Closed skeeey closed 3 years ago
[APPROVALNOTIFIER] This PR is APPROVED
This pull-request has been approved by: skeeey
The full list of commands accepted by this bot can be found here.
The pull request process is described here
/assign @qiujian16
why do we separate this to another controller? It will double the update call, right?
I think separate this make the code logic more clear, currently, we maintain this in the cluster joining controller, but collecting the node resource and handling cluster join event may be two irrelevant tasks
when the cluster starts to register, this will add an update call, but if we use cluster Available
as the judgement condition, I think we can stagger this update from others
what's your opinion? do you think we should keep this in the joining controller?
I think we can move it as part of health check (available) controller
agree with you, and I think we can also put the get version in health check controller, thus the joining controller will only handle the cluster joining
/lgtm
Signed-off-by: liuwei liuweixa@redhat.com