fix: keep HandleContext alive when a ForwardingResolver throws

[CypherPotato]

Cause

When ForwardingResolver.OnResolveClientAddress throws, HandleContext correctly answers 500, but request.remoteAddr stays null. In the finally block, the default access log format (%ri) reads HttpRequest.RemoteAddress, which calls the resolver again and throws a second time. That exception escapes HandleContext:

• default engine: it runs on a thread-pool work item, so the process terminates;
• Cadente: the connection is closed without a response, and with AsyncRequestProcessing = false the Monitor.Exit is skipped and the server stays locked.

Fix

• %ri uses the already resolved remoteAddr and no longer calls the resolver again.
• Handlers and logging in the finally block are wrapped in try/catch, and the sync completion, Monitor.Exit and request disposal always run.

Test

ThrowingForwardingResolverReturns500AndKeepsServing covers the default and Cadente engines with AsyncRequestProcessing true and false: the resolver throws, the response is 500 and the next request is 200. Without the fix the test host crashes.

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	⚠️ Failed 2026-10-16T18:06:13.256109Z	828b101	PR opened

ℹ️ About Codex in GitHub

Your team has set up Codex to 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" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.