A write to a TCP-stream overtook another write

public inbox for io-uring@vger.kernel.org
 help / color / mirror / Atom feed

* A write to a TCP-stream overtook another write
@ 2025-04-30 15:33 Andreas Wagner
  2025-04-30 18:47 ` Pavel Begunkov
  0 siblings, 1 reply; 2+ messages in thread
From: Andreas Wagner @ 2025-04-30 15:33 UTC (permalink / raw)
  To: io-uring

Dear everyone,

during development of an HTTP-server, I ran into a problem: A write to
a TCP-stream overtook another write to the same stream.
Currently, I have a broken-pipe-issue which I'd like to investigate
before sharing the sources.

Regards,
Andreas Wagner

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: A write to a TCP-stream overtook another write
  2025-04-30 15:33 A write to a TCP-stream overtook another write Andreas Wagner
@ 2025-04-30 18:47 ` Pavel Begunkov
  0 siblings, 0 replies; 2+ messages in thread
From: Pavel Begunkov @ 2025-04-30 18:47 UTC (permalink / raw)
  To: Andreas Wagner, io-uring

On 4/30/25 16:33, Andreas Wagner wrote:
> Dear everyone,
> 
> during development of an HTTP-server, I ran into a problem: A write to
> a TCP-stream overtook another write to the same stream.

Requests can be reordered in many ways, you can't submit two requests
to a tcp socket and expect that they'll get executed in order. It was
mentioned many times before, for example, look through this thread:

https://github.com/axboe/liburing/issues/1359

-- 
Pavel Begunkov


^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2025-04-30 18:46 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-04-30 15:33 A write to a TCP-stream overtook another write Andreas Wagner
2025-04-30 18:47 ` Pavel Begunkov

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox