FTTP Pings/Traceroute

Firstly this isn't an issue but more for my understanding.

I'm on Zen 900/900 and pretty happy with the overall performance.

I've seen plenty of people reporting pings to various locations of sub 10ms when on FTTP/FTTH connections and tried a few test of my own.

When pinging the usual suspects BBC and Google i never get lower than 12ms (Not a complaint). When doing a traceroute this 2nd hop is the Zen Gateway and always as the 12ms mark.

PING bbc.co.uk (151.101.192.81): 56 data bytes
64 bytes from 151.101.192.81: icmp_seq=0 ttl=59 time=12.937 ms
64 bytes from 151.101.192.81: icmp_seq=1 ttl=59 time=13.552 ms
64 bytes from 151.101.192.81: icmp_seq=2 ttl=59 time=14.330 ms
64 bytes from 151.101.192.81: icmp_seq=3 ttl=59 time=14.574 ms
64 bytes from 151.101.192.81: icmp_seq=4 ttl=59 time=14.609 ms

traceroute to bbc.co.uk (151.101.192.81), 64 hops max, 52 byte packets
1 dsldevice (192.168.0.1) 2.798 ms 1.307 ms 1.351 ms
2 vt1.cor1.lond1.ptn.zen.net.uk (51.148.72.21) 13.648 ms 13.277 ms 13.037 ms
3 lag-8.p2.thn-lon.zen.net.uk (51.148.73.172) 13.229 ms
lag-8.p1.thn-lon.zen.net.uk (51.148.73.158) 13.758 ms
lag-8.p2.thn-lon.zen.net.uk (51.148.73.172) 13.757 ms
4 lag-1.br1.thn-lon.zen.net.uk (51.148.73.153) 13.301 ms 14.314 ms
lag-2.br1.thn-lon.zen.net.uk (51.148.73.167) 13.572 ms
5 * * *
6 * * *
7 * * *

Is this normal or would you expect better?

Just curious.

Cheers

Paul

Two questions:

Wired or wireless?

Where in the UK are you? London, Cornwall, Midland, Wales, Scottish Islands &c.

12-13ms ro london is prob Glasgow / Edinburgh

15-16 in Derry N.Ireland

Tried wired and wireless and both the same and i'm in Derby.

About what you’d expect ping from Derby to London really.

I'm in Derby.

I'm assuming I'm connecting to a gateway in London as no matter where i test to, I always get the first 4/5 hops as Zen.

traceroute to google.com (142.250.178.14), 64 hops max, 52 byte packets
1 dsldevice (192.168.0.1) 3.002 ms 1.285 ms 1.174 ms
2 vt1.cor1.lond1.ptn.zen.net.uk (51.148.72.21) 13.167 ms 13.245 ms 33.743 ms
3 lag-8.p1.thn-lon.zen.net.uk (51.148.73.158) 13.334 ms
lag-8.p2.thn-lon.zen.net.uk (51.148.73.172) 13.067 ms
lag-8.p1.thn-lon.zen.net.uk (51.148.73.158) 13.427 ms
4 lag-2.br1.thn-lon.zen.net.uk (51.148.73.167) 13.152 ms 13.660 ms
lag-1.br1.thn-lon.zen.net.uk (51.148.73.153) 13.099 ms
5 72.14.223.28 (72.14.223.28) 14.398 ms 14.069 ms 13.777 ms
6 * * *
7 142.251.54.28 (142.251.54.28) 14.192 ms
142.251.54.46 (142.251.54.46) 13.078 ms
108.170.246.161 (108.170.246.161) 13.888 ms
8 108.170.246.144 (108.170.246.144) 13.290 ms
108.170.246.143 (108.170.246.143) 13.629 ms
lhr48s27-in-f14.1e100.net (142.250.178.14) 13.839 ms

If is it consistent then not too much to worry about - although maybe a little higher than could be expected for Derby. Which way is it routed to London? Straright down or via Manchester or Leeds ... you will not know as all of the routing is totally transparent.

1st world problems (Not an issue really).

