Description

Fix the e2e external connectivity check so it validates outbound connectivity instead of requiring a specific HTTP status from a single website.

Why do we need it, and what problem does it solve?

The previous CheckExternalConnectivity implementation treated external connectivity as a successful curl response with HTTP 200 from one hardcoded host. This made the test flaky and semantically incorrect: the VM could still have outbound network connectivity while the selected website was unavailable, redirected, or returned a non-200 status.

This change makes the check align with the actual test intent. The VM is now considered externally connected if it can successfully reach at least one host from a fallback list within bounded timeouts.

What is the expected result?

1. Run e2e tests that call CheckExternalConnectivity.
2. Ensure the check succeeds when the VM has outbound access and at least one fallback host is reachable.
3. Ensure the check no longer fails only because a single website returns a non-200 HTTP status.

Checklist

• The code is covered by unit tests.
• e2e tests passed.
• Documentation updated according to the changes.
• Changes were tested in the Kubernetes cluster manually.

Changelog entries

section: test
type: fix
summary: "Make the e2e external connectivity check rely on successful outbound access to fallback hosts instead of HTTP 200 from a single site."
impact_level: low