Deployment
Basic CMD
# Create
kubectl create deploy NAME --image=IMAGE --replica NUM
# Scale (or use kubectl edit)
kubectl scale deploy NAME --replicas NUM
# Change image of deploy
kubectl set image deployments/DP_NAME ContainerName=IMAGE
k set image deployments/apache httpd=httpd:alpine
When using
kubectl createcmd create the deployment, it will automatically add the samelabelsfor both deployment and pods. So if you want to customize thelabel, it’s better to print out--dry-run=client -oyamlthe yaml file and have a check.
And if not clear the container name, can use following to find it
k get pod NAME -oyaml | grep -3 -i container
Rollout
k rollout history deployment/DP_NAME
k rollout history deployment/DP_NAME --revision REV_NUM
k rollout undo deployment/DP_NAME --to-revision REV_NUM
k rollout status deployment DP_NAME
controlplane $ k rollout history deployment video-app
deployment.apps/video-app
REVISION CHANGE-CAUSE
1 <none>
2 <none>
3 <none>
controlplane $ k rollout history deployment/video-app --revision 3
deployment.apps/video-app with revision #3
Pod Template:
Labels: app=video-app
pod-template-hash=69648db755
Containers:
redis:
Image: redis:7.0.13
Port: <none>
Host Port: <none>
Environment: <none>
Mounts: <none>
Volumes: <none>
Node-Selectors: <none>
Tolerations: <none>
controlplane $ k rollout undo deployment/video-app --to-revision 3
deployment.apps/video-app skipped rollback (current template already matches revision 3)
Rolling Strategy
.spec.strategy.rollingUpdate.Unavailable and .spec.strategy.rollingUpdate.maxSurge
- Max Unavailable
- unavailable during the update process.
- Max Surge
- specifies the maximum number of Pods that can be created over the desired number of Pods
strategy:
type: RollingUpdate
rollingUpdate:
maxSurge: 30%
maxUnavailable: 45%
.spec.strategy.type==Recreate, and we can also destroy all existing pods first then create the new pods.
strategy:
type: Recreate