12ms is not a problem for me as being late 40's my gaming days ended with Doom and Quake.

I was more just trying to understand if that is perfectly normal for FTTP as you do see people post much lower pings on FTTP and sometimes on FTTC connections.

Edited by UKFOB (Sat 27-Nov-21 10:35:36)

Yes, it's normal. Depending on how Zen's network is configured, it could be that your packets flow from Derby to Manchester to London, and this all takes place at layer 2 so it doesn't appear in the traceroute.

In rough terms, all you can see is the first "routing" device that the traffic hits; you don't see all the "switching" devices in between.

PING bbc.co.uk (151.101.128.81): 56 data bytes
64 bytes from 151.101.128.81: seq=0 ttl=58 time=15.141 ms
64 bytes from 151.101.128.81: seq=1 ttl=58 time=15.160 ms
64 bytes from 151.101.128.81: seq=2 ttl=58 time=15.195 ms
64 bytes from 151.101.128.81: seq=3 ttl=58 time=15.189 ms
64 bytes from 151.101.128.81: seq=4 ttl=58 time=15.132 ms

--- bbc.co.uk ping statistics ---
5 packets transmitted, 5 packets received, 0% packet loss
round-trip min/avg/max = 15.132/15.163/15.195 ms



Gives you an idea on BT from far North West of Ireland.

That’s pretty good really. Capability of the BTW network is not to be sniffed at.

Yeah a whole 10ms at least lower than VM from my location

In reply to a post by BuckleZ:

12-13ms ro london is prob Glasgow / Edinburgh

15-16 in Derry N.Ireland

Similar latency from Edinburgh/Glasgow as Northern Ireland. Not quite as low as 12ms.

13.9ms from Edinburgh to London is the lowest I've seen on any residential ISP. That's on Openreach via BT Wholesale.
Talktalk over Openreach is 15.5ms.

Cityfibre from Edinburgh was slightly higher at around 16ms (Vodafone).
I haven't tested Zen over Cityfibre but they don't beat BT Wholesale on Openreach.

Virgin is 20ms from Edinburgh to London, though the jitter on DOCSIS is shocking meaning this bounces around.

12ms on CityFibre is likely the North of England somewhere.

Edit: just seen OP is in Derby.

Edited by j0hn83 (Sat 27-Nov-21 11:29:45)

In reply to a post by candlerb:

it could be that your packets flow from Derby to Manchester to London, and this all takes place at layer 2 so it doesn't appear in the traceroute.

From what I've read elsewhere I think that would be the case, a regionalised hub and spoke network at L2.

I've not seen any of the other major providers via Openreach get to the ping times that BTW can achieve.

My TalkTalk Biz connection has a 2 -3 ms higher base pings than the previous BTW network. Nothing at all to raise a sweat over mind you.

The only 'superior' AltNet ping times are from London-based Community Fibre on their own network, but that's pretty easy when you're operating within 10 miles of Telehouse and Equinix.

Often your routing will change with Zen as they don't run their network necessarily based on the best latency, more about spreading the load. From Leicester I get 6ms at best to sites in London, but often it can be has high as 12ms as it goes up to Manchester then back to London.

If you drop WAN/PPP and force a reconnection, it might take a few attempts, but I bet you might find that latency half as by chance you will end up on the direct route to London, in my experience anyway.

The problem it seems is many people get wise to this, so over several weeks routes to London become congested as people play the system, so you might find overnight you are unceremoniously kicked off and reconnected back to a higher latency route, and have to play the game of chance again if lowest latency is important to you.

