[ppml] Deaggregation in the ARIN region (typical gallery noise)

Chip Mefford cpm at well.com
Tue Feb 15 07:50:10 EST 2005


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Leo Bicknell wrote:
| In a message written on Sun, Feb 13, 2005 at 09:41:28AM -0800, Andrew
Dul wrote:
|
|>I was wondering if people in the ARIN region agreed with these statements
|>about the ARIN region not doing enough to educate the membership about
|>deaggregation, BGP annoucements etc...
|> snipped in original;

| You can split the issues into several categories, and I don't know
| how ARIN can help with any of them.  First, in many cases no one
| can bother to contact the offending party.  Obviously their own
| upstream is a prime candidate to notice and contact the person in
| question.  If we can't get the major ISP's who have a clue, and
| have staff, and know the issues to make it a priority to contact
| the smaller players and help them along then I don't know what hope
| we have for some outside influence to make a difference.

I'm just being a typical punter here, however, I feel that issues
happening upstream should be reported or commented on upstream.
Folks in general should have a working relationship with their
next-hop up. I know this is often hearache and pain, but
it is rational. Top down enforcement is pretty much always
resented in pretty much all things.

| Second, when contacted, one of two things happen.  The parties often
| clean up their act simply by someone asking (the, "oops, I didn't
| know I was doing that" situation).

I'm pretty much a bottom feeder, I run a few small networks,
and when I do something that my users don't like, they let me
know and I often say "Oops". I think this scales.

|In the other cases, the response
| is often "I have a reason to do this and I'm not explaining it to
| you."  I suppose ARIN might be able to do something in the former
| case although I question if it is appropriate.  In the latter case,
| unless the industry as a whole takes a much stronger stance and
| imposes sanctions (eg, route filtering) then no one complaining
| will make a difference.

route filtering gets into some very dodgy ground pretty fast.
route filtering is tantamount to broad stroke censorship.

| Third, and most important, does anyone care?  We're not at a stage
| anymore where the routers are on the verge of falling over.  Other
| than a few complainers, it's not clear this issue impacts the day
| to day operation of anyone's network in a significant way.  Obviously
| there is the potential, if it gets worse it could get real bad real
| quick.  There seems to be a level of tolerance for slop in the
| system right now, and that's probably a good thing.
|
| So, I too question if ARIN is the right place.

Not so long ago, I was blaming ARIN for all the sins in the
world, 'till a college brought me up short and told me to
go to ARIN meetings and shutup and pay attention. ARIN is
"Us" in many ways, Having spent some time talking to those
who know better than I do about a lot of stuff, ARIN stuff
included, I realised that "we" bring a lot of this on ourselves
by thinking that our networks are somehow autonomous and don't
really exist as part of a whole. Our peering providers, immho,
should be the first place to address issues, and often many issues
can be resolved, or at least better comprehended by addressing
concernings straight upstream.

In short, if you are trying to fix things globally, get them
fixed locally, and the global issues will get resolved.

And, btw, where is the current best practice on this issue documented?

- --chipper

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)

iD8DBQFCEfAC0STXFHxUucwRAv7gAJ0eTgEHYHm1ieI7cV8YYwEAc2SY0QCeOFl3
6jphhnl1rkBz3vpiw9rZ7d4=
=tBgF
-----END PGP SIGNATURE-----



More information about the ARIN-PPML mailing list