Skip to main content
Beantwoord

Severe Bad Routing/Peering

  • January 3, 2026
  • 3 reacties
  • 43 Bekeken

I have noticed a severe increase in ping related to bad routing/peering to Singaporean servers in specific, in the past this would be around 170ms ping but now it’s up to 340ms, using a VPN will sometimes help as they have a better route and this does result in the expected ping, however, ideally I wouldn’t need to use a VPN to get the expected results.

See below the traceroute without a VPN:

Tracing route to 104.225.253.226 over a maximum of 30 hops

1 4 ms 2 ms 2 ms 192.168.1.1
2 5 ms 5 ms 5 ms 1-194-212-87.ftth.glasoperator.nl [87.212.194.1]
3 6 ms 6 ms 6 ms 10.227.136.129
4 6 ms 7 ms 6 ms 10.226.11.40
5 7 ms 7 ms 7 ms adm-bb1-link.ip.twelve99.net [62.115.142.210]
6 6 ms 6 ms 6 ms adm-b10-link.ip.twelve99.net [80.239.193.252]
7 7 ms 7 ms 6 ms adm-bb1-link.ip.twelve99.net [62.115.142.210]
8 16 ms 17 ms 22 ms prs-bb1-link.ip.twelve99.net [62.115.134.97]
9 89 ms 89 ms 90 ms rest-bb1-link.ip.twelve99.net [62.115.140.105]
10 96 ms 94 ms 95 ms ash-b2-link.ip.twelve99.net [62.115.123.123]
11 * * * Request timed out.
12 95 ms 94 ms 95 ms port-channel8481.ccr92.dca04.atlas.cogentco.com [154.54.25.17]
13 88 ms 88 ms 89 ms be4188.ccr42.jfk02.atlas.cogentco.com [154.54.30.122]
14 95 ms 95 ms 96 ms be4077.ccr22.alb02.atlas.cogentco.com [154.54.90.102]
15 97 ms 95 ms 96 ms be2104.ccr22.ymq01.atlas.cogentco.com [154.54.43.21]
16 97 ms 95 ms 95 ms be2090.rcr21.ymq02.atlas.cogentco.com [154.54.45.118]
17 * * * Request timed out.
18 * * * Request timed out.
19 * * * Request timed out.
20 * * * Request timed out.
21 103 ms 102 ms 303 ms be106.bhs-g1-nc5.qc.ca [142.44.208.173]
22 104 ms 101 ms 102 ms nyc-ny1-sbb1-8k.nj.us [192.99.146.127]
23 108 ms 108 ms 108 ms lon-thw-sbb1-nc5.uk.eu [54.36.50.240]
24 108 ms 108 ms 108 ms be105.lil1-rbx8-sbb1-nc5.fr.eu [54.36.50.245]
25 112 ms 111 ms 112 ms par-th2-sbb1-nc5.fr.eu [54.36.50.226]
26 115 ms 115 ms 113 ms par-dpa1-sbb1-8k.fr.eu [57.128.121.10]
27 125 ms 127 ms 123 ms mrs-mrs2-sbb1-nc5.fr.eu [57.128.121.185]
28 * * * Request timed out.
29 348 ms 345 ms 345 ms sin-sgcs2-g2-nc5.sgp.asia [103.5.15.17]
30 * * * Request timed out.


Traceroute with VPN:​

Tracing route to 104.225.253.226 over a maximum of 30 hops

1 18 ms 24 ms 28 ms 10.0.4.1
2 22 ms 23 ms 25 ms 169.254.169.254
3 20 ms 18 ms 26 ms 100.100.100.1
4 25 ms 27 ms 17 ms 10.76.1.133
5 25 ms 26 ms 27 ms 10.76.2.1
6 28 ms 17 ms 23 ms ce-2-5-2.a00.parsfr04.fr.bb.gin.ntt.net [81.25.197.209]
7 18 ms 18 ms 26 ms ae-2.r23.parsfr04.fr.bb.gin.ntt.net [129.250.5.132]
8 38 ms 45 ms 40 ms ae-1.r22.mlanit02.it.bb.gin.ntt.net [129.250.4.246]
9 * * * Request timed out.
10 * * * Request timed out.
11 * * * Request timed out.
12 176 ms 175 ms 175 ms sin-sgcs2-g2-nc5.sgp.asia [103.5.15.17]
13 * * * Request timed out.
14 * * * Request timed out.
15 * * * Request timed out.
16 * * * Request timed out.
17 181 ms 184 ms 174 ms ns564563.ip-139-99-68.net [139.99.68.65]
18 181 ms 174 ms 174 ms 104.225.253.226

Trace complete.

​​​​​​

Beste antwoord door Pieter_B

Welcome to the World Wide Web, it’s almost impossible to ‘steer’ the hops your request makes.

A VPN could improve it somehow, but that’s also being lucky.

It looks the address you referring to is a mail relay/server, because it is advertising itself with smtp on port 25.

The problem could be temporary due to an update or maintenance extern of Odido’s network, just give it some time if you didn’t do that already.

Bekijk origineel

3 reacties

Pieter_B
Super User
Forum|alt.badge.img+4
  • Odido Superuser
  • Antwoord
  • January 4, 2026

Welcome to the World Wide Web, it’s almost impossible to ‘steer’ the hops your request makes.

A VPN could improve it somehow, but that’s also being lucky.

It looks the address you referring to is a mail relay/server, because it is advertising itself with smtp on port 25.

The problem could be temporary due to an update or maintenance extern of Odido’s network, just give it some time if you didn’t do that already.


  • Auteur
  • heeft eerste post geplaatst
  • January 4, 2026

I am aware that it is not possible to manually adjust the route the traffic takes and ensure it goes down the most optimal route, however, this is not to say that what is happening here is just a temporary problem, it has been an ongoing issue for well over a year, with recently it becoming far worse than before (jumping from 220-260ms to 340ms).

The address I used is simply an example, it is just a VPS server with nothing running off of it, it was merely to show the bad routing, this is just one of hundreds of addresses where the routing is very poor.

The only addresses that do route correctly from what I have seen recently are a few from Cloudflare, however, it seems that a traceroute to these addresses results in a loopback after a couple hops, so not very useful for data.
 


rvk01
  • January 4, 2026

KPN also routes this IP through gin.ntt.net (NTT Communications) as first external hop, and not through twelve99.net (Arelion) like Odido does. That is something that Odido could control. And gin.ntt.net routes this address differently.