On Fri, Oct 17, 2014 at 7:31 AM, Yann Ylavic <[email protected]> wrote:

> The pcap does not show the window scaling factor (SYN SYN/ACK not
> captured).
> Without this factor, the client's receive window is of 65, which would
> need a scaling factor of at least 9 (which is not the default AFAICT
> in linux) to fit the 29216 bytes (ie. 53359 - 24143) in flight at the
> time of the frame 20.
>
> Can't this be a *current* window issue (as opposed to the *max*
> window), can you still reproduce with, eg :
>    sysctl -w
> ​​
> net.ipv4.tcp_rmem="49152 49152 4194304"
> on the client side (or anything that makes the client announce a
> window > 30K from the very beginning)?
>

​Thanks for having a look -- of course the one capture I actually posted
was not running when the connection was established.  I have overlayed with
a complete one on people.a.o

The advertised scaling factor / effective window size are extremely large​
by the end of these runs, 1MB.  The client has:

net.ipv4.tcp_rmem = 2129920     2129920 882129920

One interesting part that might not have came through -- my file was
originally 130k, then 120k, and now only about 30k.  It's always only the
one last"partial" frame that is delayed by 1 RTT.



-- 
Eric Covener
[email protected]

Reply via email to