Some leaks in China/Hongkong

Fri Nov 7 10:37:22 CET 2008

Hi Carlos, Hi All,

Speaking of commercial isp's, I see a huge difference from here... (40% +/-)

is this GeoIP / DNS thingie, or something really nasty on v4 ?

# ping www.google.com
PING www.l.google.com (64.233.183.103) 56(84) bytes of data.
64 bytes from nf-in-f103.google.com (64.233.183.103): icmp_seq=1 ttl=244 time=59.6 ms
64 bytes from nf-in-f103.google.com (64.233.183.103): icmp_seq=2 ttl=244 time=56.8 ms
64 bytes from nf-in-f103.google.com (64.233.183.103): icmp_seq=3 ttl=244 time=57.1 ms
64 bytes from nf-in-f103.google.com (64.233.183.103): icmp_seq=4 ttl=244 time=58.9 ms
64 bytes from nf-in-f103.google.com (64.233.183.103): icmp_seq=5 ttl=244 time=58.7 ms
64 bytes from nf-in-f103.google.com (64.233.183.103): icmp_seq=6 ttl=244 time=58.1 ms
64 bytes from nf-in-f103.google.com (64.233.183.103): icmp_seq=7 ttl=244 time=57.3 ms
64 bytes from nf-in-f103.google.com (64.233.183.103): icmp_seq=8 ttl=244 time=57.7 ms
64 bytes from nf-in-f103.google.com (64.233.183.103): icmp_seq=9 ttl=244 time=59.6 ms
64 bytes from nf-in-f103.google.com (64.233.183.103): icmp_seq=10 ttl=244 time=60.2 ms

--- www.l.google.com ping statistics ---
10 packets transmitted, 10 received, 0% packet loss, time 8999ms
rtt min/avg/max/mdev = 56.822/58.447/60.252/1.161 ms

# ping6 ipv6.google.com
PING ipv6.google.com(2001:4860:0:1001::68) 56 data bytes
64 bytes from 2001:4860:0:1001::68: icmp_seq=1 ttl=56 time=38.7 ms
64 bytes from 2001:4860:0:1001::68: icmp_seq=2 ttl=56 time=36.6 ms
64 bytes from 2001:4860:0:1001::68: icmp_seq=3 ttl=56 time=36.8 ms
64 bytes from 2001:4860:0:1001::68: icmp_seq=4 ttl=56 time=36.3 ms
64 bytes from 2001:4860:0:1001::68: icmp_seq=5 ttl=56 time=36.2 ms
64 bytes from 2001:4860:0:1001::68: icmp_seq=6 ttl=56 time=36.5 ms
64 bytes from 2001:4860:0:1001::68: icmp_seq=7 ttl=56 time=36.2 ms
64 bytes from 2001:4860:0:1001::68: icmp_seq=8 ttl=56 time=36.8 ms
64 bytes from 2001:4860:0:1001::68: icmp_seq=9 ttl=56 time=36.9 ms
64 bytes from 2001:4860:0:1001::68: icmp_seq=10 ttl=56 time=36.7 ms

--- ipv6.google.com ping statistics ---
10 packets transmitted, 10 received, 0% packet loss, time 9033ms
rtt min/avg/max/mdev = 36.261/36.829/38.737/0.685 ms

