[atlas] 1msec RTT to ipv6.google.com?
- Previous message (by thread): [atlas] 1msec RTT to ipv6.google.com?
- Next message (by thread): [atlas] Probe downtime alert for IPv6?
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Abreu, Alexandre, Vodafone Portugal
alexandre.abreu at vodafone.com
Wed Jul 22 12:12:37 CEST 2015
Hi I've been reviewing both paths and indeed 2a00:1450:4004:801::200e is 1msec away from the probe (it is answered right at the internet exchange premises by either Google's router or a colocated server/cache) while 2a00:1450:4004:806::200e has to go out to Google DCs hence the ~12msec. Thank you all for your help. AA -----Original Message----- From: ripe-atlas [mailto:ripe-atlas-bounces at ripe.net] On Behalf Of Robert Kisteleki Sent: Tuesday, 21 de July de 2015 12:55 To: Roman Mamedov; Jen Linkova Cc: ripe-atlas at ripe.net; Tomáš Hetmer Subject: Re: [atlas] 1msec RTT to ipv6.google.com? > From the original message: > >> I've setup a ping probe (ID#2143865) to ipv6.google.com and it's consistently giving 1msec RTT >> RTT I get on the CPE the probe is connected to >> 0 2a00:1450:4003:806::200e 56 59 12ms echo reply > > My guess is that maybe the probe was set up to use a different DNS resolver > than the CPE, and it got a different IPv6 instead of 2a00:1450:4003:806::200e. Indeed the probe pinged 2a00:1450:4004:801::200e and not the address above. I have no idea if that's the root cause for this, but it's certainly a different target. It could be the same DNS resolver but it got different answers at different times? I set up one-off traceroutes 2144086 (ICMP) and 2144089 (UDP) and they seem quite reasonable to me (5 hops, ~1.2ms to target). Robert
- Previous message (by thread): [atlas] 1msec RTT to ipv6.google.com?
- Next message (by thread): [atlas] Probe downtime alert for IPv6?
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]