Automatic merge from submit-queue
Allow apiserver to choose preferred kubelet address type
Follow up to #33718 to stay compatible with clusters using DNS names for master->node communications. Adds the `--kubelet-preferred-address-types` apiserver flag for clusters that prefer a different node address type.
```release-note
The apiserver can now select which type of kubelet-reported address to use for master->node communications, using the --kubelet-preferred-address-types flag.
```
Automatic merge from submit-queue
Improve CockroachDB example
This is primarily about pulling in the init container to make the config more production-ready, but I've pulled in a few other small improvements that have been made since this was first contributed.
* Use an init container to eliminate potential edge case where losing
the first pet's data could cause it to start a second logical cluster
* Exec the cockroach binary so that it runs as PID 1 in the container
* Make some small improvements to the README
@bprashanth
```release-note
```
Automatic merge from submit-queue
Disable update-staging-client-go until code freeze
I don't want to cause developer friction until code freeze. I'll babysit the script for the moment to keep the client-go up-to-date.
Should had done this in #34489.
Automatic merge from submit-queue
quota controller uses informers if available for pod calculation
This PR does the following:
1. plumb informer factory into quota registry and evaluators
2. pod quota evaluator uses informers for determining aggregrate usage instead of making direct calls
3. admission code path does not use informers because
1. we do not want to add new watches in apiserver
2. admission code path does not require aggregate usage calculation
As a result, quota controller is much faster in re-calculating quota usage when it observes a pod deletion.
Follow-on PRs will make similar changes for other informer backed resources (pvcs next).
/cc @deads2k @mfojtik @smarterclayton @kubernetes/rh-cluster-infra
* Use an init container to eliminate potential edge case where losing
the first pet's could cause it to start a second logical cluster
* Exec the cockroach binary so that it runs as PID 1 in the container
* Make some small improvements to the README
This surfaces the token generation logic so that users can first
generate and store a token, then pass it to kubeadm init/join.
Otherwise, users have to capture and parse the output of "kubeadm init"
to feed the token to "kubeadm join."
Automatic merge from submit-queue
Initialize CIDR allocator before registering handle functions
Currently we start shared informers after everything is already created, but this change make it future-proof.
cc @davidopp @kevin-wangzefeng @foxish
Automatic merge from submit-queue
update list of vailable resources
Hi,
kubectl get --help produce a list of resource types and aliases :
```
Valid resource types include:
* clusters (valid only for federation apiservers)
* componentstatuses (aka 'cs')
...
```
``` release-note
Update the list of resources in kubectl get --help
```
The list is currently outdated (for exemple missing networkpolicy).
http://kubernetes.io/docs/user-guide/kubectl-overview/#resource-types has the same data and is also outdated.
The patch updates these 2 lists.
