[v6ops] Re: draft-ietf-v6ops-rfc7084bis

[Timothy Winters <[e-mail]>]

Hi Christian,

I think we should add a SHOULD requirement for RFC4638 (PPPoE with MTU
1500).  I will incorporate that in the next version.

For the MSS Clamping, I'm leaning towards not including it since it's
approach isn't standardized and may led to confusion.

Regards,
Tim

On Tue, Mar 17, 2026 at 2:13 AM Ole Trøan <[e-mail]> wrote:

>
>
> On 17 Mar 2026, at 05:02, Mark Smith <[e-mail]> wrote:
>
> I've always wondered how intelligent devices doing MSS hacking are. Do
> they blindly assume the location of the MSS option and just overwrite the
> values, or do they specifically find the MSS option before applying
> modifications to it. From what I can find, there is no requirement that the
> MSS option be in a fixed and the same location in a segment.
>
> MSS hacking has also been commonly enabled on ISPs' BNGs, and given how
> many packets per second they can typically deal with, and how many TCP
> connections, I've thought they might be making the blind assumption.
>
>
> mss_clamp_node.c « mss_clamp « plugins « src - vpp - Vector Packet
> Processing
> <https://git.fd.io/vpp/tree/src/plugins/mss_clamp/mss_clamp_node.c#n80>
> git.fd.io
> <https://git.fd.io/vpp/tree/src/plugins/mss_clamp/mss_clamp_node.c#n80>
> <https://git.fd.io/vpp/tree/src/plugins/mss_clamp/mss_clamp_node.c#n80>
> <https://git.fd.io/vpp/tree/src/plugins/mss_clamp/mss_clamp_node.c#n80>
>
> Typically not.
>
> Ole
> _______________________________________________
> v6ops mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]
>