[pfSense Support] Strange TCP connection behavior 2.0 RC2 (+3)
cb.list6 at gmail.com
Wed Jun 29 14:34:26 UTC 2011
On Jun 28, 2011 10:22 PM, "Mikael Abrahamsson" <swmike at swm.pp.se> wrote:
> On Tue, 28 Jun 2011, Cameron Byrne wrote:
>> My point was that if end-hosts had Hybla or something similar, these
proxies can be removed providing a better end-to-end solution.
> Well, then you run into the nice problem of the RNCs only having 400
kilobytes of buffers per session and will drop packets if they receive more
packets than that, or sometimes even just because they receive a burst of a
few tens of packets at GigE linerate (because the customer with large TCP
window size is talking to a GigE connected server).
> The recommended "solution" from the vendor was to tune the client to a
smaller TCP window size so that their RNC wouldn't get such a large burst.
Sounds like a vendor specific issue :(
Good tcp vs default tcp will not close the window tight due to some
ephemeral loss or delay. The penalties are generally too strong in tcp for
the issues of delay and loss in 3g ... this is one of the main selling
points for tcp proxies .... but better done with modern tcp on the clients
instead of a middle box
> Mikael Abrahamsson email: swmike at swm.pp.se
More information about the NANOG