[Ndn-interest] HopLimit vs Interest aggregation

[Christos Papadopoulos]

On 04/01/2018 08:19 AM, David R. Oran wrote:
> On 30 Mar 2018, at 12:38, Junxiao Shi wrote:
>
>> Dear folks
>>
>> NDN Packet Format v0.3 introduces a HopLimit element in Interest packet.
>> The spec says:
>>
>> If element is present: if the HopLimit value is larger than or equal to 1,
>> a node should accept the packet and decrease the encoded value by 1.
>>
>>
>> My question is, how should a node proceed if multiple Interests with same
>> name but different HopLimit values are aggregated?
>>
>> Consider the following topology:
>>
>> A--\
>>      G---P
>> B--/
>>
>> First, G receives an Interest /P HopLimit=10 from A and forwards it to P
>> with HopLimit=9.
>> Later, after any suppression period has elapsed but before P replies and
>> before A's Interest expires, G receives another Interest /P HopLimit=5 from
>> B.
>> Should G forward the Interest to P with HopLimit=4, or HopLimit=9?
>>
>> I can see arguments both ways.
>> The argument for choosing HopLimit=4 is: it is possible that the Interest
>> from B is the previous Interest looped back. There could be a path from P
>> to B (not shown in the topology, as G does not know the global topology),
>> and one of the routers on that path has changed the nonce for probing.
>> The argument for choosing HopLimit=9 is: when G retries the Interest, it
>> should use the maximum HopLimit among unexpired downstream nodes, to
>> maximize the possibility of reaching the content.
>>
> Safety over optimality. Hop Limit is a protection mechanism. Be conservative.
> Always decrement the hop limit when forwarding. NEVER increase it.

In addition, Hop Limit (or the equivalent TTL in IP) has proven useful 
in scoped communication, especially in multicast. I think the wishes of 
the Interest sender should be respected, there may be a reason for the 
selected value of Hop Limit. Moreover, changing the value in the middle 
of the network without notifying the sender makes things very hard to debug.

Christos.

>
>> What do others think?
>>
>> Yours, Junxiao
>> _______________________________________________
>> Ndn-interest mailing list
>> Ndn-interest at lists.cs.ucla.edu
>> http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest
> DaveO
> _______________________________________________
> Ndn-interest mailing list
> Ndn-interest at lists.cs.ucla.edu
> http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest