Re: [lwip-users] Zero window and refused data problem

[Simon Goldschmidt]

Valery Ushakov wrote:
> I don't want to argue this for the general case but for the specific
> zero window case this argument doesn't apply, I think, and quite
> obviously (IMHO) so.

No, of course it doesn't. And I'm sorry if it sounded like that. I wasn't
thinking about the zero window case when writing that but the general case
where this would hurt. Looks like the zero-window would have to behave
differently here...

> [..] Alternatively we can accept the probe as refused data, but then we
> should relax the logic that decides to not send any ACKs when refused
> data is present and do send the ACK if it also announces zero window.
> In this case we can expect the next probe soon, that we will have to
> reject and keep acking the first one.

That would be OK for me.

> [..] I don't think we should worry about the reaction of "all remote
> stacks" to this as anything is better than actively causing the
> connection to get aborted by not responding.

I think we should. In lwIP we're often faced with problems that protocol
developers haven't thought about because we can be very resource limited.
Again, I'm not talking about the zero window case but about the "refused_data"
case in general. You'd have to be careful not producing endless data/ACK
loops when just sending an old ackno again, since that might trigger
retransmission in remote stacks...

But that doesn't mean I'nm not open to changes in this area, only it's
not as easy as the OP wanted it to fix.

As always, this is probably best kept in a bug report, so that we can
discuss possible fixes/patches there.


Simon