For clarity, can you share prefix or add details. You should see withdraws propagate in seconds for most vantage points. In low single digit minutes anywhere but broken vantage points. When you say you see the prefix, do you mean you see it eveywhere, or you see it somewhere? Eveywhere, F5 for sure. Anywhere important, F5 probably Some odd irrelevant corner, almost certainly not F5 but broken vantage point not actually pushing traffic ________________________________ Lähettäjä: Saku Ytti <saku@ytti.fi> Lähetetty: Tuesday, March 3, 2026 10:49:29 AM Vastaanottaja: North American Network Operators Group <nanog@lists.nanog.org> Kopio: Elmar K. Bins <elmi@4ever.de> Aihe: Re: Can route-views be trusted? On Tue, 3 Mar 2026 at 10:41, Elmar K. Bins via NANOG <nanog@lists.nanog.org> wrote:
Local withdrawal does not result in *quick* global withdrawal. There's so much dampening involved, that you can get lucky, but that's not a given.
If you still see the prefix two hours later, that's when I'd start thinking deeper about the heat death of the universe. Until then, it is what it is.
Sir, are you saying Internet convergence typically takes hours? Almost no one uses dampening, and it hasn't been BCP in years. Perhaps if we'd have dampening that reduces local-pref, it could be useful. DFZ all the time has extremely flappy prefixes, which see many thousands of updates per day propagating invariably to any vantage point that publishes the data. Slow propagation is an exception, not rule. Most likely reason for such would not be dampening, but would be outQ that cannot be starved for example because lack of DRAM by the far end forced the window to zero. -- ++ytti