phlax

heres a more helpful failure on the packaging pr https://dev.azure.com/cncf/envoy/_build/results?buildId=82041&view=logs&j=7d299c4a-9a35-59b9-677d-f6996e8fc292&t=2e3bbaf1-9c4a-5c1a-ee33-a8cc5407b4d9&l=5227 you can see the logs and response there

i added a loopback, so there is a proxy in this example now

@mattklein123 @alyssawilk i tested again today to see whether any changes at netlify had resolved the problem that this addresses, unfortunately it doesnt seem to have: ```console PackagesDistroChecker ERROR [distros]...

ci should be fixed

@alyssawilk @mattklein123 this is no longer required for my current work as the netlify issues appears to be fixed - so im happy to close that said, i quite like...

> but that wasn't the bug with netlify because it was happening with Chrome and an https url which wouldn't be doing x-forward-proto: http. i believe due to caching it...

> alternately if you want best of both worlds, we could try to hit up envoyproxy.io and have a custom error rewrite handler which notes lack of network connectivity and...

this is still an issue as for CI (and i guess end users) as the website can still be flakey

> we could have default config be to proxy and retry config serve from cache? i guess that would be possible - iiuc then adding adding a warm cache to...

@soulxu as mentioned previously should this not be raised against `main` first ?