RE: [lwip-users] Simultaneous TCP client and server on same platform

[Muhamad Ikhwan Ismail]

I am using the sockets API, which means I am also using the callback API rite ?
The problem doesn't occur if the server runs as a standalone, without the TCP client.
I am tracing where the reset comes from using breakpoint and the link register. Ill post it once I get the source..

---

> Subject: RE: [lwip-users] Simultaneous TCP client and server on same platform
> From: address@hidden
> To: address@hidden
> Date: Mon, 21 Jan 2008 12:58:09 +0000
>
> On Mon, 2008-01-21 at 12:13 +0000, Kieran Mansley wrote:
> > On Mon, 2008-01-21 at 11:45 +0000, Muhamad Ikhwan Ismail wrote:
> > >
> > > I am quite sure that is not the case, since I dont have a strict
> > > memory restriction
> > > I have allocated each kind of pcb with 40 and according to lwip stats
> > > I only used
> > > 3, 1 UDP and 2 TCP.
> > >
> >
> > Try turning on the LWIP_DEBUGF statements. You should then get a reason
> > for the reset in the form or a print statement.
> >
> > The most likely one from what you describe is the reset sent at line 628
> > of file lwip/src/core/tcp_in.c. This is triggered if the ACK in the SYN
> > handshake has the wrong ACK number, but from the trace you provide that
> > doesn't seem to be the case.
>
> I missed a more likely explanation: if the accept callback function
> returns an error the connection is aborted, which sends a RST. Which
> API are you using? Have you requested a callback when a connection is
> accepted?
>
> Kieran
>
>
>
> _______________________________________________
> lwip-users mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/lwip-users

---

Climb to the top of the charts! Play the word scramble challenge with star power. Play now!