This is from Leicestershire, but it's a leased line service (Exponential-e on-net, Openreach EAD), to give an idea of what is possible on some networks. First hop is always the London core.

gateway # execute ping outlook.office.com
PING outlook.ha.office365.com (52.97.241.162): 56 data bytes
64 bytes from 52.97.241.162: icmp_seq=0 ttl=243 time=5.6 ms
64 bytes from 52.97.241.162: icmp_seq=1 ttl=243 time=5.6 ms
64 bytes from 52.97.241.162: icmp_seq=2 ttl=243 time=5.6 ms
64 bytes from 52.97.241.162: icmp_seq=3 ttl=243 time=5.5 ms
64 bytes from 52.97.241.162: icmp_seq=4 ttl=243 time=5.5 ms

--- outlook.ha.office365.com ping statistics ---
5 packets transmitted, 5 packets received, 0% packet loss
round-trip min/avg/max = 5.5/5.5/5.6 ms

gateway # execute ping bbc.co.uk
PING bbc.co.uk (151.101.0.81): 56 data bytes
64 bytes from 151.101.0.81: icmp_seq=0 ttl=61 time=4.5 ms
64 bytes from 151.101.0.81: icmp_seq=1 ttl=61 time=4.5 ms
64 bytes from 151.101.0.81: icmp_seq=2 ttl=61 time=4.5 ms
64 bytes from 151.101.0.81: icmp_seq=3 ttl=61 time=4.4 ms
64 bytes from 151.101.0.81: icmp_seq=4 ttl=61 time=4.5 ms

--- bbc.co.uk ping statistics ---
5 packets transmitted, 5 packets received, 0% packet loss
round-trip min/avg/max = 4.4/4.4/4.5 ms

A Community Fibre service in London:

traceroute to bbc.co.uk (151.101.0.81), 30 hops max, 46 byte packets
1 x.x.x.x (x.x.x.x) 1.846 ms 1.741 ms 1.734 ms
2 94.177.138.97 (94.177.138.97) 1.986 ms 1.782 ms 1.986 ms
3 xge-0-2-5.ar03.thn.lon.network.as201838.net (94.247.86.34) 2.111 ms 1.882 ms 1.983 ms
4 lag-1.edge1.thn.lon.network.as201838.net (94.247.87.53) 1.858 ms 1.862 ms 1.730 ms

I'm in Twickenham and the BBC ( presumably also in the London area ) and I get:

for IPV6

Pinging bbc.co.uk [2a04:4e42::81] with 32 bytes of data:
Reply from 2a04:4e42::81: time=5ms
Reply from 2a04:4e42::81: time=5ms
Reply from 2a04:4e42::81: time=5ms
Reply from 2a04:4e42::81: time=5ms

Ping statistics for 2a04:4e42::81:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 5ms, Maximum = 5ms, Average = 5ms


Tracing route to bbc.co.uk [2a04:4e42::81]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms fritz.box [2a02:8011:700d:0:e228:6dff:feb8:9fab]
2 6 ms 23 ms 6 ms lo-0.cor1.lond1.ptn.zen.net.uk [2a02:8010::100]
3 6 ms 6 ms 6 ms ae-8.p1.thn-lon.zen.net.uk [2a02:8010:0:700::3a]
4 6 ms 5 ms 5 ms fastly-gw1.no-dns-yet.lonap.net [2001:7f8:17::d361:1]
5 10 ms 5 ms 5 ms 2a04:4e42::81

Trace complete.

and for IPV4

ping -4 bbc.co.uk

Pinging bbc.co.uk [151.101.0.81] with 32 bytes of data:
Reply from 151.101.0.81: bytes=32 time=6ms TTL=59
Reply from 151.101.0.81: bytes=32 time=5ms TTL=59
Reply from 151.101.0.81: bytes=32 time=5ms TTL=59
Reply from 151.101.0.81: bytes=32 time=5ms TTL=59

Ping statistics for 151.101.0.81:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 5ms, Maximum = 6ms, Average = 5ms

tracert -4 bbc.co.uk

Tracing route to bbc.co.uk [151.101.0.81]
over a maximum of 30 hops:

1 <1 ms <1 ms 7 ms fritz.box [194.32.67.8]
2 112 ms 85 ms 5 ms vt1.cor1.lond1.ptn.zen.net.uk [51.148.72.21]
3 6 ms 6 ms 8 ms lag-8.p2.thn-lon.zen.net.uk [51.148.73.172]
4 6 ms 5 ms 5 ms lag-2.br1.thn-lon.zen.net.uk [51.148.73.167]
5 6 ms 5 ms 5 ms ip81-59.fastly-gw1.lonap.net [5.57.81.59]
6 7 ms 5 ms 6 ms 151.101.0.81

Trace complete.

Leased line in London (Virtual1, Openreach EAD)

$ ping outlook.office.com
PING outlook.ha.office365.com (52.97.208.50): 56 data bytes
64 bytes from 52.97.208.50: icmp_seq=0 ttl=243 time=1.704 ms
64 bytes from 52.97.208.50: icmp_seq=1 ttl=243 time=1.761 ms
64 bytes from 52.97.208.50: icmp_seq=2 ttl=243 time=1.789 ms
64 bytes from 52.97.208.50: icmp_seq=3 ttl=243 time=1.811 ms
64 bytes from 52.97.208.50: icmp_seq=4 ttl=243 time=1.686 ms
64 bytes from 52.97.208.50: icmp_seq=5 ttl=243 time=1.726 ms
64 bytes from 52.97.208.50: icmp_seq=6 ttl=243 time=1.576 ms
64 bytes from 52.97.208.50: icmp_seq=7 ttl=243 time=1.730 ms

--- outlook.ha.office365.com ping statistics ---
8 packets transmitted, 8 packets received, 0.0% packet loss
round-trip min/avg/max/stddev = 1.576/1.723/1.811/0.068 ms


$ ping bbc.co.uk
PING bbc.co.uk (151.101.0.81): 56 data bytes
64 bytes from 151.101.0.81: icmp_seq=0 ttl=56 time=1.172 ms
64 bytes from 151.101.0.81: icmp_seq=1 ttl=56 time=1.157 ms
64 bytes from 151.101.0.81: icmp_seq=2 ttl=56 time=1.254 ms
64 bytes from 151.101.0.81: icmp_seq=3 ttl=56 time=1.285 ms
64 bytes from 151.101.0.81: icmp_seq=4 ttl=56 time=1.263 ms
64 bytes from 151.101.0.81: icmp_seq=5 ttl=56 time=1.250 ms
64 bytes from 151.101.0.81: icmp_seq=6 ttl=56 time=1.189 ms
64 bytes from 151.101.0.81: icmp_seq=7 ttl=56 time=1.291 ms

--- bbc.co.uk ping statistics ---
8 packets transmitted, 8 packets received, 0.0% packet loss
round-trip min/avg/max/stddev = 1.157/1.233/1.291/0.049 ms

TalkTalk Business FTTP (Openreach) from EAIPS (Suffolk):

>> ping outlook.office.com
SEQ HOST SIZE TTL TIME STATUS
0 52.97.146.210 56 243 10ms
1 52.97.146.210 56 243 10ms
2 52.97.146.210 56 243 10ms
3 52.97.146.210 56 243 10ms
4 52.97.146.210 56 243 10ms
5 52.97.146.210 56 243 10ms
6 52.97.146.210 56 243 10ms
7 52.97.146.210 56 243 10ms
8 52.97.146.210 56 243 10ms
9 52.97.146.210 56 243 10ms
10 52.97.146.210 56 243 10ms
sent=11 received=11 packet-loss=0% min-rtt=10ms avg-rtt=10ms max-rtt=10ms

>> ping bbc.co.uk
SEQ HOST SIZE TTL TIME STATUS
0 151.101.192.81 56 59 10ms
1 151.101.192.81 56 59 9ms
2 151.101.192.81 56 59 9ms
3 151.101.192.81 56 59 9ms
4 151.101.192.81 56 59 9ms
5 151.101.192.81 56 59 9ms
6 151.101.192.81 56 59 9ms
7 151.101.192.81 56 59 9ms
8 151.101.192.81 56 59 9ms
9 151.101.192.81 56 59 9ms
10 151.101.192.81 56 59 9ms
sent=11 received=11 packet-loss=0% min-rtt=9ms avg-rtt=9ms max-rtt=10ms

Intersite from London to Suffolk via IPSEC is around 10ms.

This was pretty much what i was thinking it should be given its full fibre and a new network. Shouldn't be any over subscription.

