On Mar 30, 2011, at 2:23 PM, "Justin Dean" <[email protected]> wrote:

> "Deprecating the use of the IPv4 ID field for non-reassembly uses
>   should have little - if any - impact."
> 
> That statement wasn't true for very long was it...

Technically even if your devices rely on it it wouldn't be long before they 
started dropping valid packets, IMO.

> Well section 9 may provide a way out.
> 
> "Network address translators (NATs) and address/port translators
>   (NAPTs) rewrite IP fields, and tunnel ingresses (using IPv4
>   encapsulation) copy and modify some IPv4 fields, so all are
>   considered sources, as do any devices that rewrite any portion of the
>   source address, destination address, protocol, and ID tuple for non-
>   atomic datagrams [RFC3022]. As a result, they are subject to all the
>   requirements of any source, as has been noted."
> 
> With the current draft incoming packets can have their id fields modified by
> the SMF gateway before being forwarded in the MANET.  I would say this makes
> the gateway a source.  Then we just need to fragment everything!  I can see
> the reason for banning the behavior in high bandwidth wired links where
> duplicates only occur with routing loops and the id wraps so fast it is
> useless.  But the way this is written it bans ALL use of the IPv4 ID field
> for duplicate detection.
> 
> "The IPv4 ID field is no longer permitted for duplicate detection." 
> 
> Does this mean we can't even hash this field for duplicate detection?!

The field and the other parts of the header are not ensured unique for 
fragments right now - many hosts break the rules. A packet as a whole isn't 
either. 

It is much safer to expect the higher layers to drop such duplicates. You can 
drop but you need more requirements - e.g., that the higher layer won't send 
valid dups (TCP seqnos accomplish this), and you check/hash the whole packet.

The header *isn't* enough right now, unless you ensure rate limiting.

Joe

> 
>> -----Original Message-----
>> From: [email protected] [mailto:[email protected]] On Behalf
>> Of Teco Boot
>> Sent: Wednesday, March 30, 2011 7:29 AM
>> To: [email protected]; MANET IETF
>> Subject: [manet] SMF and intarea-ipv4-id-update
>> 
>> Sorry for x-posting. But there is a conflict in:
>> http://tools.ietf.org/html/draft-ietf-manet-smf
>> http://tools.ietf.org/html/draft-ietf-intarea-ipv4-id-update
>> 
>> SMF has a duplicate packet detection function based on the IPv4
>> ID field. So text in ietf-intarea-ipv4-id-update section 4
>> is not correct, in that there would be no deployments for such.
>> That said, SMF deployment with IPv4 DPD on IP-ID would be limited.
>> 
>> What to do?
>> 
>> Teco
>> 
>> 
>> 
>> 
>> _______________________________________________
>> manet mailing list
>> [email protected]
>> https://www.ietf.org/mailman/listinfo/manet
> 
> _______________________________________________
> Int-area mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/int-area
_______________________________________________
Int-area mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/int-area

Reply via email to