        periodSeconds: 5

        timeoutSeconds: 30

      resources:

        requests:

          memory: "10Mi"

          cpu: "100m"

        limits:

          memory: "20Mi"

          cpu: "200m"

[root@ip-10-0-0-29 ~]# kubectl create -f pod.yml

pod/team-a created

--See here Livness probe failed after 31 second create new container 

[root@ip-10-0-0-29 ~]# kubectl describe pod team-a

Name:             team-a

Namespace:        dev

Priority:         0

Service Account:  default

Node:             minikube/192.168.49.2

 bytes.

  Warning  Unhealthy  76s (x3 over 86s)    kubelet            Liveness probe failed: cat: /tmp/ccit: No such file or directory

  Normal   Killing    76s                  kubelet            Container container1 failed liveness probe, will be restarted

  Normal   Pulling    46s (x2 over 3m56s)  kubelet            Pulling image "nginx"

  Normal   Created    45s (x2 over 3m55s)  kubelet            Created container: container1

  Normal   Started    45s (x2 over 3m55s)  kubelet            Started container container1

  Normal   Pulled     45s                  kubelet            Successfully pulled image "nginx" in 795ms (795ms inc

Step2: Delete ccit file one more time ,1 second pull the image 0 second created one more container

every 5 second it will send the request if 30 seconds not respond ,create new container.

 Type     Reason     Age                  From               Message

  ----     ------     ----                 ----               -------

  Normal   Scheduled  9m41s                default-scheduler  Successfully assigned dev/team-a to minikube

  Normal   Pulled     9m40s                kubelet            Successfully pulled image "nginx" in 786ms (786ms including waiting). Image size: 192385800 bytes.

  Normal   Pulled     6m30s                kubelet            Successfully pulled image "nginx" in 795ms (795ms including waiting). Image size: 192385800 bytes.

  Warning  Unhealthy  31s (x6 over 7m11s)  kubelet            Liveness probe failed: cat: /tmp/ccit: No such file or directory

  Normal   Killing    31s (x2 over 7m1s)   kubelet            Container container1 failed liveness probe, will be restarted

  Normal   Pulling    1s (x3 over 9m41s)   kubelet            Pulling image "nginx"

  Normal   Created    0s (x3 over 9m40s)   kubelet            Created container: container1

  Normal   Started    0s (x3 over 9m40s)   kubelet            Started container container1

  Normal   Pulled     0s                   kubelet            Successfully pulled image "nginx" in 798ms (798ms including waiting). Image size: 192385800 bytes.

So far deployment these are completed.

Deploy

   pods 

     containers 

   name,images.ports

   resources,volume,cm & sec & probes

--Thanks