Re: [mif] mif Digest, Vol 5, Issue 18

Min Hui escribió:
> Hi, marcelo
>
>
> 2009/3/19 marcelo bagnulo braun <marcelo@it.uc3m.es>:
>   
>> Min Hui escribió:
>>     
>>> Hi,  Yuanchen
>>>
>>> pls see my reply in line. thx.
>>>
>>> 2009/3/17 ma yc <ycma610103@gmail.com>:
>>>
>>>       
>>>> Hi Min
>>>>
>>>> Thank you for your reply. Please check inline.
>>>>
>>>> Yuanchen
>>>>
>>>> 2009/3/17 Min Hui <huimin.cmcc@gmail.com>:
>>>>
>>>>         
>>>>> Hi, Yuanchen
>>>>>
>>>>> For the first question, there are two situations.
>>>>>
>>>>> 1. the host sends data using one of its interfaces, and the correspond
>>>>> node reply to the address of this interface. it's has no problem, the
>>>>> host will receive the reply by the same interface.
>>>>> 2. the network side initiate the communication, the host name will be
>>>>> used in the DNS query to get a list of addresses, which one should be
>>>>> choosed depends on the network side policy.
>>>>>
>>>>>           
>>>> yuanchen> i agree with this scenario that the DNS query may result
>>>> in a list of addresses. But i still don't understand what "network
>>>> side initiated"
>>>> means. IP flow is end-to-end and the other end node should
>>>> make the DNS query and make the decision. Do you mean that the other end
>>>> node should get some information from somewhere in the network to help
>>>> the
>>>> decision making? I think the requirement should be clarified.
>>>>
>>>>
>>>>         
>>> Sorry for misleading the "network side", the meaning is when the other
>>> end node initiates the communication, it will not directly target one
>>> of the addresses, the network element will make this decision and
>>> return one address to this node.
>>>
>>>       
>> I don't understand this
>> Usually, the DNS is the one that returns the addresses, Are you considering
>> that the network influences the DNS responses or are you thinking in
>> something else?
>>     
>
> I mean the network influences the DNS responses. DNS need to pick one
> address among multiple addresses and return it to the node, the
> network could have some selection policy to influence the DNS
> responses.
>   

we need to look into this a bit more in detail...i am not sure it is the 
best idea to have the network tweaking the dns responses.

regards, marcelo

>   
>> Regards, marcelo
>>
>>     
>>>>> For the second question, the element would be the gateway which in
>>>>> charge of all the interfaces, e.g. PDN GW in 3GPP architecture, LMA in
>>>>> PMIP domain.
>>>>>
>>>>>           
>>>> yuanchen> I do not think LMA is a good example for MIF, since the LMA is
>>>> for
>>>> mobility management. PND GW in 3GPP also has the function of LMA.
>>>> Such scenario sounds more like extension to PMIP for multiple interface
>>>> node
>>>> support. Netlmm may be better place for such topic.
>>>>
>>>>
>>>>         
>>> I'll avoid to discuss mobility management in mif, but I'm not very
>>> sure whether PDN GW example isn't suitable to be discussed in mif,
>>> because it's a network element which is responsible for all the
>>> interfaces in the 3GPP architecture, and doesn't only appear as a LMA.
>>>
>>>
>>>
>>>       
>>>>> Thanks for your comments.
>>>>>
>>>>> - Hui Min
>>>>>
>>>>> 2009/3/17 ma yc <ycma610103@gmail.com>:
>>>>>
>>>>>           
>>>>>> Hi, Min Hui,
>>>>>>
>>>>>> I an also confused on the requirement 5.
>>>>>>
>>>>>> As for the the flow directed to the multiple
>>>>>> interface node, why does the network side need
>>>>>> to decide which interface to forward the data?
>>>>>>
>>>>>> The dst address of the packet should be used for routing.
>>>>>> The network just forwards the data according to the routing table.
>>>>>> I do not see the needs for interface selection to route the flow.
>>>>>>
>>>>>> Also if you have the use case, i have the same concern
>>>>>> as macelo, which elememt will be in chcarge of
>>>>>> distributing the traffic?
>>>>>>
>>>>>> Thank you.
>>>>>> Regards
>>>>>> Yuanchen
>>>>>>
>>>>>>
>>>>>>             
>>>>>>> Date: Mon, 16 Mar 2009 18:13:29 +0100
>>>>>>> From: marcelo bagnulo braun <marcelo@it.uc3m.es>
>>>>>>> Subject: Re: [mif] about draft-hui-ip-multiple-connections-ps-02
>>>>>>> To: Min Hui <huimin.cmcc@gmail.com>
>>>>>>> Cc: mif <mif@ietf.org>
>>>>>>> Message-ID: <49BE88B9.6010306@it.uc3m.es>
>>>>>>> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>>>>>>>
>>>>>>>
>>>>>>>               
>>>>>>>>  5 Network side should be capable of distributing the IP flow
>>>>>>>>  according to some parameters, such as IP address prefix, network
>>>>>>>> type
>>>>>>>>  and so on.
>>>>>>>>
>>>>>>>> i don't understnad what you mean by this one, could you expand?
>>>>>>>>
>>>>>>>> A: That is the policy in the network side. Corresponding to the
>>>>>>>> policy
>>>>>>>> of sending data mentioned in the fourth bullet, the policy of
>>>>>>>> receiving data is also needed, which can be apply in the network
>>>>>>>> side.
>>>>>>>> The network can determine forward a specific IP flow to which
>>>>>>>> interface of the destination host according to the policy.
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>                 
>>>>>>> right, i see what you mean. I am not sure which element of the network
>>>>>>> would do that... i mean, are you assuming that all interfaces are
>>>>>>> being
>>>>>>> connected to the same ISP? If not, i am not sure how would you do
>>>>>>> this...
>>>>>>>
>>>>>>>               
>>>>>>>> Regards, marcelo
>>>>>>>>
>>>>>>>>
>>>>>>>>                 
>>>>>> _______________________________________________
>>>>>> mif mailing list
>>>>>> mif@ietf.org
>>>>>> https://www.ietf.org/mailman/listinfo/mif
>>>>>>
>>>>>>
>>>>>>             
>>> _______________________________________________
>>> mif mailing list
>>> mif@ietf.org
>>> https://www.ietf.org/mailman/listinfo/mif
>>>
>>>
>>>       
>>     
>
>   