From mboxrd@z Thu Jan  1 00:00:00 1970
From: tc@forest.one.pl
Date: Sun, 24 Apr 2005 11:03:30 +0000
Subject: Re: [LARTC] Strange pings.
Message-Id: <200504241303.30357.tc@forest.one.pl>
List-Id: <lartc.vger.kernel.org>
References: <200504200055.42395.tc@forest.one.pl>
In-Reply-To: <200504200055.42395.tc@forest.one.pl>
MIME-Version: 1.0
Content-Type: text/plain; charset="windows-1252"
Content-Transfer-Encoding: quoted-printable
To: lartc@vger.kernel.org

On Friday 22 of April 2005 01:34, you wrote:
> tc@forest.one.pl wrote:
>=20
> > 64 bytes from 192.168.1.1: icmp_seq=95 ttld time=89.0 ms
>=20
> Which is about 2 packets @ 256kbit. I tested and got the same behaviour=20
> with a simple setup, but max ping about 45 because htb dequeues in pairs =

> by default. If you change #define hysteresis from 1 to 0 in=20
> net/sched/sch_htb.c then it's more accurate. Setting Quantum to your MTU =

> may also help.
ok. I changed hysteresis and I changed quantum.
>=20
> > Then it is stable (0.4ms) for some time.
> >=20
> > The same situation repeats in about 60 second delay.
> >=20
> > Maybe someone solved this problem?
>=20
> Give interactive class more rate than it needs. Latency was OK for me=20
> with ICMP class rate 255kbit ceil 256kbit.
>=20
I can't give 255 kbit to icmp traffic :) But for me 4kbit seems to be enoug=
h.=20
On testing environment it works fine. But on "real" configuration
the pings are still too long.=20
I wonder if one-level configuration (with just one root qdisc) would give
better latency.

> >=20
> > PPS. Is it possible to create a filter that will match all packets?
>=20
> I don't know about all, but all per protocol like -
>=20
> .. protocol ip prio 10 u32 match u32 0 0 ..
> .. protocol arp prio 11 u32 match u32 0 0 ..
>=20
It works ok for me, but only with prio 0.

Thanks for help.

Maciek.
_______________________________________________
LARTC mailing list
LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/cgi-bin/mailman/listinfo/lartc