Recover All Changes Happens in Child Clusters
Clusternet-agent provide a feature gate
Recoverywhich allows to rollback any changes inside a child cluster. It’sfalseby default.
So after you try Deploying Applications to Multiple Clusters with Replication Scheduling,
in child cluster child-1, and scale the replicas of nginx-deploy from 3 to 1, the replicas remains at 1:
root@child-1:~# kubectl scale deploy my-nginx -n foo --replicas=1
root@child-1:~# kubectl get po -n foo -w
NAME READY STATUS RESTARTS AGE
my-nginx-66b6c48dd5-9dpxt 1/1 Running 0 58s
As has described the feature gate Recovery is false by default in clusternet-agent, so if you want to ensure the resources deployed
by clusternet exist persistently in a child cluster, you should set this feature gate to true in clusternet-agent:--feature-gates=Recovery=true
Then try again! It will work like below, any changes in child clusters will be recovered.
root@child-1:~# kubectl get po -n foo
NAME READY STATUS RESTARTS AGE
my-nginx-66b6c48dd5-6vqtx 1/1 Running 0 13s
my-nginx-66b6c48dd5-9dpxt 1/1 Running 0 2m40s
my-nginx-66b6c48dd5-jw9kr 1/1 Running 0 13s
root@child-1:~# kubectl scale deploy my-nginx -n foo --replicas=1
deployment.apps/my-nginx scaled
root@child-1:~# kubectl get po -n foo
NAME READY STATUS RESTARTS AGE
my-nginx-66b6c48dd5-9dpxt 1/1 Running 0 3m9s
root@child-1:~# kubectl get po -n foo -w
NAME READY STATUS RESTARTS AGE
my-nginx-66b6c48dd5-9dpxt 1/1 Running 0 2m25s
my-nginx-66b6c48dd5-jw9kr 0/1 Pending 0 0s
my-nginx-66b6c48dd5-jw9kr 0/1 Pending 0 0s
my-nginx-66b6c48dd5-6vqtx 0/1 Pending 0 0s
my-nginx-66b6c48dd5-6vqtx 0/1 Pending 0 0s
my-nginx-66b6c48dd5-jw9kr 0/1 ContainerCreating 0 0s
my-nginx-66b6c48dd5-6vqtx 0/1 ContainerCreating 0 0s
my-nginx-66b6c48dd5-jw9kr 1/1 Running 0 1s
my-nginx-66b6c48dd5-6vqtx 1/1 Running 0 1s
Feedback
Was this page helpful?
Glad to hear it! Please tell us how we can improve.
Sorry to hear that. Please tell us how we can improve.