# traceroute www.google.com
traceroute: Warning: www.google.com has multiple addresses; using 64.233.183.103
traceroute to www.l.google.com (64.233.183.103), 30 hops max, 40 byte packets
 1  ge130-1000m.cr2.lisboa.nfsi.pt (81.92.200.126)  0.139 ms  0.199 ms  0.113 ms
 2  xe12-10GE.xmr1.as25137.net (81.92.201.10)  1.359 ms  0.348 ms  0.365 ms
 3  if-1-2-151.core1.PV9-Lisbon.as6453.net (195.219.186.9)  0.487 ms  0.346 ms  0.240 ms
 4  if-12-0-0.mcore3.L78-London.as6453.net (195.219.144.5)  27.349 ms  27.073 ms  26.976 ms
 5  if-5-0-0.mcore3.LDN-London.as6453.net (195.219.195.9)  32.463 ms  32.324 ms  32.613 ms
 6  ldn-b4-link.telia.net (213.248.74.1)  32.446 ms  34.502 ms  32.539 ms
 7  ldn-bb1-link.telia.net (80.91.249.77)  32.622 ms ldn-bb1-link.telia.net (80.91.250.234)  32.700 ms  32.452 ms
 8  adm-bb2-pos6-0-0.telia.net (213.248.65.158)  39.836 ms adm-bb1-pos6-0-0.telia.net (213.248.65.150)  43.489 ms  43.314 ms
 9  adm-b4-link.telia.net (80.91.250.66)  43.338 ms adm-b4-link.telia.net (80.91.253.86)  39.976 ms adm-b4-link.telia.net (80.91.253.82)  39.699 ms
10  google-118151-adm-b4.c.telia.net (213.248.72.14)  54.257 ms google-ic-127680-adm-bb2.c.telia.net (213.248.91.54)  48.193 ms google-ic-126116-adm-b4.c.telia.net (80.239.193.182)  45.820 ms
11  209.85.251.12 (209.85.251.12)  52.971 ms 209.85.251.14 (209.85.251.14)  56.089 ms  53.862 ms
12  64.233.175.246 (64.233.175.246)  51.057 ms 72.14.233.114 (72.14.233.114)  49.855 ms 64.233.175.246 (64.233.175.246)  48.394 ms
13  209.85.255.23 (209.85.255.23)  58.856 ms 216.239.46.51 (216.239.46.51)  49.990 ms 72.14.233.79 (72.14.233.79)  48.822 ms
14  209.85.249.133 (209.85.249.133)  57.250 ms 216.239.43.30 (216.239.43.30)  61.987 ms  65.345 ms
15  nf-in-f103.google.com (64.233.183.103)  61.596 ms  61.217 ms  60.348 ms

# traceroute6 ipv6.google.com
traceroute to ipv6.l.google.com (2001:4860:0:1001::68) from 2001:b18::211:43ff:fefd:90f6, 30 hops max, 16 byte packets
 1  ge-1.Edge1.ip6.Lisbon.NFSi.pt (2001:b18::ffff)  0.346 ms  0.31 ms  0.234 ms
 2  2001:5a0:1500::1 (2001:5a0:1500::1)  0.729 ms  0.456 ms  0.358 ms
 3  2001:5a0:1500::e (2001:5a0:1500::e)  27.217 ms  27.205 ms  27.213 ms
 4  2001:5a0:c00:100::e (2001:5a0:c00:100::e)  32.949 ms  36.937 ms  32.967 ms
 5  2001:5a0:200::5 (2001:5a0:200::5)  68.051 ms  37.071 ms  37.088 ms
 6  pr61.ams04.net.google.com (2001:7f8:1::a501:5169:1)  33.581 ms  36.939 ms  33.709 ms
 7  * * *
 8  * * *
 9  * * *
10  * * *
11  * 2001:4860:0:1001::68 (2001:4860:0:1001::68)  39.995 ms  36.551 ms

Regards,
---
Nuno Vieira
nfsi telecom, lda.

nuno.vieira at nfsi.pt
Tel. (+351) 21 949 2300 - Fax (+351) 21 949 2301
http://www.nfsi.pt/

----- "Carlos Friacas" <cfriacas at fccn.pt> wrote:

> On Wed, 29 Oct 2008, Gert Doering wrote:
> 
> > To be a bit more constructive: you could maintain a list of AS
> numbers
> > that belong to known research networks and should be reached via
> Geant2,
> > and local-pref these.  Anything else coming in via Geant2 should not
> be
> > local-prefed (don't necessarily *drop*, even if that might be
> prudent given
> > some of the crap I2 is leaking, but at least do not *force*) - which
> would
> > avoid routing Google traffic to Hongkong, instead of Amsterdam.
> 
> Hello,
> 
> I can say that from my small corner, something has really improved.
> 
> In Abril, when ipv6.google.com was launched, the avg latency from my 
> network was about 140% higher when comparing to www.google.com.
> 
> *Now* it's about 10% *lower*! (about ~5ms in absolute values). So,
> someone 
> tweaked something, somewhere regarding v6 routing :-)
> 
> IMHO, this also may indicate that there is still something along the 
> path in the v4 side which may need some tunning... ;-)
> 
> 
> Best Regards,
> Carlos
> (Portuguese NREN)