Re: [v6ops] draft-gont-v6ops-ipv6-ehs-in-real-world: clarification text

["Eric Vyncke (evyncke)" <[e-mail]>]

Fred,

The performance impact/network health issues are mainly related to
hop-by-hop which is per RFC 2460 inspected by every router on the path.
This EH should be ignore IMHO by all routers over the Internet (private
networks may still inspect/act on it for RSVP for examine).

If for some reasons, a router in the middle needs access to layer-4
information (IPFIX? DDoS mitigation? ... ?), then the EH chain must be
parsed which can cause a performance impact.

Do not take me wrong: beside ignoring HbH, I really hope that routers on
the paths simply look at layer-3 header and do nothing else. But, for this
I-D, I want to be sure to cover all aspects, hence the modified sentence

-éric

On 18/04/15 03:46, "Fred Baker (fred)" <[e-mail]> wrote:

>
>> On Apr 17, 2015, at 3:17 PM, Eric Vyncke (evyncke) <[e-mail]>
>>wrote:
>> 
>> Your proposal is fine for me, but, I would suggest a slightly stronger
>> text:
>> "The results presented in this document indicate that in the scenarios
>> where the corresponding measurements were performed, the use of IPv6
>> extension headers can lead to packet drops. We note that
>> packet drops occurring at transit networks is undesirable
>> and it is hoped and expected that this situation will improve over
>>time."
>> 
>> Should we say something around the lines of "... Undesirable except when
>> Those packets cannot be forwarded without impacting the performance and
>> the health of the network devices" ?
>
>Well, question for you. If we follow RFC 2460, the router in the middle
>doesn’t know whether the header is there or not. The only systems that
>should have a performance impact are systems that parse to them or
>interpret them. I personally would like to believe that we aren’t making
>*other* systems more complex to save them. If this is being done in a
>firewall or load balancer, the limitations of the device aren’t primary,
>the correct operation of the protocol is. Would we be saying that it’s OK
>for a firewall to drop a packet because it doesn’t feel like dealing with
>it that day?
>
>What specific devices did you have in mind? What would the argument be?