Routing issues

Forums › 2degrees (including Slingshot, Orcon, Flip, Stuff Fibre, MyRepublic, 2talk and Vocus) › Routing issues

1 | 2

Thepilgrim

13 posts

Geek

#469230 15-May-2011 10:33

[Edit]

; <<>> DiG 9.7.3 <<>> @ns2 cdn.geekz1.com a
; (1 server found)
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 24603
;; flags: qr rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 4, ADDITIONAL: 0

;; QUESTION SECTION:
;cdn.geekz1.com. IN A

;; ANSWER SECTION:
cdn.geekz1.com. 3587 IN CNAME nzcdn.geekz1.com.
nzcdn.geekz1.com. 3587 IN A 202.175.128.164

;; AUTHORITY SECTION:
geekz1.com. 172787 IN NS c.dxmx.com.
geekz1.com. 172787 IN NS d.dxmx.com.
geekz1.com. 172787 IN NS b.dxmx.com.
geekz1.com. 172787 IN NS a.dxmx.com.

;; Query time: 2 msec
;; SERVER: 192.168.0.6#53(192.168.0.6)
;; WHEN: Sun May 15 10:36:03 2011
;; MSG SIZE rcvd: 137

[/Edit] - this server had forwarders added this morning (after your recommendation) forgot to restart bind after removing the config

GoodSync

GoodSync. Easily back up and sync your files with GoodSync. Simple and secure file backup and synchronisation software will ensure that your files are never lost (affiliate link).

freitasm

BDFL - Memuneh
79042 posts

Uber Geek

Administrator

ID Verified

Trusted

Geekzone

Lifetime subscriber

#469255 15-May-2011 12:14

*sigh*

Just because we are talking about CDN, someone sent me an anonymous email, with an invalid email address with some confused information about CDN usage, DNS and more.

Obviously the person didn't want to actually get involved in any meaningful discussion, otherwise a valid email address would be used.

Trolls...

My technology disclosure

Thepilgrim

13 posts

Geek

#469264 15-May-2011 12:32

fun and games - thank you for your help this morning btw

I've re-added forwarding configuration to both name servers - in theory if it's not in my Zones it'll use slingshot from now on to resolve hosts...I've not seen any change.

I do however notice massive latency spikes when i hit asianetcom.net

e.g.

www.facebook.com
2 8 ms 7 ms 7 ms p32.argent.cust.commverge.net.nz [202.180.81.32]
3 8 ms 7 ms 8 ms vlan93-cpcak3-e1.tranzpeer.net [202.180.76.68]
4 8 ms 7 ms 8 ms vlan7-cpcak3-s1.tranzpeer.net [202.180.81.49]
5 8 ms 8 ms 8 ms ge-1-0-0-401.gw1.akl2.asianetcom.net [203.192.130.193]
6 134 ms 132 ms 132 ms ip-203-192-185-49.asianetcom.net [203.192.185.49]
7 133 ms 133 ms 132 ms facebook.com.any2ix.coresite.com [206.223.143.161]
8 141 ms 141 ms 151 ms ge-3-0-0.br01.sjc.tfbnw.net [204.15.23.0]
9 168 ms 168 ms 168 ms ae10.bb01.prn1.tfbnw.net [74.119.76.69]
10 165 ms 166 ms 165 ms ae0.dr02.prn1.tfbnw.net [74.119.79.103]
11 169 ms 168 ms 165 ms po1022.csw01a.prn1.tfbnw.net [74.119.79.51]
12 165 ms 165 ms 165 ms www-12-01-prn1.facebook.com [69.171.224.13]

24.media.tumblr.com
2 8 ms 8 ms 7 ms p32.argent.cust.commverge.net.nz [202.180.81.32]
3 8 ms 8 ms 8 ms vlan93-cpcak3-e1.tranzpeer.net [202.180.76.68]
4 8 ms 7 ms 8 ms vlan7-cpcak3-s1.tranzpeer.net [202.180.81.49]
5 8 ms 8 ms 8 ms ge-1-0-0-401.gw1.akl2.asianetcom.net [203.192.130.193]
6 132 ms 132 ms 132 ms ip-203-192-185-49.asianetcom.net [203.192.185.49]
7 134 ms 133 ms 135 ms edgecast.com.any2ix.coresite.com [206.223.143.112]
8 133 ms 134 ms 132 ms 68.232.44.19

Ragnor

8196 posts

Uber Geek

Trusted

#469616 16-May-2011 13:16

120-200ms is entirely normal for US based hops in routes from NZ, traffic has to cross the pacific!

http://en.wikipedia.org/wiki/Southern_Cross_Cable

Regs

4064 posts

Uber Geek

Trusted

Snowflake

#469915 17-May-2011 00:12

not sure what the current situation is, but after the japan tsunami/earthquake pa cnet/asianetcom lost one or two of their USJAPAN links. As a result they started routing traffic via alternate routes, the one via SYDJAPAN being one of them. this no doubt caused a bit more stress on the SYD POP and routers so could potentially explain traffic anomolies there.

prior to tsunami, my traffic for SIN used to transit via Los Angeles and Narito(Japan). Now it goes via Sydney and Narito(Japan) instead.

Sales Engineer
Snowflake
www.snowflake.com

about.me/nzregs
Twitter: @nzregs

Thepilgrim

13 posts

Geek

#471259 19-May-2011 19:13

tbh I just don't know enough about internet routing (really need to sit down a read more!)

I've switched back to my original config (dns pointing towards root and not forwarding through to slingshot) as I didn't see any huge difference, however as stated before I see "TCP Previous segment lost" for almost every web request (response from destination seq is a off) - however...if i leave the pages - they do load...hours later, no errors, no css issues - they load and when i click a link, we do this slow dance again.

I can't sniff off of my ADSL router, however I will be trying another one out this weekend if I have time, but maybe someone can explain the very long load time for a single website?

Thepilgrim

13 posts

Geek

#472046 22-May-2011 00:28

So...an hour wait time - one conversation later and it's all fixed.

Not sure how but i did learn the following things:

I'm on EUBA0, there was issues with this service, those are now fixed (historically fixed)
AYCE is currently working well (so sayith the service desk)
MTU wasn't the issue
My router wasn't the issue
IP address allocation is for a set period of time (loooong lease times)
I've only had three IP addresses in since early December 2010
New IP address allocated = fixed

I was fortunate to talk with someone who knew a thing or two - annoyed at the wait time, happy with the outcome.

Thanks for your help :)

1 | 2