Fix: inventory builder overwrites the already existing nodes specified in the host.yaml

[VladMasarik]

Environment:

• Version of Python (python --version):
• Python 3.8.5

Kubespray version (commit) (git rev-parse --short HEAD):
5ea2d1e

Expected behavior:
With the following file I execute

KUBE_MASTERS_MASTERS=1 CONFIG_FILE=inventory/mycluster/hosts.yaml python contrib/inventory_builder/inventory.py 10.0.0.8

hosts.yaml file. (I think this should show that there is a bug as well, but I accidentally deleted my previous file which worked. In case you need it, I can try replicating it.)

all:
  hosts:
    main:
      ansible_host: 40.114.118.34
      ip: 10.0.1.6
      access_ip: 40.114.118.34
    node1:
      ansible_host: 10.0.0.9
      ip: 10.0.0.9
      access_ip: 10.0.0.9
  children:
    kube-master:
      hosts:
        main:
    kube_node:
      hosts:
        main:
        node1:
    etcd:
      hosts:
        main:
    k8s_cluster:
      children:
        kube_control_plane:
        kube_node:
    calico_rr:
      hosts: {}
    kube_control_plane:
      hosts:
        main:
        node1:

I expect the inventory.py script to add new entry while keeping the old ones:

all:
  hosts:
    main:
      ansible_host: 40.114.118.34
      ip: 10.0.1.6
      access_ip: 40.114.118.34
    node1:
      ansible_host: 10.0.0.9
      ip: 10.0.0.9
      access_ip: 10.0.0.9
    node2:
      ansible_host: 10.0.0.8
      ip: 10.0.0.8
      access_ip: 10.0.0.8

Actual behavior:
All hosts are deleted, except for the newly specified one:

all:
  hosts:
    node1:
      ansible_host: 10.0.0.8
      ip: 10.0.0.8
      access_ip: 10.0.0.8

I believe I have also found the bug+fix, but I got no response on Slack yet. When I wanted to submit a PR you mentioned I should create an issue, and I also think I found other bugs, with which I am not sure how to deal with. So my questions are:

1. Is this an actual issue or I am just doing something wrong?
2. How should I proceed if I think I have a fix, but I encountered other bugs, with which I am not sure how to deal with?

[k8s-triage-robot]

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-contributor-experience at kubernetes/community.
/lifecycle stale

[VladMasarik]

/remove-lifecycle stale
PR exists, but it is not merged in yet