fix(iroh): fix very slow initial connection establishment

[rklaehn]

Description

Fixes very slow (~1s) initial connection establishment

Also adds a test for initial connection delay.

Initial connection was very slow even with relays disabled and static discovery. There was a fixed delay.

Breaking Changes

None

Notes & open questions

Question: not sure how we can turn this into a proper test without making it super flaky, since it is a timing test. Maybe measure the ratio between first and second round and have some very conservative bounds?

Changed the test name and put the timings in the assertion. I think the test is good to go as is. The ratio is extremely conservative, so it should rarely flake out.

Note: I do 32 connections as an attempt to make the timings more reliable, but strictly speaking it would also work with 1 connection.

thread 'endpoint::tests::initial_connection_time' panicked at iroh/src/endpoint.rs:3269:9:
First round: 32.230450542s, second round 0.063144708s

Change checklist

• Self-review.
• Documentation updates following the style guide, if relevant.
• Tests if relevant.
• All breaking changes documented.
  • List all breaking changes in the above "Breaking Changes" section.
  • Open an issue or PR on any number0 repos that are affected by this breaking change. Give guidance on how the updates should be handled or do the actual updates themselves. The major ones are:
    • quic-rpc
    • iroh-gossip
    • iroh-blobs
    • dumbpipe
    • sendme

[github-actions]

Documentation for this PR has been generated and is available at: https://n0-computer.github.io/iroh/pr/3434/docs/iroh/

Last updated: 2025-08-14T12:18:56Z

[github-actions]

Netsim report & logs for this PR have been generated and is available at: LOGS
This report will remain available for 3 days.

Last updated for commit: b0d9c28

[matheus23]

I'm not super happy with the code duplication, I think it's probably possible to factor out common code for constructing a Transmit from send_ping_actions and try_send_ping_actions, but it's fine, especially since this code won't live that long.