fix(test): retry loopback accept in stale-session poll tests

[cursor]

Summary

macOS Cross-Platform CI on PR #304 failed in server::tests::paused_players_are_closed_during_poll because TcpStream::connect() can return before the listener's non-blocking accept() sees the socket.

Changes

• Use the existing accept_pending_connections helper in stale-session poll teardown tests
• Apply the same retry pattern to related max-connection and per-IP cap accept tests

Test plan

• Cross-Platform cargo test (macOS, all features) passes

  

---

Note

Low Risk
Only test harness timing changes; no production server or accept-path behavior is modified.

Overview
Test-only hardening for loopback TCP acceptance races (especially on macOS), where TcpStream::connect() can succeed before the server's non-blocking accept() sees the socket.

Server tests that set up connections now call accept_pending_connections instead of a single accept_new_connections() plus an immediate length assert. That helper retries accepts with a short sleep until the expected connection count is reached or a 2s deadline.

The max global connections and per-IP cap tests use the same retry loop when asserting a third client stays rejected, so flaky CI from timing is avoided without changing server admission logic.

^{Reviewed by Cursor Bugbot for commit 008ceb8. Bugbot is set up for automated code reviews on this repo. Configure here.}

[coderabbitai]

Important

Review skipped

Bot user detected.

To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Advanced

Run ID: ca9b2d59-e166-4f10-8617-3ace6752ce6c

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[cursor]

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: serverGenReqId_15baf00f-e2ea-4cca-940e-2e630da6e0d8)