"Poor scores" is often discussed on community.ntppool.org. I gave a talk at the 2020 PTTI conference that gives some examples: http://www.leapsecond.com/ntp/NTP_Suitability_PTTI2020_Revised_Sommars.pdf Some ISPs/IXPs rate-limit NTP due to its historical misuse as a DDoS source.
Sometimes we're able to deduce where this is taking place. I've gotten no useful feedback from the organizations doing the rate-limitng, unfortunately. Steve Sommars On Tue, Apr 21, 2020 at 1:53 AM Jan Kasprzak <[email protected]> wrote: > Hello, Tom, > > Tom Yates wrote: > : According to https://www.ntppool.org/a/MadHatter , my server's not > : doing so well, since the afternoon of 18/4. Interesting thing is > : that the graph below is the same server, reached via ipv6. So the > : server's not the problem. > : > : I've asked my ISP if they're doing any kind of anti-DDoS filtering > : that might be blocking ipv4, and so far they say no. > : > : Any ideas where I can look next? I assume everyone's not seeing > : similar issues, or someone would have said so already here. > > I see similar problems with my server: > > https://www.ntppool.org/a/Yenya > > unlike yours, I see only occasional but deeper drops in IPv4 score, > IPv6 is fine. It seems that filtering or rate limiting (or packet loss > in general) happens several ASs upstream. What I usually do is to > look at the score of other pool members near me (in the same AS, > then in the same city in another AS). If I see a similar pattern in their > score graphs, the problem is probably not anywhere near me. > > I also tried to catch NTP traffic comming through my router/fw with > tcpdump, > and the number of incoming and outgoing NTP packets is about the same. > Which means my server and my router are not the ones who drop them. > > Sometimes I see low score only in the NTP Pool Beta monitoring site, > but sometimes only in the production site. > > It seems that the times when the Internet was TCP/IP are gone :-), > now anything beyond ICMP, DNS, and HTTP is suspicious... > > -Yenya > > -- > | Jan "Yenya" Kasprzak <kas at {fi.muni.cz - work | yenya.net - private}> > | > | http://www.fi.muni.cz/~kas/ GPG: 4096R/A45477D5 > | > IORING_OP_NOP ... the benefits of doing nothing asynchronously are minimal, > but sometimes a placeholder is useful. --Jonathan Corbet at LWN > _______________________________________________ > pool mailing list > [email protected] > http://lists.ntp.org/listinfo/pool > _______________________________________________ pool mailing list [email protected] http://lists.ntp.org/listinfo/pool
