[Cake] Questions on Cake/cobalts experiment ingress mode

[Cake] Questions on Cake/cobalts experiment ingress mode

[Kevin Darbyshire-Bryant]

Hi All,

I've been out of the loop for a while and just noticed the 'ingress' 
mode for cake in the cobalt experimental branch.

Colour me confused on how to use this.  My (typical) situation is I've 
'eth0' as my WAN port connected to an ISP VDSL2 modem - this is 
naturally asymmetric in capacity (80Mbps in to me, 20Mbps out from me)

Egress shaping is easy with cake....just add the qdisc to 'eth0' and it 
all works.  Ingress requires generating a 'clone' interface (IFB) of 
eth0 to get the incoming packets and shaping the egress of the IFB.

Can I now apply two qdiscs to eth0 directly???  An ingress mode cake and 
an egress mode cake?  Or rather do I just need to tell my existing IFB 
cake instance to run in 'ingress' mode.


Like I said..confused :-)

Kevin

Re: [Cake] Questions on Cake/cobalts experiment ingress mode

[Jonathan Morton]

> On 6 Apr, 2017, at 12:17, Kevin Darbyshire-Bryant <kevin@darbyshire-bryant.me.uk> wrote:
> 
> Can I now apply two qdiscs to eth0 directly???  An ingress mode cake and an egress mode cake?  Or rather do I just need to tell my existing IFB cake instance to run in 'ingress' mode.

The latter.  The fundamental limitation has not been removed (and is not in Cake itself); this is just a modification to how dropped packets are accounted for when the bottleneck is upstream of Cake.

 - Jonathan Morton

Re: [Cake] Questions on Cake/cobalts experiment ingress mode

[Kevin Darbyshire-Bryant]

Ahh, cheers Jonathan, understand! (and thanks for the really quick reply)


Kevin (much less confused)

On 06/04/17 10:19, Jonathan Morton wrote:
>
>> On 6 Apr, 2017, at 12:17, Kevin Darbyshire-Bryant <kevin@darbyshire-bryant.me.uk> wrote:
>>
>> Can I now apply two qdiscs to eth0 directly???  An ingress mode cake and an egress mode cake?  Or rather do I just need to tell my existing IFB cake instance to run in 'ingress' mode.
>
> The latter.  The fundamental limitation has not been removed (and is not in Cake itself); this is just a modification to how dropped packets are accounted for when the bottleneck is upstream of Cake.
>
>  - Jonathan Morton
>