[Idr] Re: BGP Origin Scrubbing (draft-marenamat-idr-scrub-bgp-origin-00)

[David Farmer <[e-mail]>]

“MUST NOT” is not going to change anything at this point. It’s not going to
make us into the protocol police, but would make us into protocol nannies,
waging our finger ISPs. So let’s NOT!

Changing the Origin Type doesn’t break the protocol, and if fact I would
argue it is using the protocol as it is intended.

I think the RIPE presenter has the right idea, default to setting the
Origin Type to IGP, but provide an opt out for customers.

But I do think the use of EGP in the wild is kind of weird.

Thanks

On Mon, Nov 3, 2025 at 18:01 Robert Raszuk <[e-mail]> wrote:

> Hi,
>
> Reading the RIPE91 slides from James in search for justification of this
> idea behind this draft it seems that the real issue is that some Tier1s are
> overwriting ORIGIN Attribute.
>
> So if anything I would suggest to open a github ticket towards RFC4271bis
> to make change in section 5.1.1:
>
> from:
>
> Its value SHOULD NOT be changed by any other speaker.
>
> changed into:
>
> Its value MUST NOT be changed by any other speaker.
>
> Regards,
> Robert
>
> On Tue, Nov 4, 2025 at 12:37 AM heasley <[e-mail]> wrote:
>
>> Mon, Nov 03, 2025 at 11:48:22PM +0100, Maria Matejka:
>> > >    The BGP Origin attribute in its original meaning has been out of
>> use
>> > >    for years.  Yet, the BGP Origin attribute has high priority in the
>> > >    best route selection algorithm, right after the AS Path length, and
>> > >    it's being used inconsistently over the Internet to manipulate the
>> > >    route preference.
>>
>> Yes, it is being used to affect route selection.  There are few knobs that
>> one can turn for that purpose.  I suggest that the attribute be left
>> alone.
>>
>> _______________________________________________
>> Idr mailing list -- [e-mail]
>> To unsubscribe send an email to [e-mail]
>>
> _______________________________________________
> Idr mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]
>