feat: add vm-migration-network setting

[FrankYang0529]

Problem:

Solution:

Related Issue(s):

harvester/harvester#5848

Test plan:

Additional documentation or context

[github-actions]

Name	Link
🔨 Latest commit	4dc3f88
😎 Deploy Preview	https://689476ab87317246c3e866f3--harvester-preview.netlify.app

[innobead]

In general, LGTM. Just a few feedback.

[jillian-maroket]

Review done. Note that the structural changes that I suggested here will be applied to the "Storage Network" page as well. I will create the PR after we merge this one.

[innobead]

Some minor feedback.

[innobead]

You already explained the IP range for the VM migration network, but why is it necessary to mention "the VM migration network must not use"? It’s a little confusing here.

[jillian-maroket]

This is legacy text that was copied from the storage network doc and reorganized a bit. I recently updated the storage network doc. LMK if you want to reuse the wording that what Ivan and I agreed on:

https://docs.harvesterhci.io/v1.6/advanced/storagenetwork#prerequisites

[FrankYang0529]

Remove You must exclude IP addresses that KubeVirt pods and the VM migration network must not use..

[innobead]

The storage network is required for Longhorn pods for data plane (IM and share manager pods), while the VM migration network is needed for each VM/virt-launcher pod. I suggest updating the explanation of the IP range as below. WDYT?

cc @Vicente-Cheng @WebberHuang1118

Suggested change

	Specify an IP range in the IPv4 CIDR format. The number of IP addresses must be larger than or equal to the number of your cluster nodes.
	Specify an IP range in the IPv4 CIDR format. The number of IP addresses must be larger than or equal to the number of VMs.

[FrankYang0529]

Based on my testing, only virt-handler get live-migration IP like:

    k8s.v1.cni.cncf.io/network-status: |-
      [{
          "name": "k8s-pod-network",
          "ips": [
              "10.52.0.93"
          ],
          "default": true,
          "dns": {}
      },{
          "name": "harvester-system/vm-migration-network-g2cr7",
          "interface": "migration0",
          "ips": [
              "10.1.2.2"
          ],
          "mac": "ae:20:fb:20:67:94",
          "dns": {}
      }]

Other kubevirt pods including virt-lanucher, they don't have live-migration IP:

    k8s.v1.cni.cncf.io/network-status: |-
      [{
          "name": "k8s-pod-network",
          "ips": [
              "10.52.1.32"
          ],
          "default": true,
          "dns": {}
      }]

Each node has one virt-handler, I think we can just mention "cluster nodes" here.

[innobead]

Yeah that's fair, the same as storage network.