Why TCP connections must be handled concurrently #718
Replies: 2 comments 1 reply
-
|
This is a issue about accepted socket missing. Lines 363 to 376 in e359e5e I have fixed it in my fork. |
Beta Was this translation helpful? Give feedback.
-
|
Works for me. Does making requests with curl instead of chrome work? Note that you don't read the request, so who is to say there's someone on the other end reading? One gotcha with chrome is that it often makes multiple simultaneous connections to the same domain for performance reasons, without necessarily reading or writing data on those extra connections (only if there are more resources to fetch, like css or script files.) |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
From #715, I still didn't get why TCP connections must be handled concurrently.
As the code telling, the server will tell Chrome "don't keep tcp connection alive" and close the connection after data sent, server should be ready for next connection. But this simple HTTP server will stcuk at
server.acceptafter accepted a connection from Chrome.I write a (maybe just looks like) same thing in Python3, and it just working fine.
I do not know much about network things, is there anything I missing out?
Beta Was this translation helpful? Give feedback.
All reactions