From mboxrd@z Thu Jan 1 00:00:00 1970 From: David McNab Date: Sun, 14 Mar 2004 04:11:54 +0000 Subject: Re: [LARTC] Specific question on ingress policing Message-Id: <4053DB8A.5000805@rebirthing.co.nz> List-Id: References: <4053C9BB.4050209@rebirthing.co.nz> In-Reply-To: <4053C9BB.4050209@rebirthing.co.nz> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: lartc@vger.kernel.org Roy wrote: > As I observed, setting 'rate 32kbit index 20' for each will not do anything > good, > they were behaving like independent clasess, so total trafic was sum of all > rates. > then I reduced rates for all clases su that sum of the rates was max rate > alocated for that poll > and it worked more or less. That's a good approach. I can easily change pyshaper (mentioned in earlier post) so that once it determines the number of connections matching a criterion, it divides the bandwidth allocation by the number of such connections, and generates tc ingress policing filter commands accordingly. > I am not sure if this was correct. > > probably the best way to find how index works is to look at source code. I'm very time-constrained. Given that I'm often having to study source code just to find out how to use other programs, I don't have the time to do this with tc, sadly. When someone writes a thorougly detailed manual on tc usage (including a complete tc-filters manpage), it'll be a good thing. > class www policer rate 100kbit index 20 > class p2p policer index 20 drop Can you please provide all your *complete* tc commands, so I can see this in context and form an understanding? > more or less it was working Good to know. -- Kind regards David -- leave this line intact so your email gets through my junk mail filter _______________________________________________ LARTC mailing list / LARTC@mailman.ds9a.nl http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/