fix not programmed listeners for merge gateways#8744
fix not programmed listeners for merge gateways#8744cnvergence wants to merge 5 commits intoenvoyproxy:mainfrom
Conversation
Signed-off-by: Karol Szwaj <karol.szwaj@gmail.com>
✅ Deploy Preview for cerulean-figolla-1f9435 ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
Signed-off-by: Karol Szwaj <karol.szwaj@gmail.com>
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #8744 +/- ##
==========================================
+ Coverage 74.39% 74.44% +0.04%
==========================================
Files 246 246
Lines 39221 39223 +2
==========================================
+ Hits 29180 29199 +19
+ Misses 8017 8003 -14
+ Partials 2024 2021 -3 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: be40f08593
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
| if t.MergeGateways && !listener.IsReady() { | ||
| continue |
There was a problem hiding this comment.
Exclude non-ready listeners from TLS overlap checks
In merge-gateways mode this new continue drops non-programmed listeners from xDS generation, but checkOverlappingTLSConfig still iterates all HTTPS listeners from gateway.listeners (including non-ready ones) and can set TLSOverlaps on ready listeners. That causes internal/xds/translator/listener.go to force ALPN to http/1.1 for a listener that no longer has any overlapping peer in xDS, so valid HTTPS listeners can lose HTTP/2 unnecessarily whenever an overlapping listener is invalid/unprogrammed.
Useful? React with 👍 / 👎.
Signed-off-by: Karol Szwaj <karol.szwaj@gmail.com>
What type of PR is this?
What this PR does / why we need it:
Which issue(s) this PR fixes:
Fixes #
Release Notes: Yes/No