I have a cluster and node creates in AWS EKS. I applied the deployment to that cluster as under
kubectl apply -f deployment.yaml
Where deployment.yaml contains the containers' specification along with DockerHub repo and image
However, I did a mistake in deployment.yaml and I need to re-apply it to the configuration
My question is:
1 - How do I reapply a deployment.yaml to the AWS EKS cluster using kubectl?
Just running the above command is not working (kubectl apply -f deployment.yaml)
2- After I re-apply the deployment.yaml , will the node will go an pick up the DockerHub image or do I still need to do something else( supposing all the other details are ok)
Some outputs below:
>> kubectl get pods
my-app-786dc95d8f-b6w4h 0/1 ImagePullBackOff 0 9h
my-app-786dc95d8f-w8hkg 0/1 ImagePullBackOff 0 9h
kubectl describe pod my-app-786dc95d8f-b6w4h
Name: my-app-786dc95d8f-b6w4h
Namespace: default
Priority: 0
Node: ip-192-168-24-13.ec2.internal/192.168.24.13
Start Time: Fri, 10 Jul 2020 12:54:38 -0400
Labels: app=my-app
pod-template-hash=786dc95d8f
Annotations: kubernetes.io/psp: eks.privileged
Status: Pending
IP: 192.168.7.235
IPs:
IP: 192.168.7.235
Controlled By: ReplicaSet/my-app-786dc95d8f
Containers:
simple-node:
Container ID:
Image: BAD_REPO/simple-node
Image ID:
Port: 80/TCP
Host Port: 0/TCP
State: Waiting
Reason: ImagePullBackOff
Ready: False
Restart Count: 0
Environment: <none>
Mounts:
/var/run/secrets/kubernetes.io/serviceaccount from default-token-mwwvl (ro)
Conditions:
Type Status
Initialized True
Ready False
ContainersReady False
PodScheduled True
Volumes:
default-token-mwwvl:
Type: Secret (a volume populated by a Secret)
SecretName: default-token-mwwvl
Optional: false
QoS Class: BestEffort
Node-Selectors: <none>
Tolerations: node.kubernetes.io/not-ready:NoExecute for 300s
node.kubernetes.io/unreachable:NoExecute for 300s
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal BackOff 17m (x2570 over 9h) kubelet, ip-192-168-24-13.ec2.internal Back-off pulling image "BAD_REPO/simple-node"
Warning Failed 2m48s (x2634 over 9h) kubelet, ip-192-168-24-13.ec2.internal Error: ImagePullBackOff
BR
if you need to change image:
kubectl set image deployment.v1.apps/{your_deployment_name} image_name:tag
but you always can do
kubectl delete -f deployment.yaml
kubectl create -f deployment.yaml
since your image is in ImagePullBackOff - it doesn't work anyway and you can just recreate deployment. Usually you don't do drop/create on prod. that is why i am using image change all the time. just have to change tag on every new image.
ImagePullBackOff means that kubernetes is not able to pull the image.
Specially, the service account "default" is not able to pull the image.
To fix this issue, you need two checks: