Re: How long AS-PATH policies have you used
Saku Ytti via NANOG <[email protected]>
| Newsgroups | gmane.org.operators.nanog |
|---|---|
| Message-ID | <CAAeewD8AaDYWQhHHXi=Jqg08mF+uQK2UQZAgaSaEBENjZOJ4Qg@mail.gmail.com> |
Thank you Jakob. This is exactly what I need, and indeed scaling will be terrific. Meaning, you can immediately prove it scales to the entire DFZ and has a margin. That is, if you change AS-SET prefix-list to AS-SET asn origin + synthesized you get configuration which will always commit, pathological AS-SET are an Internet problem, not a Me problem. While improving security, because non ROAd route objects will get their origin checked as well. On Sat, 7 Mar 2026 at 04:20, Jakob Heitz (jheitz) via NANOG <[email protected]> wrote: > > > That was fixed in IOS-XR about 5 years ago. > Previously, we could only do it with as-path-set in route-policy. > as-path-set allows whole paths to be specified, so it's slow. > If all you want is to match on a single ASN, it can be done much faster using a tree lookup instead of linear. > So, I invented the as-set for route-policy configuration. > Here is a config sample: > > as-set asset1 > 2914, > 174, > 3356 > end-set > ! > route-policy asd > if as-path originates-from asset1 then > pass > endif > end-policy > > If that doesn't meet your needs, let me know. > Kind Regards, > Jakob Heitz > > Date: Tue, 24 Feb 2026 20:15:00 +0200 > From: Saku Ytti <[email protected]> > > > For SROS, IOSXR 90k would be quite a ridiculous attempt, and it's > probably cheaper just to expand to a million lines of prefix-list, > since prefix-list scale is more tested than AS_PATH scale. > In SROS as-path-group can contain only 128 lines, so if you match a > single ASN per line, you'd need 700 terms just to check the origin, > unless you use regexp OR in the lines to put multiple origins per > line. > > _______________________________________________ > NANOG mailing list > https://lists.nanog.org/archives/list/[email protected]/message/EAP7ZNT32OGAQM4PMTNQQE76DMTEWNMV/ -- ++ytti _______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/[email protected]/message/Q263NVIISFQ4QYHVQUX4T42AH4Q4JZEA/