I'd be interested to see what any CF/Zen customers get in Leicester.

I did reboot the router recently for the IPV6 config to be added but still connected to the same first visable hop.

I'm on 900/100 from Zen and . To BBC/Google I'm getting around 8 to 11 ms via WiFI 5 to 6 from wired and 5 from the router itself.

Text

1 23 45 67 89 1011 12

traceroute to www.bbc.co.uk (212.58.233.251), 30 hops max, 60 byte packets  1  router ()  0.680 ms  0.634 ms  0.665 ms 2  vt1.cor1.lond2.ptn.zen.net.uk (51.148.72.23)  5.543 ms  5.563 ms  5.631 ms  3  lag-8.p2.ixn-lon.zen.net.uk (51.148.73.206)  5.508 ms lag-7.p1.ixn-lon.zen.net.uk (51.148.73.186)  5.465 ms  5.438 ms 4  lag-2.p1.thn-lon.zen.net.uk (51.148.73.132)  5.556 ms  5.525 ms  5.517 ms  5  lag-1.br1.thn-lon.zen.net.uk (51.148.73.153)  5.486 ms  5.434 ms lag-2.br1.thn-lon.zen.net.uk (51.148.73.167)  5.370 ms 6  82.71.254.134 (82.71.254.134)  5.252 ms  4.635 ms  4.607 ms  7  132.185.248.51 (132.185.248.51)  4.642 ms  4.717 ms  4.660 ms 8  ae11.cr02.thdow.bbc.co.uk (132.185.254.166)  4.645 ms  4.595 ms  4.554 ms  9  * * *10  ae2.er01.lbh.bbc.co.uk (132.185.249.7)  4.867 ms  4.827 ms  4.789 ms 11  132.185.252.126 (132.185.252.126)  169.204 ms  169.164 ms  169.126 ms

I'm am using my own router which is more powerful than the Fritzbox Zen provided

In reply to a post by UKFOB:

This was pretty much what i was thinking it should be given its full fibre and a new network. Shouldn't be any over subscription.

I'd be interested to see what any CF/Zen customers get in Leicester.

I did reboot the router recently for the IPV6 config to be added but still connected to the same first visable hop.

I doesn't necessarily follow that way. What is likely happening is the CityFibre connection is just getting you to the nearest point where it can hand over to Zen's own network, and from that point on it behaves like any Zen connection, i.e. you are bundled up with everyone else and treated exactly the same.

You may need to drop the PPP and reconnect several times to route hop, it can take several tries, although being FTTP no worry of DLM kicking in at least. You still end up seeing the same gateway(s), the route change happens at a lower level that isn't visible via traceroute except for seeing the change in latency to the gateway.

Vodafone's Gigafast using CityFibre's network.

Traceroute to bbc.co.uk
Smokeping to fastly.com showing the last week~

Edited by 0007 (Sat 27-Nov-21 16:20:31)

E300 / UKFOB

Depends on where his resilient node is and the routing to it., Your scenario only works where it is London / Manchester and you are south of Manchester by a significant distance. Because of Derby location it may route vis Manchester even if it is resilient on London. If they have a 3rd node ( Glasgow) it could be by there and even higher.. Direct routing to London from Derby won't be much shorter than going via Manchester due to transmission routes.

i thought they would of used AS8468 ex ENTANET network and not Vodafone's for CityFibre's network.

Few from Manchester that have IXManchester if any use to someone, Quite surprised BT dont have any public peering exchange points in IXManchester or any LINX Internet Exchange's only the one in London https://www.peeringdb.com/net/281

speedtest.tnp.net.uk
mirrors.manchester.m247.com
mirror-man.bytemark.co.uk

Edited by francisuk25 (Sun 28-Nov-21 22:51:57)

In reply to a post by francisuk25:

i thought they would of used AS8468 ex ENTANET network and not Vodafone's for CityFibre's network.

Cityfibre's residential FTTP network is wholesale layer 2, very similar to Openreach. All layer 3 (i.e. IP routing) is performed by the ISP.