Hi Fred,

On 05/29/2013 08:32 PM, Templin, Fred L wrote:
> Hi Lori,
>
>> -----Original Message-----
>> From: Lori Jakab [mailto:[email protected]]
>> Sent: Wednesday, May 29, 2013 9:58 AM
>> To: Templin, Fred L
>> Cc: Paul Vinciguerra; Lori Jakab; Brian Haberman; draft-ietf-lisp-
>> [email protected]; [email protected]
>> Subject: Re: [lisp] AD Evaluation: draft-ietf-lisp-deployment
>>
>> On 05/28/2013 11:38 PM, Templin, Fred L wrote:
>>>> -----Original Message-----
>>>> From: Paul Vinciguerra [mailto:[email protected]]
>>>> Sent: Tuesday, May 28, 2013 12:58 PM
>>>> To: Templin, Fred L; Lori Jakab; Brian Haberman
>>>> Cc: [email protected]; [email protected]
>>>> Subject: RE: [lisp] AD Evaluation: draft-ietf-lisp-deployment
>>>>
>>>>>   "Encapsulation results in an end-to-end path MTU of less than
>>>>>   1500 bytes, if encapsulated packets need to travel over links
>>>>>   with MTU lower than 1500 bytes + LISP encapsulation overhead."
>>>>>
>>>>> Thanks - Fred
>>>>> [email protected]
>>>>>
>>>> [PV]
>>>> How about
>>>> "Packet Encapsulation may exceed the MTU of the underlying physical
>>>> media.   To address the MTU constraints of a given physical media,
>> the
>>>> MTU of the original payload may need to be reduced or be fragmented
>> and
>>>> encapsulated in multiple packets."
>>> LISP and other tunnel types need to concern themselves with the path
>>> MTU between the tunnel ingress and egress, so it is about more than
>>> just the underlying physical media of the first hop into the tunnel.
>> I think the wording by Brian is both clear and concise, and addresses
>> your concerns (doesn't use "may" and refers to the end-to-end path).
>> Unless there is strong objection, I will use that wording.
> Almost. Suggest changing Brian's wording slightly as follows:
>
>   "Encapsulation increases the amount of overhead associated with each
>    packet.  For LISP, this added overhead decreases the effective
>    end-to-end path MTU."
>
> Reason is that this statement is true for LISP, but not necessarily
> true for all other tunnel types.

I tried finding a tunneling protocol where it isn't true, but I came up
empty.  Can you please give an example of a tunnel where the
encapsulation overhead does not result in a decrease of the E2E PMTU? 
Even if the subject matter is LISP, and even if there is an exception, I
wouldn't single it out when *most* tunneling protocols have the same issue.

-Lori

>
> Thanks - Fred
> [email protected]
>
>
>> Thanks,
>> -Lori

_______________________________________________
lisp mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/lisp

Reply via email to