keep Gateway's AttachedRoutes field up to date

[mlavacca]

What this PR does / why we need it:

We need to update and keep up to date the Gateway's AttachedRoutes field, with the actual number of routes accepted by it.

Which issue this PR fixes:

Fixes #3120

Special notes for your reviewer:

PR Readiness Checklist:

Complete these before marking the PR as ready to review:

• the CHANGELOG.md release notes have been updated to reflect any significant (and particularly user-facing) changes introduced by this PR

[codecov]

Codecov Report

Patch coverage: 55.6% and no project coverage change.

Comparison is base (753e91f) 60.0% compared to head (55f96c7) 60.1%.

Additional details and impacted files

@@          Coverage Diff           @@
##            main   #4052    +/-   ##
======================================
  Coverage   60.0%   60.1%            
======================================
  Files        151     151            
  Lines      16612   16735   +123     
======================================
+ Hits        9982   10066    +84     
- Misses      5999    6035    +36     
- Partials     631     634     +3     

Impacted Files	Coverage Δ
internal/controllers/gateway/gateway_controller.go	12.6% <13.0%> (+<0.1%)	⬆️
internal/controllers/gateway/gateway_utils.go	44.8% <58.4%> (+1.7%)	⬆️
internal/controllers/gateway/route_utils.go	76.8% <75.8%> (+1.8%)	⬆️
internal/util/builder/allowedroutes.go	66.6% <100.0%> (+16.6%)	⬆️

... and 3 files with indirect coverage changes

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

No remaining blockers.

[shaneutt]

Once the linter is happy and @pmalek's comments are resolved, LGTM

[gjkim42]

I found that when we create more than 101 httproutes, the kong-ingress-controller cannot determine the attachedRoutes field correctly and the value keeps fluctuating.
It ends up making a huge amount of requests to kube-apiserver to update its fluctuating attachedRoutes continually.

Any idea in here?

@mlavacca

[mlavacca]

Hi, @gjkim42, thanks for reporting this issue! does this behavior happen consistently at 101? What I mean is, can you confirm it does not happen with 100 routes, but it does with 101? Furthermore, do you have logs with throttling from the API server? It shouldn't happen, as the local cache should be used to retrieve Routes and Gateways 🤔

[gjkim42]

I filed the bug and it has been addressed.

Please see #4456 (comment) here.

[mlavacca]

Cool, thanks!