BT Bad routing

Permabanned
Joined
28 Dec 2009
Posts
13,052
Location
london
I am having a problem with slow communication to my dedicated server.

Communication from my home pc to the dedicated server in france is ok. But communication from dedicated server to my home pc fails.

traceroute to *homeip* (*homeip*), 64 hops max, 52 byte packets
1 rbx-3-m1.fr.eu (213.251.187.253) 0.659 ms 0.409 ms 0.532 ms
2 rbx-2-6k.fr.eu (213.251.191.130) 0.602 ms 0.375 ms *
3 rbx-g2-a9.fr.eu (91.121.131.10) 2.065 ms 2.621 ms 1.958 ms
4 gsw-g1-a9.fr.eu (91.121.215.150) 4.228 ms 4.556 ms 4.626 ms
5 gsw-1-6k.fr.eu (213.186.32.158) 7.452 ms 8.860 ms 8.363 ms
6 gblx.as3549.fr.eu (213.186.32.198) 6.494 ms 6.950 ms 8.570 ms
7 ae2.scr4.CDG2.gblx.net (67.16.129.194) 45.665 ms 7.629 ms 7.516 ms
8 * xe8-2-0-10G.scr4.LON3.gblx.net (67.16.165.230) 13.222 ms 10.791 ms
9 lag2.ar9.LON3.gblx.net (67.17.106.149) 11.196 ms 13.009 ms 12.896 ms
10 bt-group-plc.ethernet4-4.ar9.lon3.gblx.net (64.214.144.218) 15.461 ms 16.126 ms 18.206 ms
11 166-49-168-94.eu.bt.net (166.49.168.94) 17.889 ms 16.194 ms 16.860 ms
12 * 62.172.103.140 (62.172.103.140) 22.814 ms 24.798 ms
13 host213-121-193-53.ukcore.bt.net (213.121.193.53) 21.753 ms 21.724 ms 23.974 ms
14 acc1-10GigE-0-2-0-0.l-far.21cn-ipp.bt.net (109.159.249.24) 17.762 ms 18.173 ms 19.209 ms
15 * 109.159.249.100 (109.159.249.100) 17.701 ms
109.159.249.98 (109.159.249.98) 18.443 ms
16 217.41.168.106 (217.41.168.106) 19.241 ms 18.977 ms 16.687 ms
17 217.41.168.76 (217.41.168.76) 17.740 ms 19.159 ms 19.184 ms
18 213.120.177.41 (213.120.177.41) 33.252 ms
213.120.156.65 (213.120.156.65) 18.444 ms
213.120.177.17 (213.120.177.17) 19.459 ms
19 * * *
20 * * *
21 * * *
22 * * *
23 * * *
24 * * *
25 * * *
26 * * *
27 * * *
28 * * *
29 * * *
30 * * *
31 * * *
32 * * *
33 *

While the other way is fine:

1 217.32.146.69 (217.32.146.69) 6.292 ms 6.542 ms 6.358 ms
2 217.32.146.94 (217.32.146.94) 6.447 ms 6.445 ms 6.476 ms
3 213.120.156.66 (213.120.156.66) 6.726 ms 7.051 ms 6.965 ms
4 217.41.168.77 (217.41.168.77) 7.208 ms 7.293 ms 6.993 ms
5 217.41.168.107 (217.41.168.107) 7.468 ms 7.559 ms 7.228 ms
6 acc1-10GigE-0-7-0-4.l-far.21cn-ipp.bt.net (109.159.249.74) 7.188 ms
109.159.249.114 (109.159.249.114) 7.990 ms
acc1-10GigE-0-0-0-4.l-far.21cn-ipp.bt.net (109.159.249.66) 7.079 ms
7 core2-te0-7-0-5.faraday.ukcore.bt.net (109.159.249.15) 8.685 ms
core2-te0-7-0-4.faraday.ukcore.bt.net (109.159.249.11) 9.718 ms
core2-te0-0-0-7.faraday.ukcore.bt.net (109.159.249.7) 12.990 ms
8 host213-121-193-76.ukcore.bt.net (213.121.193.76) 17.956 ms 11.496 ms 11.734 ms
9 core2-pos5-0-0.telehouse.ukcore.bt.net (194.74.65.126) 7.949 ms 7.750 ms 8.230 ms
10 t2as2-tge1-4.uk-lon1.eu.bt.net (166.49.214.129) 7.967 ms 8.015 ms 7.751 ms
11 166-49-211-34.eu.bt.net (166.49.211.34) 7.959 ms 7.509 ms 7.774 ms
12 ldn-bb1-link.telia.net (80.91.246.144) 8.450 ms
ldn-bb2-link.telia.net (80.91.249.181) 8.011 ms
ldn-bb2-link.telia.net (80.91.246.146) 8.768 ms
13 adm-bb2-link.telia.net (80.91.250.126) 16.198 ms
adm-bb2-link.telia.net (80.91.253.146) 16.221 ms
adm-bb1-link.telia.net (213.155.130.90) 16.185 ms
14 adm-b4-link.telia.net (80.91.253.182) 35.621 ms 16.180 ms
adm-b4-link.telia.net (80.91.253.152) 16.331 ms
15 ams-1-6k.nl.eu (213.251.128.101) 24.319 ms * 24.318 ms
16 rbx-g1-a9.fr.eu (178.33.100.224) 25.238 ms 26.246 ms 24.987 ms
17 rbx-1-6k.fr.eu (94.23.122.105) 44.689 ms
rbx-2-6k.fr.eu (213.186.32.234) 24.216 ms
rbx-1-6k.fr.eu (94.23.122.105) 24.974 ms
18 rbx-3-m1.routers.chtix.eu (213.251.191.4) 24.200 ms
rbx-3-m1.fr.eu (213.251.191.132) 23.251 ms
rbx-3-m1.routers.chtix.eu (213.251.191.4) 23.744 ms
19 *dedicatedbox* (*dedicatedbox*) 23.960 ms 24.030 ms 23.711 ms


Any tips in fixing this that don't include a painful phone call to BT?
 
I have been using ovh for years without any problems. I have been getting a steady 4mbyte download from it since i got fiber.

Thanks for pointing out that my home pc is blocking icmp.

That makes sense. But what about the slow transfer speeds ? I get only 200kbyte a second?
 
Are you using ethernet on the HH3? There's a known issue with its ports where speeds end up ~ 1mbit, give or take, and one's network card reports a link speed of only 10Mb.

Temporarily fixed by rebooting the HH, I think.
 
No i use pfsense and the fiber modem only.

no i don't think it is icmp being blocked that is causing the traceroute timeout.

I enabled icmp from that host and did a test icmp and it works fine. Then reran the traceroute and it did the same thing. I definitely think there is a routing problem on bt network.
 
BT's routing is pretty naff tho, mine literally takes a leisurely stroll around the country before exiting onto the internet :( I live in South Somerset and at the moment its routing to London via Sheffield... at one point I caught it routing via Birmingham->Cardiff->Sheffield->London :(
 
No i use pfsense and the fiber modem only.

no i don't think it is icmp being blocked that is causing the traceroute timeout.

I enabled icmp from that host and did a test icmp and it works fine. Then reran the traceroute and it did the same thing. I definitely think there is a routing problem on bt network.

Trust me the IP.

EDIT:

Home IP - 31 packets transmitted, 0 received, 100% packet loss, time 30026ms
 
Last edited:
I only enabled icmp for my homeip from my dedicated box. if you want me to enable icmp i will need an ipaddress. What are you trying to do traceroute it ?

I guess i could enable icmp for the time being while we testing..

edit: i have enabled icmp from all ip.
 
I can download that file from my homeip at fullspeed 4.7mbyte sec
From my dedicated box i can full speed 10mbyte sec.

So both hosts have no throughput problems to other hosts. only to each other. As the routing problem, is within bt network it points to a bt routing issue.
 
I am aware that it is an ooh speed test file. This is means that the problem is not ovh related or even specific to my line. But a routing issue between my ovh box and and my home ip, specifically in the bt network.

This is why it is a realy odd problem.

Edit: this appears to have solved itself. Even though the traceroute is still doing the same thing, bandwidth is back to normal.
 
Last edited:
Back
Top Bottom