[Oisf-devel] RFC: vlan_id in flow tracking

[Victor Julien]

On 04/08/2013 05:19 PM, Eric Leblond wrote:
> Hi,
> 
> On Mon, 2013-04-08 at 16:53 +0200, Victor Julien wrote:
>> On 04/08/2013 03:47 PM, Eric Leblond wrote:
>>> Hi,
>>>
>>> On Mon, 2013-04-08 at 15:13 +0200, Victor Julien wrote:
>>>> On 04/08/2013 03:00 PM, Eric Leblond wrote:
>>>>> Hi,
>>>>>
>>>>> On Mon, 2013-04-08 at 14:35 +0200, Victor Julien wrote:
>>>>>> On 04/08/2013 01:44 PM, Eric Leblond wrote:
>>>>>>> Hi,
> ...
>>>>> where Network (private IP for example) may be shared among client. And
>>>>> if we have no luck, then we can cross beams and have two clients with
>>>>> same network under the same VLAN. 
>>>>
>>>> Right, I think I get the point. If we would have:
>>>>
>>>> [vlan 1][vlan 2][client net 1]
>>>> [vlan 2][vlan 2][client net 2]
>>>>
>>>> We might mix both as we first peel off the outer vlan and then have no
>>>> way of distinguishing between the 2 client nets anymore, right?
>>>
>>> Exactly!
>>>
>>>> So if we have multiple layers, we'd need multiple tags in our flow hashing?
>>>>
>>>> [vlan 1][vlan 2][client net 1]
>>>> [vlan 2][vlan 2][client net 2]
>>>>
>>>> Here both tag 1 and 2 would be needed to get flows for "client net 1"
>>>> and tag 2 and 2 would be needed to get "client net 2". Make sense?
>>>
>>> Yes, this is the best behavior. 
>>
>> RFC:
>> https://github.com/inliniac/suricata/commit/e2693b6ce6ae5f7e43da26b47fdc2220dc1c3fa5
> 
> I've got few comments on it but to consider the perspective of MPLS
> support. In this case, we will have the MPLS label to separate
> clients/network in the same way has it is done with vlan id. Maybe we
> could use a pointer inside the flow structure to indicate the
> "separator" field. This would also be of use if we consider a set of GRE
> tunnels used to separate network.

Not sure if we should consider MPLS at this time. We've discussed
several ways to support it, including a flow engine per label. We'll get
to that when the time is right :)

Wrt GRE, I've been browsing my pcap collection, but I can't seem to find
something like a "vlan id" or "mpls label" in it. They "key" field isn't
it I think.

>>>> In my patch I have a u16 for padding, so I could add support for dual
>>>> layer vlans easily. For more... well that would be more work. Maybe we
>>>> can just limit to 2 layers.
>>>
>>> From http://en.wikipedia.org/wiki/IEEE_802.1Q triple tagging is not
>>> standard.
>>
>> I added a check for this in my RFC patch above.
> 
> It seems clean.
> 
>>> By the way, tagging host in the same way will be needed (thinking to
>>> threshold mainly).
>>
>> Right. That is going to be interesting wrt IPrep :) Maybe I should move
>> iprep out of the hosts.
> 
> Yes, multi-client is something that would result in some complex and
> huge work :)

Yeah. Taking it a step at a time...

-- 
---------------------------------------------
Victor Julien
http://www.inliniac.net/
PGP: http://www.inliniac.net/victorjulien.asc
---------------------------------------------