Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 12 Dec 2014 10:35:10 -0700
From:      John Nielsen <lists@jnielsen.net>
To:        Brett Glass <brett@lariat.net>
Cc:        Luigi Rizzo <rizzo@iet.unipi.it>, "freebsd-net@freebsd.org" <net@freebsd.org>
Subject:   Re: Can DUMMYNET handle weighting of traffic according to firewall rules?
Message-ID:  <E1DD5C68-B4ED-4D10-B1D5-E0EED17D8C8B@jnielsen.net>
In-Reply-To: <201412121523.IAA03923@mail.lariat.net>
References:  <201412120711.AAA00622@mail.lariat.net> <CA%2BhQ2%2Bg40aZO%2B6JJsvDU8GG_UGp=rO1tQQoaETRe%2BBc-iyBNKA@mail.gmail.com> <201412121523.IAA03923@mail.lariat.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Dec 12, 2014, at 8:23 AM, Brett Glass <brett@lariat.net> wrote:

> At 03:06 AM 12/12/2014, Luigi Rizzo wrote:
>=20
>> you can set the limit for the pipe, create two queues with different
>> weights attached to the pipe, and then schedule.
>>=20
>> ipfw pipe 12 config bw 3456 Kbit/s
>> ipfw queue 34 config weight 2 pipe 12
>> ipfw queue 56 config weight 1 pipe 12
>> ipfw add queue 34 in recv halfduplexlink0
>> ipfw add queue 56 out xmit halfduplexlink0
>=20
> Alas, as I understand it (and also based on my empirical tests), this =
will give downstream traffic priority but will still let the same amount =
of upstream traffic through per second if there is no downstream =
traffic... because the capacity of the pipe is still the same. What I =
want to do is have the pipe, not the queue, weight the upstream traffic =
twice as heavily.

Is there a reason you can't use a separate pipe for each direction?




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?E1DD5C68-B4ED-4D10-B1D5-E0EED17D8C8B>