Skip to content

Conversation

@michael-go
Copy link

On MacOS, with listen(1), when starting more than ~3 concurrent clients, new clients often get ECONNREFUSED as it can take few ms for the server to accept() each connection. By not setting the backlog in listen(), it seems to handle even 1000 concurrent connections just fine.
Tested also on Linux, where it behaves the same with and without the param.

On MacOS, with listen(1), when starting more than ~3 concurrent clients,
new clients often get ECONNREFUSED as it can take few ms for the server
to accept() each connection. By not setting the backlog in listen(),
it seems to handle even 1000 concurrent connections just fine.
Tested also on Linux, where it behaves the same with and without the
param.
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