Can someone explain how kubectl rollout pause/resume works under the hood? I get how to use it, but I don't understand how to replicate the CLI functionality using the API, or how the controller was made to respect it if it's not a Deployment spec change.
-
Show this thread
-
Can't find a KEP or proposal. Code is very abstract. /cc
@bgrant0607@xkarga00 for involvement with PRs https://github.com/kubernetes/kubernetes/pull/20087 … https://github.com/kubernetes/kubernetes/pull/19882 …1 reply 0 retweets 1 likeShow this thread -
Cli Rollout as far as i remember from old reading of code , is not based on api(not rolling update of deployment) . It was replica set management on the client side . Not sure if that has changed recently
1 reply 0 retweets 0 likes -
The rollout command was co-developed with the Deployment API. It was rolling-update that predates kubectl. Pause and resume patch the paused API field. https://github.com/kubernetes/kubernetes/blob/5b0d8d2ce118d44731c9eb4c52b285d676424e01/staging/src/k8s.io/api/apps/v1/types.go#L303 …
2 replies 0 retweets 0 likes -
ah thanks, i was thinking more about "kubectl rolling-update" command , i could not find an official documentation but this https://jamesdefabia.github.io/docs/user-guide/kubectl/kubectl_rolling-update/ … . wonder why we never have pause for daemonset
1 reply 0 retweets 0 likes
It is documented on the deployment page: https://kubernetes.io/docs/concepts/workloads/controllers/deployment/ …. I don't think I have ever seen usage data for the feature.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.