routed from london to manchester and back to london

Hello Zen users

Is anyone eles experiencing this issue also on the london gateway (lond1 and lond2) that there connection is been routed from London to Manchester and back to London with increasing latency? i find it affects the following;

Cloudfire e.g 1.1.1.1
Microsoft
iomart (e.g Rapidswitch)
M247 via London server

ISPs
TalkTalk (example IP from a core IP 78.144.1.11)
Virgin Media (IP example on a watford core 62.253.174.21)

Microsoft (wont let me ping the IP but routes are like so)

C:\Users\Francis>tracert 51.140.229.88

Tracing route to 51.140.229.88 over a maximum of 30 hops

1 <1 ms <1 ms <1 ms fritz.box [192.168.178.1]
2 5 ms 5 ms 5 ms vt1.cor1.lond1.ptn.zen.net.uk [51.148.72.21]
3 7 ms 5 ms 5 ms ae-3.pe2.thn-lon.zen.net.uk [51.148.73.12]
4 13 ms 12 ms 12 ms ae-13.cor2.manc1.ptn.zen.net.uk [51.148.73.25]
5 11 ms 11 ms 11 ms 195.66.244.82
6 13 ms 13 ms 17 ms ae23-0.icr01.lon22.ntwk.msn.net [104.44.40.157]
7 * * * Request timed out.
8 * * * Request timed out.
9 * * * Request timed out.
10 17 ms 17 ms 18 ms be-100-0.ibr01.lon22.ntwk.msn.net [104.44.21.87]
11 17 ms 16 ms 15 ms be-5-0.ibr02.cwl20.ntwk.msn.net [104.44.18.99]
12 16 ms 16 ms 16 ms ae122-0.icr02.cwl20.ntwk.msn.net [104.44.20.180]
13 * * * Request timed out.

Cloudflare for example

C:\Users\Francis>tracert cdn.discordapp.com

Tracing route to cdn.discordapp.com [162.159.133.233]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms fritz.box [192.168.178.1]
2 5 ms 5 ms 5 ms vt1.cor1.lond1.ptn.zen.net.uk [51.148.72.21]
3 6 ms 6 ms 6 ms ae-3.pe2.thn-lon.zen.net.uk [51.148.73.12]
4 13 ms 13 ms 13 ms ae-13.cor2.manc1.ptn.zen.net.uk [51.148.73.25]
5 16 ms 12 ms 12 ms ixmanchester.as13335.net [195.66.244.71]
6 13 ms 13 ms 12 ms 162.159.133.233

Trace complete.

I have contacted Zen technical support via email https://www.zen.co.uk/contact-us explainning that this has been going on for a month and a half and this is happening at night (checked at 3am) and also morning about 7am again routed to london > manchester > london again.

I was advised to power down my router for 5 - 10 minutes and then power it back on so done this today for 1 hour but didnt help,and still getting routed to Manchester and back.

Edited by francisuk25 (Fri 07-Aug-20 17:38:33)

Just the same here. It seems to be since problems on the Manchester gateway when they temporarily switched folk via London.
Tracing route to 51.140.229.88 over a maximum of 30 hops

1 <1 ms <1 ms <1 ms 192.168.1.254
2 7 ms 7 ms 7 ms losubs.subs.bng2.th-lon.zen.net.uk [62.3.80.21]
3 7 ms 8 ms 8 ms ae1-183.cr2.th-lon.zen.net.uk [62.3.86.82]
4 8 ms 7 ms 10 ms vl-50.ae-5.pe2.thn-lon.zen.net.uk [51.148.73.54]
5 16 ms 18 ms 15 ms ae-13.cor2.manc1.ptn.zen.net.uk [51.148.73.25]
6 14 ms 15 ms 15 ms 195.66.244.82
7 15 ms 15 ms 16 ms ae24-0.icr01.lon24.ntwk.msn.net [104.44.239.115]
8 * * * Request timed out.

In reply to a post by professor973:

Just the same here. It seems to be since problems on the Manchester gateway when they temporarily switched folk via London.

I have been on the London gateway ever since i joined zen on the 10th of June just gone and i dont think i ever had a routing problem and all my connections would terminate London what is great but now mostly my Gaming traffic would be round 12-14ms even if i choice a London server what was 7ms now added 12-14ms

https://www.zen.co.uk/blog/posts/zen-blog/2018/06/04...

Yes, same here (FTTP)

Having some dropped calls on work laptop using Skype for business (SFB) so traced route and queried it with tech support as wanted to see if I could get the latency down before I start talking to the people that manage the SFB servers.

They said there shouldn't be an issue (there shouldn't) but didn't answer why it wasn't going out for the connection to LINX in London which would make sense.

Completely unrelated but I've been tinkering with Sipgate and that traffic takes a more direct route to Germany.

Tracing route to sipgate.co.uk [217.10.79.23]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms fritz.box [192.168.8.1]
2 5 ms 8 ms 4 ms vt1.cor2.lond2.ptn.zen.net.uk [51.148.72.24]
3 5 ms 4 ms 5 ms ae-4.pe2.ixn-lon.zen.net.uk [51.148.73.110]
4 63 ms 6 ms 6 ms ae-1.pe2.thn-lon.zen.net.uk [51.148.73.96]
5 5 ms 4 ms 4 ms ae-1.br2.thn-lon.zen.net.uk [51.148.73.119]
6 5 ms 4 ms 4 ms linx.ae4-irt1.lon37.uk.as13237.net [195.66.224.99]
7 9 ms 9 ms 9 ms ae1.irt1.ams01.nl.as13237.net [82.197.136.22]
8 12 ms 12 ms 12 ms ae1-irt1.dus03.de.as13237.net [82.197.128.21]
9 13 ms 13 ms 12 ms Primacall-FRA-9-eth11-104.de.lambdanet.net [217.71.104.82]
10 13 ms 12 ms 12 ms sipgate.co.uk [217.10.79.23]

Trace complete.


I could perhaps understand if they were doing this to ensure any equipment likely to need quick maintenance is closer to home but surely this would involve switching traffic termination to Manchester?

Edited by techguy (Fri 07-Aug-20 20:02:55)

In reply to a post by techguy:

Completely unrelated but I've been tinkering with Sipgate and that traffic takes a more direct route to Germany.

Tracing route to sipgate.co.uk [217.10.79.23]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms fritz.box [192.168.8.1]
2 5 ms 8 ms 4 ms vt1.cor2.lond2.ptn.zen.net.uk [51.148.72.24]
3 5 ms 4 ms 5 ms ae-4.pe2.ixn-lon.zen.net.uk [51.148.73.110]
4 63 ms 6 ms 6 ms ae-1.pe2.thn-lon.zen.net.uk [51.148.73.96]
5 5 ms 4 ms 4 ms ae-1.br2.thn-lon.zen.net.uk [51.148.73.119]
6 5 ms 4 ms 4 ms linx.ae4-irt1.lon37.uk.as13237.net [195.66.224.99]
7 9 ms 9 ms 9 ms ae1.irt1.ams01.nl.as13237.net [82.197.136.22]
8 12 ms 12 ms 12 ms ae1-irt1.dus03.de.as13237.net [82.197.128.21]
9 13 ms 13 ms 12 ms Primacall-FRA-9-eth11-104.de.lambdanet.net [217.71.104.82]
10 13 ms 12 ms 12 ms sipgate.co.uk [217.10.79.23]

Trace complete.

No issues with Sipgate VoIP as this is a pretty stright forward to london to netherlands and into germany as mentioned.

francis@archlinuxlabs:~# traceroute sipgate.co.uk
traceroute to sipgate.co.uk (217.10.79.23), 30 hops max, 38 byte packets
1 192.168.178.1 (192.168.178.1) 0.357 ms 0.438 ms 0.233 ms
2 vt1.cor2.lond1.ptn.zen.net.uk (51.148.72.22) 5.754 ms 5.762 ms 5.992 ms
3 ae-4.pe1.thn-lon.zen.net.uk (51.148.73.0) 5.655 ms 5.977 ms 6.016 ms
4 ae-1.br2.thn-lon.zen.net.uk (51.148.73.119) 6.155 ms 5.799 ms ae-2.br2.thn-lon.zen.net.uk (51.148.73.115) 5.794 ms
5 linx.ae4-irt1.lon37.uk.as13237.net (195.66.224.99) 6.210 ms 5.715 ms 5.837 ms
6 ae1.irt1.ams01.nl.as13237.net (82.197.136.22) 11.174 ms 10.574 ms 10.637 ms
7 ae1-irt1.dus03.de.as13237.net (82.197.128.21) 13.516 ms 13.813 ms 20.983 ms
8 Primacall-FRA-9-eth11-104.de.lambdanet.net (217.71.104.82) 14.014 ms 13.989 ms 14.027 ms
9 * * *
10 * * *

Clouvider Limited
Hydra Communications Ltd
Vultr / Choopa LLC

all above have zero issues and routed into London without going into London > Manchester then back into London again.

So the ones below are getting re-routed into Manchester

Cloudfire e.g 1.1.1.1 or 1.0.0.1
Microsoft
iomart (e.g Rapidswitch)
M247 via London server

Edited by francisuk25 (Fri 07-Aug-20 23:04:01)

Yes indeed. That 6 ms is killing me.

I can't post the server I'm trying to connect to at work as I don't want someone to hit it with a DDOS (we've had those before) but I am seeing it, it actually gets handed off at what appears to be Zen's connection to LINX in Manchester to our ISP, it makes it's way back down to Telehouse then back up to Bucks.

Could any Zenistas shed a light on this?

Hi All,

I'm the Head of Network & Infrastructure at Zen and I'm responsible for Zen's network. I rarely post on here, but I can reassure I frequently read and investigate/act upon issues which are raised. I thought I'd quickly reach out specifically about this issue though.

We are aware of an ongoing routing issue which causes some traffic to take a suboptimal route through our network, my engineers have been working through this issue and have identified a fix. I expect the fix to be implemented today - I will further update when the work has been completed.

Richard

Richard,
i can confirm that the issue is now resolved (checked at 15:20 today) -Thanks again for the fix ather
explain many times to technical support but got nowhere and marked it down as congestion.

Richard,

I don't envy you working with customers that can hop between Manchester and London as single hosts - can't advertise those /32s out to peers!

Thanks for sorting this.