Skip to content

Conversation

johnhaddon
Copy link
Member

In c357ea1, we switched from making an IPv4 connection, to an IPv6 one. This fixed some nasty lag on Windows, but it turns out that it also broke IPv4-only environments completely. So now we fall back to opening a v4 connection if the v6 connection fails.

@ericmehl, if you could check that the lag is still banished on Windows, that would be great. I think it should be fine because we open the v6 connection first still, but I'm a bit out of my comfort zone with this, so the more testing the better.

In c357ea1, we switched from making an IPv4 connection, to an IPv6 one. This fixed some nasty lag on Windows, but it turns out that it also broke IPv4-only environments completely. So now we fall back to opening a v4 connection if the v6 connection fails.
@johnhaddon johnhaddon requested a review from ericmehl September 26, 2025 10:21
@johnhaddon johnhaddon self-assigned this Sep 26, 2025
@johnhaddon
Copy link
Member Author

Hmm, Windows CI is telling me I definitely messed something up.

The previous commit was failing on Windows. Maybe we can't call `local_endpoint()` before `bind()`?
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant