rssLink RSS for all categories
 
icon_red
icon_green
icon_red
icon_red
icon_blue
icon_green
icon_green
icon_red
icon_red
icon_red
icon_orange
icon_green
icon_green
icon_green
icon_green
icon_blue
icon_red
icon_orange
icon_red
icon_red
icon_red
icon_red
icon_green
icon_red
icon_red
icon_red
icon_red
icon_orange
icon_green
 

FS#310 — FS#4341 — Telia (Paris) 10G

Attached to Project— Network
Maintenance
Whole Network
CLOSED
100%
The 10G port between Telia and Paris is down. There no consequence on the traffic flow, the traffic was automatically rerouted by Amsterdam (GBLX),
however, the ingoing traffic is arriving from Madrid. Sometimes it could influence the latency time of some networks of North Europe:
http://smokeping.ovh.net/ovh-server-statistics/show.cgi?target=PING.AS21211

But I believe that some AS are sending us the traffic via Telia
http://smokeping.ovh.net/ovh-server-statistics/show.cgi?target=PING.AS8612
http://smokeping.ovh.net/ovh-server-statistics/show.cgi?target=PING.AS5089
http://smokeping.ovh.net/ovh-server-statistics/show.cgi?target=PING.AS16338

that's to expect.
Date:  Saturday, 17 July 2010, 02:02AM
Reason for closing:  Done
Comment by OVH - Saturday, 17 July 2010, 01:31AM

We had blocked OVH's notification at Tiscali via Seabone.
http://smokeping.ovh.net/ovh-server-statistics/show.cgi?target=PING.AS8612


Comment by OVH - Saturday, 17 July 2010, 01:34AM

We had blocked OVH's notification via Seabone towards CW.
That did not change anything. That is not probably the source.

We had redirected the traffic instead of GBLX/CW/AUNA (3549 1273 16338) via TATA/LEVEL3/AUNA (6453 3356 16338).
We are back to 31ms.
http://smokeping.ovh.net/ovh-server-statistics/show.cgi?target=PING.AS16338


Comment by OVH - Saturday, 17 July 2010, 01:36AM

There is not peering between Seabone and Virgine. It is no longer Tiscali. With the CW, the problem seems to be resolved
http://smokeping.ovh.net/ovh-server-statistics/show.cgi?target=PING.AS5089


Comment by OVH - Saturday, 17 July 2010, 01:39AM

64 bytes from www.virginmedia.com (212.250.162.12): icmp_seq=56 ttl=114 time=35.5 ms
64 bytes from www.virginmedia.com (212.250.162.12): icmp_seq=57 ttl=114 time=35.7 ms
64 bytes from www.virginmedia.com (212.250.162.12): icmp_seq=58 ttl=48 time=81.6 ms
64 bytes from www.virginmedia.com (212.250.162.12): icmp_seq=59 ttl=114 time=17.0 ms
64 bytes from www.virginmedia.com (212.250.162.12): icmp_seq=60 ttl=114 time=17.4 ms

The CW was causing the traffic problem. It is fixed now.


Comment by OVH - Saturday, 17 July 2010, 01:42AM

However, it seems that a bug set community 1299:2849 was announcing all notifications instead of only 1299:284x Cable & Wireless DE/1273 1299:584x Cable & Wireless DE/1273 -
http://www.onesc.net/communities/as1299/

hence the slight increase of latency on ONO:
http://smokeping.ovh.net/ovh-server-statistics/show.cgi?target=PING.AS6739


Comment by OVH - Saturday, 17 July 2010, 01:46AM

The port is UP, but the BGP session is down. From Telia's side, the engineer believes that the port is DOWN. It seems that someone has disconnect us from the router and reconnected on the bad one...


Comment by OVH - Saturday, 17 July 2010, 01:50AM

The community problem is fixed. But the traffic is again coming from Telia.

Via what?

No traceroute nor looking glass upon Virgin. we are doing it manually
http://smokeping.ovh.net/ovh-server-statistics/show.cgi?target=PING.AS5089

BT AS5400 ?
1299:260x BT/Concert/5400 - -
non

LINX ?
1299:208x Peers at LINX -
Mieux

....

20 Minutes later. Visibly Virgin gars have forced the traffic via Telia.

3 osr02hari-ge422.network.virginmedia.net (81.96.16.134) 40.724 ms 42.704 ms 44.932 ms
4 popl-bb-1b-ae1-0.network.virginmedia.net (195.182.178.169) 47.914 ms 49.894 ms 51.626 ms
5 leed-bb-1a-as1-0.network.virginmedia.net (62.253.185.237) 58.600 ms 60.580 ms 68.059 ms
6 tcl3-ic-1-ae0-0.network.virginmedia.net (212.43.163.198) 69.787 ms 43.913 ms 46.640 ms
7 ldn-b2-link.telia.net (213.248.93.69) 48.566 ms 49.795 ms 51.776 ms
8 ldn-bb1-link.telia.net (80.91.250.225) 53.759 ms 55.740 ms 57.716 ms
9 prs-bb1-link.telia.net (80.91.247.254) 66.444 ms 69.424 ms 70.653 ms
10 mad-b1-link.telia.net (80.91.254.133) 91.684 ms 91.685 ms 75.190 ms
11 10g.mad-1-6k.routers.chtix.eu (213.251.130.17) 67.192 ms * *
12 20g.gsw-2-6k.routers.ovh.net (213.251.128.41) 77.940 ms * *
13 80g.p19-2-6k.routers.chtix.eu (213.186.32.153) 67.942 ms * *
14 * * *
15 ping.ovh.net (213.186.33.13) 76.939 ms 73.690 ms 72.939 ms

We cancel OVH's notifications via Telia.

3 osr02hari-ge422.network.virginmedia.net (81.96.16.134) 41.502 ms 42.472 ms 44.705 ms
4 popl-bb-1b-ae1-0.network.virginmedia.net (195.182.178.169) 47.685 ms 49.666 ms 51.648 ms
5 popl-bb-1a-ae0-0.network.virginmedia.net (213.105.174.229) 53.624 ms 55.606 ms 56.586 ms
6 amst-ic-1-as0-0.network.virginmedia.net (213.105.175.6) 68.564 ms 41.907 ms 43.883 ms
7 amsix.routers.ovh.net (195.69.145.231) 46.807 ms * 50.019 ms
8 40g.rbx-2-6k.routers.chtix.eu (91.121.131.9) 56.001 ms * *
9 160g.th2-1-6k.routers.chtix.eu (213.186.32.202) 66.935 ms * *
10 80g.p19-2-6k.routers.chtix.eu (213.186.32.150) 68.130 ms * *
11 10g.p19-57-6k.routers.chtix.eu (213.251.128.34) 50.949 ms * 10g.p19-52-6k.routers.chtix.eu (213.186.32.230) 49.698 ms
12 ping.ovh.net (213.186.33.13) 49.946 ms 49.696 ms 50.697 ms


Comment by OVH - Saturday, 17 July 2010, 02:01AM

Some ONO customers in ES are forwarding problems from Valencia 81.184.0.X

netname: MENTA-CABLEMODEMS
descr: AUNA - CTC
descr: Internet de Banda Ampla
descr: CableModem

I have heard about network migration that ONO bought from South Spain ... It Might be this. It is then, AUNA which has already migrated towards the ONO network (from AS16338 to l'AS6739). They are not all working properly.

10 minutes later:

got something that we do not see everyday:

th2-1-6k#sh ip bgp 81.184.0.0
3549 1299 6739
3320 3356 16338

The concerned IP band is announced via 2 different networks !
We will attempt to route 81.184.0.0 via 3320 3356 16338

5 minutes later

# ping 81.184.0.99
PING 81.184.0.99 (81.184.0.99) 56(84) bytes of data.
64 bytes from 81.184.0.99: icmp_seq=1 ttl=240 time=50.8 ms
64 bytes from 81.184.0.99: icmp_seq=2 ttl=240 time=50.5 ms

--- 81.184.0.99 ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 1002ms
rtt min/avg/max/mdev = 50.597/50.720/50.844/0.256 ms

So nice this attempt :) it is pinging ! :)

CONCLUSION:
ONO has then problems in their notifications and problems between their internal AS.

Ping is important. Could we ameliorate it?
1 p19-52-6k.routers.ovh.net (213.186.33.159) 0.441 ms * 0.448 ms
2 10G.p19-7-6k.routers.ovh.net (213.186.32.229) 0.588 ms * 72.060 ms
3 80G.th1-1-6k.routers.ovh.net (213.186.32.134) 0.571 ms * 1.114 ms
4 80G.th2-1-6k.routers.ovh.net (213.186.32.166) 10.702 ms 22.720 ms *
5 20G.dtag.th2-1-6k.routers.ovh.net (213.186.32.206) 1.503 ms 0.461 ms 1.368 ms
6 ams-sa2-i.AMS.NL.NET.DTAG.DE (62.154.5.65) 10.969 ms 10.568 ms 11.070 ms
7 194.25.211.34 (194.25.211.34) 14.077 ms 74.705 ms 10.467 ms
8 ae-33-51.ebr1.Amsterdam1.Level3.net (4.69.139.129) 10.681 ms 10.806 ms 10.590 ms
9 ae-47-47.ebr2.Dusseldorf1.Level3.net (4.69.143.206) 16.442 ms ae-45-45.ebr2.Dusseldorf1.Level3.net (4.69.143.198) 14.319 ms 14.193 ms
10 ae-1-100.ebr1.Dusseldorf1.Level3.net (4.69.141.149) 20.076 ms 13.794 ms 13.583 ms
11 ae-47-47.ebr2.Frankfurt1.Level3.net (4.69.143.174) 18.553 ms ae-45-45.ebr2.Frankfurt1.Level3.net (4.69.143.166) 14.134 ms ae-48-48.ebr2.Frankfurt1.Level3.net (4.69.143.178) 13.773 ms
12 ae-72-72.csw2.Frankfurt1.Level3.net (4.69.140.22) 19.525 ms 16.398 ms ae-82-82.csw3.Frankfurt1.Level3.net (4.69.140.26) 22.700 ms
13 ae-81-81.ebr1.Frankfurt1.Level3.net (4.69.140.9) 14.798 ms ae-91-91.ebr1.Frankfurt1.Level3.net (4.69.140.13) 14.113 ms ae-71-71.ebr1.Frankfurt1.Level3.net (4.69.140.5) 15.558 ms
14 ae-2-8.bar1.Barcelona1.Level3.net (4.69.141.57) 37.314 ms 37.107 ms 37.461 ms
15 213.242.114.6 (213.242.114.6) 35.808 ms 35.615 ms 35.809 ms

Il y a moyen:
6453 3356 16338
5511 3356 16338
6762 3356 16338
3320 3356 16338
3549 1299 6739
1299 6739

Essayons OTI:
# ping 81.184.0.99
PING 81.184.0.99 (81.184.0.99) 56(84) bytes of data.
64 bytes from 81.184.0.99: icmp_seq=1 ttl=241 time=38.9 ms
64 bytes from 81.184.0.99: icmp_seq=2 ttl=241 time=40.0 ms

--- 81.184.0.99 ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 1008ms
rtt min/avg/max/mdev = 38.954/39.504/40.054/0.550 ms

Cool :)

How to lose 2 hours ... then it works :)