On Thu, 2003-05-29 at 23:54, David S. Miller wrote:
> From: Martin Josefsson <gandalf@xxxxxxxxxxxxxx>
> Date: 29 May 2003 21:57:11 +0200
>
> I've just tested 2.5-bk and I can still reproduce it.
>
> Can you get a tcpdump trace at the sender side for the complete life
> of the connection for which the hang occurs?
>
> There isn't much we can do about this without a trace.
I've already sent the trace twice to netdev, I've attached it again.
I can reproduce it quite easily, both with TCP_CORK and without.
--
/Martin
stall-57798-anon
Description: Text document
|