fix: add retry logic to vpc-egress-gateway e2e curl check

[oilbeater]

Summary

• Replace RunHostCmdOrDie with WaitUntil + RunHostCmd in checkEgressAccess() to add retry logic (2s interval, 30s timeout) for the curl connectivity check
• Fixes flaky e2e test "should be able to create vpc-egress-gateway with underlay subnet" which intermittently fails with curl exit code 28 (connection timeout)
• Root cause: VEG Status.Ready only indicates control plane readiness; OVN dataplane convergence (NB→SB propagation, ovn-controller flow installation, BFD session establishment) is asynchronous and may take several seconds

Test plan

• Run vpc-egress-gateway e2e tests multiple times to verify flakiness is resolved
• Verify the retry pattern matches existing CheckPodEgressRoutes usage in the same test suite

🤖 Generated with Claude Code

[gemini-code-assist]

Summary of Changes

Hello @oilbeater, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request enhances the robustness of VPC Egress Gateway end-to-end tests by introducing a retry mechanism for network connectivity checks. This change specifically targets a flaky test that intermittently failed due to the asynchronous nature of network dataplane convergence, ensuring that tests wait for network readiness before asserting connectivity.

Highlights

• E2E Test Stability: Implemented retry logic for curl connectivity checks within the checkEgressAccess() function to mitigate intermittent failures in VPC Egress Gateway e2e tests.
• Retry Mechanism: Replaced RunHostCmdOrDie with a WaitUntil loop wrapping RunHostCmd, allowing for retries every 2 seconds with a total timeout of 30 seconds.
• Root Cause Addressed: Addressed test flakiness caused by asynchronous OVN dataplane convergence, which can lead to temporary network unavailability even after the control plane indicates readiness.

Changelog

• test/e2e/vpc-egress-gateway/e2e_test.go
  • Replaced a direct curl command execution with a WaitUntil retry loop to handle transient network issues.
  • Updated error handling for RunHostCmd within the retry block.

Activity

• The author plans to run vpc-egress-gateway e2e tests multiple times to verify flakiness is resolved.
• The author plans to verify the retry pattern matches existing CheckPodEgressRoutes usage in the same test suite.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request introduces retry logic to an e2e test to address flakiness, which is a solid improvement. The use of WaitUntil is appropriate for handling the asynchronous nature of network convergence. My review includes a suggestion to enhance debuggability by adding logging within the retry loop. I've also noted a point of confusion in the WaitUntil helper function that could be addressed in a future refactoring to improve the test framework's clarity.

[gemini-code-assist]

This retry logic is a good improvement for the flaky test. I have two suggestions for further improvement:

1. WaitUntil function signature: The framework.WaitUntil function is a bit misleading. It accepts an interval as its first argument, but this argument is ignored, and a hardcoded 2-second interval is used instead. Your call works because you've passed 2*time.Second, matching the hardcoded value. This is worth noting as it has caused issues elsewhere (e.g., CheckPodEgressRoutes passes a 3-second interval that is ignored). A follow-up to refactor WaitUntil to use the interval parameter would improve the framework's clarity.

2. Debugging: To make this test easier to debug if it times out, consider adding logging within the retry loop to show why a retry is happening. For example:

   func(_ context.Context) (bool, error) {
       output, err := e2epodoutput.RunHostCmd(pod.Namespace, pod.Name, cmd)
       if err != nil {
           framework.Logf("curl command failed, will retry: %v", err)
           return false, nil
       }
       clientIP, _, err = net.SplitHostPort(strings.TrimSpace(output))
       if err != nil {
           framework.Logf("failed to parse curl output, will retry. output: %q, error: %v", output, err)
       }
       return err == nil, nil
   }