Re: TCP socket handling errors

> On 3. Apr 2024, at 19:46, Sad Clouds <[email protected]> wrote:
> 
> On Wed, 3 Apr 2024 17:28:52 +0200
> Michael Tuexen <[email protected]> wrote:
> 
>>> On 3. Apr 2024, at 15:44, Sad Clouds <[email protected]> wrote:
>>> 
>>> I found a bug that is still open from May 2010 and describes the same
>>> behaviour that I see with my application:
>>> 
>>> https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=146845
>>> 
>>> If this hasn't been fixed over the last 14 years, then I guess I will
>>> add some code to simply ignore ECONNRESET on close(2) for FreeBSD and
>>> MacOS. This seems the be the general advice from other people who hit
>>> this issue.
>> I'll bring this up on the bi-weekly FreeBSD transport call.
>> 
>> Best regards
>> Michael
>>> 
>> 
> 
> Hello, I've attached a test program, this easily reproduces the issue
> on Raspberry Pi 4 within a few seconds of running it.
> 
> Server output:
> $ ./econnreset server
> Server: accept()
> Server: shutdown()
> Server: close()
> ...
> Server: accept()
> Server: shutdown()
> Server: close()
> close() failed, error=Connection reset by peer
> 
> 
> Client output (aborts when server exists due to close() failure):
> $ while true; do ./econnreset client || break; done
> Client: connect()
> Client: shutdown()
> Client: close()
> ...
> Client: connect()
> Assertion failed: (int_val == 0), function client, file econnreset.c,
> line 156. Abort trap (core dumped)
> <econnreset.c>
I can reproduce this locally.

Best regards
Michael