netdev
[Top] [All Lists]

Re: [PATCH] netem: account for packets in delayed queue in qlen

To: Patrick McHardy <kaber@xxxxxxxxx>
Subject: Re: [PATCH] netem: account for packets in delayed queue in qlen
From: jamal <hadi@xxxxxxxxxx>
Date: Thu, 21 Apr 2005 20:32:19 -0400
Cc: Stephen Hemminger <shemminger@xxxxxxxx>, "David S. Miller" <davem@xxxxxxxxxxxxx>, netdev@xxxxxxxxxxx
In-reply-to: <426840E3.8000203@trash.net>
Organization: unknown
References: <20050329152110.38d50653@dxpl.pdx.osdl.net> <4252EB9D.9070305@trash.net> <20050407120417.4297cd14@dxpl.pdx.osdl.net> <42628300.9010007@trash.net> <20050419110639.47767113@localhost.localdomain> <42666098.5060409@trash.net> <20050421132020.41858bc4@localhost.localdomain> <426832E1.7020003@trash.net> <20050422092221.4950e23c@localhost.localdomain> <426839B4.2090502@trash.net> <1114128336.10572.12.camel@localhost.localdomain> <426840E3.8000203@trash.net>
Reply-to: hadi@xxxxxxxxxx
Sender: netdev-bounce@xxxxxxxxxxx
On Fri, 2005-22-04 at 02:10 +0200, Patrick McHardy wrote:
> jamal wrote:
> > Duplication of packets would be trivial to do with mirred mirror option
> > attached to the qdisc.
> 
> The problem with duplication and queueing to the device is that
> the packet needs to take the same classification-path as the
> original packet to end up in the same leaf-queue. This can only
> be guaranteed by restricting which classifiers can be used.

What i described happens before the queueing. So packets should end up
in the same leaf queue because the same classifier rule is used to
select the dup action. I should actually try to make sure it works;

cheers,
jamal


<Prev in Thread] Current Thread [Next in Thread>