Resolve control plane endpoint when updating service

[lubronzhan]

What this PR does / why we need it:

Which issue(s) this PR fixes:

Fixes ##178

Describe testing done for PR:

Special notes for your reviewer:

Release note:

New PR Checklist

• Ensure PR contains only public links or terms
• Use good commit messages
• Squash the commits in this branch before merge to preserve our git history
• If this PR is just an idea or POC, use a Draft PR instead of a full PR
• Add appropriate labels according to what type of issue is being addressed.

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 73.68421% with 5 lines in your changes are missing coverage. Please review.

Project coverage is 36.54%. Comparing base (a70caef) to head (9be9129).

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #180      +/-   ##
==========================================
+ Coverage   36.34%   36.54%   +0.19%     
==========================================
  Files          20       20              
  Lines        2724     2731       +7     
==========================================
+ Hits          990      998       +8     
  Misses       1675     1675              
+ Partials       59       58       -1     

Files	Coverage Δ
pkg/ako-operator/lib.go	100.00% <ø> (ø)
pkg/haprovider/haprovider.go	68.96% <73.68%> (+0.93%)	⬆️

[flawedmatrix]

For the most part, looks good. Just a question though, does the DNS resolver support IPv6?

[lubronzhan]

For the most part, looks good. Just a question though, does the DNS resolver support IPv6?

yes it does

// LookupIP looks up host using the local resolver.
// It returns a slice of that host's IPv4 and IPv6 addresses.

Once we support ipv6 as control plane HA, when AKO supports it, we could add more tests

[flawedmatrix]

Yeah that makes sense. For some reason I thought this was part of the change

load-balancer-operator-for-kubernetes/pkg/haprovider/haprovider.go

Line 459 in a70caef

return ips[0].String(), err

and I wasn't sure which IP family this returned.

But it seems like existing behavior so this PR looks good to me.