Re: [lwip-members] TCP ACK bug

[Jani Monoses]

Hi
it has been fixed a couple of months ago in CVS :)
to bad your tree is out of sync and we duplicate effort in 
some situations.

Jani.

> 
> Hi
> 
> I have found a situation where LwIP gets stuck. My tree is waaaay
> out of sink, so i have no idea if it has long since been fixed.
> 
> There is a rare situation when you download in one direction
> from a remote non-LwIP stack on a very unreliable link, where
> packets can get reordered, that you get lwip stuck on
> "tcp_receive: duplicate seqno"...
> 
> the problem happens when an ack from lwip has been lost,
> but out-of-ordering makes the remote think its further
> behind than it really is, and retransmits a packet that
> ends up at the line "tcp_receive: duplicate seqno" and
> is processed no further.
> 
> simply adding:
> 
>        else{
>       /* the whole segment is < rcv_nxt */
>       /* must be a duplicate of a packet that has already been
>       correctly handled */
>       
> +     tcp_ack(pcb);
>       DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: duplicate seqno %ld\n",
>       seqno));
>        }
> 
> fixes this.
> 
> -paul
> 
> 
> 
> 
> _______________________________________________
> lwip-members mailing list
> address@hidden
> http://mail.nongnu.org/mailman/listinfo/lwip-members