HPA + Auto Deploy

Horizontal Pod AutoScaler

Deployment placement

Note

the resource property of container must be set, otherwise the scaler can’t get your container runtime metrics

kind: Deployment
apiVersion: apps/v1
metadata:
  name: <deployment name>
  namespace: <namespace>
  labels:
    app: <deployment name>
    k8s-app: <deployment name>
  annotations:
    deployment.kubernetes.io/revision: '21'
spec:
  selector:
    matchLabels:
      k8s-app: <deployment name>
  template:
    metadata:
      name: <deployment name>
      creationTimestamp: null
      labels:
        app: <deployment name>
        k8s-app: <deployment name>
    spec:
      containers:
        - name: <deployment name>
          image: <container-registery/image>
          ports:
            - name: http
              containerPort: 9000
              protocol: TCP
            - name: grpc
              containerPort: 9001
              protocol: TCP
          resources:
            limits:
              cpu: 200m
              # memory: 1024Mi
            requests:
              cpu: 50m
              # memory: 128Mi
          imagePullPolicy: Always
          securityContext:
            privileged: false
      restartPolicy: Always
      terminationGracePeriodSeconds: 30
      dnsPolicy: ClusterFirst
      securityContext: {}
      imagePullSecrets:
        - name: ghcr
      schedulerName: default-scheduler
  strategy:
    type: RollingUpdate
    rollingUpdate:
      maxUnavailable: 25%
      maxSurge: 25%
  revisionHistoryLimit: 10
  progressDeadlineSeconds: 600

hpa placement

apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: <hpa-name>
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: <your desired deployment name>
  minReplicas: 1
  maxReplicas: 10
  metrics:
  - type: Resource
    resource:
      name: cpu
      target:
        type: Utilization
        averageUtilization: 50 # unit: percent

gitlab runner auto rollout to kubernetes cluster

gitlab -> runner -> auto trigger kubectl roll update (/home/ubuntu/.kube)

runner on kubernetes

if we want to let multiple repo run on same gitlab runner, we need to use register group runner (need to create a group)

• offical description
• directly tutorial

create agent config file in project (optional)

To create an agent configuration file:

Choose a name for your agent. The agent name follows the DNS label standard from RFC 1123. The name must:

• Be unique in the project.
• Contain at most 63 characters.
• Contain only lowercase alphanumeric characters or -.
• Start with an alphanumeric character.
• End with an alphanumeric character.
• In the repository, in the default branch, create an agent configuration file at the root:

.gitlab/agents/<agent-name>/config.yaml

You can leave the file blank for now, and configure it later.

Register the agent with GitLab

1. Select Operate > Kubernetes clusters

2. Select Connect a cluster (agent)

* If you want to create a configuration with CI/CD defaults, type a name.
* If you already have an agent configuration file, select it from the list.

3. if you already have config file in your project, selec one. otherwise create a new one

4. get the access token

Update your .gitlab-ci.yml file to run kubectl commands

1. install kubernetes agent in your cluster through helm with the provided token

helm repo add gitlab https://charts.gitlab.io
helm repo update
helm upgrade --install <xxxxx> gitlab/gitlab-agent \
    --namespace gitlab-agent-<xxxxx> \
    --create-namespace \
    --set image.tag=v16.5.0 \
    --set config.token=glagent-YoxxFv-5HxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxUeA \
    --set config.kasAddress=wss://kas.gitlab.com

2. check if gitlab has connected to the cluster agent

3. run kubectl in .gitlab-ci.yml

deploy:
  image:
    name: bitnami/kubectl:latest
    entrypoint: ['']
  script:
    - kubectl config get-contexts
    - kubectl config use-context path/to/agent/repository:agent-name
    - kubectl get pods

notice: If you are not sure what your agent’s context is, open a terminal and connect to your cluster. Run kubectl config get-contexts. (in my case, I need to execute the command directly in the gitlab ci triggered job)

1. push the change to trigger the gitlab workflow and run your own custom kubectl command in the gitlab ci

reference

• k8s agent + gitlab CI/CD
• register the agent with gitlab
• configure agent config file