nanog mailing list archives
Re: ECN
From: Matt Corallo <nanog () as397444 net>
Date: Wed, 13 Nov 2019 12:00:25 -0500
Not ideal, sure, but if it’s only for the SYN (as you seem to indicate), splitting the flow shouldn’t have material performance degradation?
On Nov 13, 2019, at 11:51, Toke Høiland-Jørgensen <toke () toke dk> wrote: On 13 November 2019 17:20:18 CET, Matt Corallo <netadmin () as397444 net> wrote: This sounds like a bug on Cloudflare’s end (cause trying to do anycast TCP is... out of spec to say the least), not a bug in ECN/ECMP.Even without anycast, an ECMP shouldn't hash on the ECN bits. Doing so will split the flow over multiple paths; avoiding that is the whole point of doing the flow-based hashing in the first place. Anycast "only" turns a potential degradation of TCP performance into a hard failure... :) -Toke
Current thread:
- Re: ECN, (continued)
- Re: ECN Saku Ytti (Nov 14)
- Re: ECN Tore Anderson (Nov 13)
- Re: ECN Warren Kumari (Nov 13)
- TCP and anycast (was Re: ECN) Anoop Ghanwani (Nov 13)
- Re: TCP and anycast (was Re: ECN) Bill Woodcock (Nov 14)
- Re: TCP and anycast (was Re: ECN) William Herrin (Nov 14)
- Re: TCP and anycast (was Re: ECN) Randy Bush (Nov 14)
- Re: TCP and anycast (was Re: ECN) Christopher Morrow (Nov 14)
- Re: TCP and anycast (was Re: ECN) Randy Bush (Nov 14)
- Message not available
- Re: ECN Toke Høiland-Jørgensen via NANOG (Nov 13)
- Re: ECN Matt Corallo (Nov 13)
- Re: ECN Anoop Ghanwani (Nov 13)
- Re: ECN Owen DeLong (Nov 13)
- Re: ECN Toke Høiland-Jørgensen via NANOG (Nov 14)
- Re: ECN Jon Lewis (Nov 13)