On Jun 17, 2012, at 0:12, Kiss Gábor wrote: > Would you mind to check page http://www.pool.ntp.org/user/kissg ?
Thank you for having marked your profile public. I try not to put peoples IPs and such on the list when we discuss things unless they've done so, and obviously having specific cases to talk about is much more productive. > 1. Please compare IPv4 and IPv6 grapsh. > Can you see the different characteristics? Yes, I think it's a good example of how IPv4 routing and connectivity is just better (still). Also, I think there are some local IPv6 problems by the monitoring server I am trying to figure out with the help of the bandwidth provider. Here is an IPv4 traceroute, after running for about 10-15 minutes. Notably the "standard deviation" (in ping times) is very small. The slowest round trip (186.3) to the destination is just 9ms slower than the fastest and the average (177.8) just 0.5ms slower than the fastest (177.3). Very stable. Packets Pings Host Loss% Last Avg Best Wrst StDev 1. fe2-8.hr01.lax2.phyber.com 0.0% 0.8 10.9 0.5 201.3 35.3 2. xe1-1.cr02.lax1.phyber.com 0.0% 0.7 2.9 0.6 120.9 11.0 3. ge-6-24-284.car2.LosAngeles1.Level3.net 0.0% 0.6 4.2 0.5 62.6 10.5 4. vlan60.csw1.LosAngeles1.Level3.net 0.0% 1.3 3.9 0.6 13.2 3.8 5. ae-63-63.ebr3.LosAngeles1.Level3.net 0.0% 1.3 1.3 0.6 6.8 0.7 6. 4.69.132.82 0.0% 66.7 66.0 64.6 82.7 2.1 7. ae-71-71.csw2.Washington1.Level3.net 0.0% 73.3 68.5 64.6 94.9 4.1 8. ae-72-72.ebr2.Washington1.Level3.net 0.0% 66.0 65.6 64.6 69.3 0.6 9. ae-44-44.ebr2.Paris1.Level3.net 0.0% 144.1 144.7 144.0 156.6 1.1 ae-42-42.ebr2.Paris1.Level3.net ae-41-41.ebr2.Paris1.Level3.net ae-43-43.ebr2.Paris1.Level3.net 10. ae-46-46.ebr1.Frankfurt1.Level3.net 0.0% 154.8 153.4 152.7 160.4 0.9 ae-48-48.ebr1.Frankfurt1.Level3.net 11. ae-1-12.bar1.Budapest1.Level3.net 12.9% 168.5 170.8 167.2 230.7 8.9 12. ae-0-11.bar2.Budapest1.Level3.net 7.2% 168.6 170.6 168.3 278.9 9.0 13. dialup-212.162.26.186.frankfurt1.mik.net 0.0% 169.2 169.5 168.7 207.7 2.7 14. be13.rtr1.vh.hbone.hu 0.0% 183.7 179.7 179.0 193.6 1.1 15. vl799.sw.b1.vh.hbone.hu 0.0% 179.5 179.6 178.5 250.2 4.2 16. cirkusz.lvs.iif.hu 0.0% 177.6 177.8 177.3 186.3 0.8 Now IPv6, running 30 seconds or so already shows much "messier" data: Packets Pings Host Loss% Last Avg Best Wrst StDev 1. fe2-8.hr01.lax2.phyber.com 0.0% 1.4 1.6 0.7 4.6 0.9 2. 2607:f238:0:10::2 0.0% 1.8 1.3 0.8 3.1 0.6 3. xe2-3.cr02.lax1.phyber.com 0.0% 0.8 6.6 0.8 96.2 21.2 4. 2607:f238:0:14::2 0.0% 0.8 1.4 0.8 2.1 0.4 5. 2001:504:13::1a 0.0% 1.5 3.9 0.8 13.1 3.3 6. 10gigabitethernet7-3.core1.sjc2.he.net 0.0% 9.2 11.7 9.2 20.0 3.0 7. sjo-eqx-s1-link.telia.net 0.0% 9.1 11.0 9.0 29.5 4.7 8. bpt-b4-v6.telia.net 0.0% 192.9 197.6 189.8 241.1 13.7 9. 2001:2000:3080:17::2 0.0% 198.3 196.6 194.1 198.4 1.6 10. lo0.sw.b1.vh.hbone.hu 0.0% 197.2 196.2 193.8 198.1 1.7 11. cirkusz.lvs.iif.hu 0.0% 186.3 194.7 186.3 201.8 4.9 Best round trip is almost the same as the IPv4 path (186.3ms), but the average is 194.7 and the slowest is 201.8ms. Much harder to get an accurate time if each packet can take that much different time. Ask _______________________________________________ pool mailing list pool@lists.ntp.org http://lists.ntp.org/listinfo/pool