On 30/11/2017, at 11:15 PM, Nathan Ward wrote:
I maybe missing on some last century/decade history, but what's the current
story with Spark/Vodafone public peering in Australia but not in New
Zealand? With <somebody> has stopped paid(?) peering with these guys
recently, is it a today's case that rest of the country talks to Spark
(government owned?)/Vodafone via 50ms Sydney path and everybody is happy
with it?
If you’d like some operational feedback, give some examples of where you’re
seeing this. Source/dest/traceroutes/etc. etc. are all useful.
Here's an example of what I'm thinking is maybe sub-optimal
peering/routing, albeit at least in-country...
My example traffic is sourced from a VF UFB static IP and heading to 2talk,
but appears to be crossing the Spark network?!
With all the industry consolidation these days it's hard to keep up. Is
this expected behaviour these days, or something I should escalate?
And presumably, if I escalate, it should be to 2talk not VF because the
reverse direction seems to be OK?
ladmin(a)fpl-rpi32 ~ $ mtr -n -c 2 sip.2talk.co.nz
My traceroute [v0.85]
fpl-rpi32 (0.0.0.0)
Tue Dec 12 07:21:29 2017
Keys: Help Display mode Restart statistics Order of fields quit
Packets
Pings
Host Loss% Snt
Last Avg Best Wrst StDev
1. 192.168.1.1 0.0% 2
1.2 1.1 1.0 1.2 0.0
2. 210.246.31.254 0.0% 2
52.5 44.7 37.0 52.5 11.0
3. 203.167.245.102 0.0% 2
5.0 4.3 3.6 5.0 1.0
4. 203.167.245.101 0.0% 2
3.3 4.7 3.3 6.0 1.7
5. 203.98.18.65 0.0% 2
14.6 14.2 13.9 14.6 0.0
6. 203.98.18.66 0.0% 2
16.2 16.0 15.8 16.2 0.0
7. 122.56.127.210 0.0% 2
14.3 14.1 13.8 14.3 0.0
8. 122.56.118.114 0.0% 2
14.1 14.8 14.1 15.5 1.0
9. 114.31.203.130 0.0% 2
14.4 13.9 13.4 14.4 0.0
10. 27.111.14.77 0.0% 2
16.5 14.8 13.2 16.5 2.2
11. 27.111.14.65 0.0% 2
14.8 15.1 14.8 15.5 0.0
Once more with DNS on:
ladmin(a)fpl-rpi32 ~ $ mtr -c 2 sip.2talk.co.nz
My traceroute [v0.85]
fpl-rpi32 (0.0.0.0)
Tue Dec 12 07:20:21 2017
Keys: Help Display mode Restart statistics Order of fields quit
Packets
Pings
Host Loss% Snt
Last Avg Best Wrst StDev
1. 192.168.1.1 0.0% 2
1.0 1.1 1.0 1.3 0.0
2. 210-246-31-254.dsl.dyn.ihug.co.nz 0.0% 2
7.9 6.3 4.6 7.9 2.2
3. 203.167.245.102 0.0% 2
5.9 5.9 5.9 5.9 0.0
4. 203.167.245.101 0.0% 2
3.7 4.0 3.7 4.4 0.0
5. 203.98.18.65 0.0% 2
15.1 15.6 15.1 16.1 0.0
6. 203.98.18.66 0.0% 2
15.9 15.4 14.9 15.9 0.0
7. ae10-44.tkcr5.global-gateway.net.nz 0.0% 2
16.1 14.6 13.1 16.1 2.0
8. vocus-dom.tkcr5.global-gateway.net.nz 0.0% 2
14.4 14.9 14.4 15.5 0.0
9. as55561.cust.bdr01.alb01.akl.VOCUS.net.nz
http://alb01.akl.vocus.net.nz/ 0.0% 2 14.5 16.2
14.5 17.8 2.2
10. 27-111-14-77.2talk.co.nz 0.0% 2
13.6 14.2 13.6 14.8 0.0
11. phone.nzpbx.com 0.0% 2
15.7 14.9 14.1 15.7 1.0
Reverse direction appears OK (akl-ix), but visibility from my endpoint also
isn't great:
root(a)2tvm:~# mtr -c 2 210.246.28.87
My traceroute [v0.80]
2tvm.fiberphone.co.nz (0.0.0.0) Tue Dec
12 09:43:09 2017
Keys: Help Display mode Restart statistics Order of fields quit
Packets
Pings
Host Loss% Snt Last Avg
Best Wrst StDev
1. 103-29-30-1.static.2talk.co.nz 0.0% 2 0.9 2.3
0.9 3.7 2.0
2. ???
3. as55561.akl-ix.nz 0.0% 2 1.1 0.9
0.8 1.1 0.2
4. 43.243.21.64 0.0% 2 1.8 1.9
1.8 2.1 0.2
5. ???
Or am I just confused and this is all optimal?
-
Pete Mundy
