Document stance on static IP addresses

[ncdc]

User Story

As a user/operator, I would like to understand Cluster API's stance on static IP addresses so that I can understand if it will affect my ability to use Cluster API in my environment.

Detailed Description

We should document that individual infrastructure providers may, at their discretion, add support for static IP address assignment at the individual machine level, but, when you move up to higher-level abstractions such as MachineSet, MachineDeployment, KubeadmControlPlane, there is no support for customizing each replica. If someone wanted to use static IPs with these abstractions, there would need to be a coordinated effort between the infrastructure provider and some sort of IPAM controller (outside the scope of Cluster API itself).

/kind documentation
/priority important-soon
/milestone v0.3.0

cc @akutz

[akutz]

While it may be implied/inferred, it also (I think) bears explicit mention in such documentation that the upgrade capabilities for MachineDeployments and the upcoming control plane upgrades won't work without DHCP or some sort of IPAM (as @ncdc mentioned above).

[ncdc]

cc @rosskukulinski

[sheriefhammam]

A relevant observation
I haven't debugged it yet but for some reason i can assign static IPs to worker nodes but it affects somehow the control plane

apiVersion: infrastructure.cluster.x-k8s.io/v1alpha2
kind: VSphereMachineTemplate
metadata:
  name: workload-cluster-02-md-0
  namespace: default
spec:
  template:
    spec:
      datacenter: NIC-Darmstadt
      diskGiB: 20
      memoryMiB: 2048
      network:
        devices:
        - dhcp4: false
          dhcp6: false
          gateway4: 10.100.146.1
          ipAddrs:
          - 10.100.146.50/24
          networkName: ds146
      numCPUs: 2
      template: T_centos-7-kube-v1.16.3.a4

kubectl get nodes -o wide
NAME                                        STATUS   ROLES    AGE   VERSION   INTERNAL-IP     EXTERNAL-IP     OS-IMAGE                KERNEL-VERSION               CONTAINER-RUNTIME
workload-cluster-02-controlplane-0          Ready    master   36m   v1.16.3   10.100.146.50   10.100.146.50   CentOS Linux 7 (Core)   3.10.0-1062.9.1.el7.x86_64   containerd://1.3.0
workload-cluster-02-md-0-5b7df745f6-bgs2z   Ready    <none>   35m   v1.16.3   10.100.146.84   10.100.146.84   CentOS Linux 7 (Core)   3.10.0-1062.9.1.el7.x86_64   containerd://1.3.0

[vincepri]

/help
/milestone Next

[k8s-ci-robot]

@vincepri:
This request has been marked as needing help from a contributor.

Please ensure the request meets the requirements listed here.

If this request no longer meets these requirements, the label can be removed
by commenting with the /remove-help command.

In response to this:

/help
/milestone Next

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[luniHw]

Issue:
Used capv 0.4.2 and centos-7-kube-v1.15.6 to create centos k8s cluster on nimbus env, it failed to assign static ip, a dynamical ip is assigned to vm. (Nimbus env, some ips are dhcp, some ips are static ips)

Help:
I need to upgrade capv the version or do others? thanks

[MnrGreg]

@luniHw static IP assignment in CAPV is fixed with: kubernetes-sigs/cluster-api-provider-vsphere#707 but there hasn't been a CAPV release since.

@yastij any idea on timing for the next release?

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[vincepri]

/remove-lifecycle stale

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[fabriziopandini]

/remove-lifecycle stale

[vincepri]

/lifecycle frozen

[vincepri]

/close

Cluster API doesn't do ip management today, this is left for infrastructure providers

[k8s-ci-robot]

@vincepri: Closing this issue.

In response to this:

/close

Cluster API doesn't do ip management today, this is left for infrastructure providers

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[jessehu]

update: the IPAM Proposal is added in CAPI 1.2.0:
#6000
#6313