Re: TCP_DEFER_ACCEPT is missing counter update

[Eric Dumazet <eric.dumazet@gmail.com>]

Willy Tarreau a écrit :
> On Fri, Oct 16, 2009 at 07:00:49AM +0200, Eric Dumazet wrote:
>> Eric Dumazet a écrit :
>>>
>>> So, it appears defer_accept value is not an inherited attribute,
>>> but shared by all embryons. Therefore we should not touch it.
>>>
>>> Of course it should be done, or add a new connection field to count number
>>> of pure ACKS received on each SYN_RECV embryon.
>>>
>> Could be something like this ? (on top of net-next-2.6 of course)
>>
>> 7 bits is more than enough, we could take 5 bits IMHO.
> 
> Couldn't we just rely on the retrans vs rskq_defer_accept comparison ?
> 

In this case, we lose TCP_DEFER_ACCEPT advantage in case one SYN-ACK was dropped
by the network : We wakeup the listening server when first ACK comes from client,
instead of really wait the request.

I think being able to count pure-acks would be slighly better, and cost nothing.


retrans is the number of SYN-RECV (re)sent, while req_acks would count number of
pure ACK received.

Those numbers, in an ideal world should be related, but could differ in real world ?