Webpages not found unless page is refreshed multiple times

Forums › 2degrees (including Slingshot, Orcon, Flip, Stuff Fibre, MyRepublic, 2talk and Vocus) › Webpages not found unless page is refreshed multiple times

acsi

11 posts

Geek

#306912 3-Sep-2023 10:21

I've got a problem where a webpage will come up with "Cannot connect to server". It will eventually connect if the page is refreshed (sometimes up to around 40-50 times). This also affects the sending and receiving of emails.

This has been on and off for the last year and a bit but is now pretty much constant. It's the reason I swapped provider from Slingshot to 2Degrees (Who were promptly then brought out by Slingshot).

Multiple calls to the support desk seem to get it working again for a week or so (Its becoming more constant). Turning the ONT off and back on sometimes also gets it working again. Sometimes it will work fine for a month or a week but it now is constantly there.

I've swapped cables, routers and tried disconnecting devices from the network to no avail, and I don't think I can handle another call to tech support.

Once mine or my missus work VPN connects they do not drop out at all even though on other devices which aren't connected via the VPN you'll get server not found. So i'm now sure it's the ONT or further down the fibre as it's the same behaviour on different routers (I'm back to using the provided Fritzbox).

I'm now at the point of seriously considering swapping to hyper fibre just because it'll replace the ONT (I think).

Is there anyway to check the ONT before I go down the rabbit hole?

1 | 2

842 posts

Ultimate Geek

ID Verified

Lifetime subscriber

#3123211 3-Sep-2023 10:30

Example websites?

Traceroutes? Pings?

Mucked around with your DNS settings?

acsi

11 posts

Geek

#3123250 3-Sep-2023 12:53

huckster:

Example websites?

Traceroutes? Pings?

Mucked around with your DNS settings?

Websites - pretty much any but for example trademe.co.nz eBay.co.uk stuff.co.nz tvnz.co.nz

Ping - Its sort of behaving itself at the moment but

PING trademe.co.nz (151.101.66.137): 56 data bytes

64 bytes from 151.101.66.137: icmp_seq=0 ttl=57 time=10.426 ms

64 bytes from 151.101.66.137: icmp_seq=1 ttl=57 time=44.001 ms

64 bytes from 151.101.66.137: icmp_seq=2 ttl=57 time=87.606 ms

64 bytes from 151.101.66.137: icmp_seq=3 ttl=57 time=12.202 ms

64 bytes from 151.101.66.137: icmp_seq=4 ttl=57 time=63.017 ms

64 bytes from 151.101.66.137: icmp_seq=5 ttl=57 time=106.729 ms

64 bytes from 151.101.66.137: icmp_seq=6 ttl=57 time=10.261 ms

64 bytes from 151.101.66.137: icmp_seq=7 ttl=57 time=306.424 ms

64 bytes from 151.101.66.137: icmp_seq=8 ttl=57 time=125.074 ms

64 bytes from 151.101.66.137: icmp_seq=9 ttl=57 time=132.860 ms

64 bytes from 151.101.66.137: icmp_seq=10 ttl=57 time=211.861 ms

64 bytes from 151.101.66.137: icmp_seq=11 ttl=57 time=138.059 ms

64 bytes from 151.101.66.137: icmp_seq=12 ttl=57 time=141.005 ms

64 bytes from 151.101.66.137: icmp_seq=13 ttl=57 time=182.694 ms

64 bytes from 151.101.66.137: icmp_seq=14 ttl=57 time=228.791 ms

Traceroute -

traceroute: Warning: trademe.co.nz has multiple addresses; using 151.101.2.137

traceroute to trademe.co.nz (151.101.2.137), 64 hops max, 52 byte packets

1 192.168.1.1 (192.168.1.1) 34.825 ms 1.157 ms 1.111 ms

2 * * *

3 192.168.255.252 (192.168.255.252) 8.937 ms 3.642 ms 2.335 ms

4 192.168.255.253 (192.168.255.253) 2.337 ms 2.192 ms 2.871 ms

5 192.168.255.253 (192.168.255.253) 2.706 ms !X 2.118 ms !X 2.532 ms !X

But then I also get this

traceroute: Warning: trademe.co.nz has multiple addresses; using 151.101.66.137

traceroute to trademe.co.nz (151.101.66.137), 64 hops max, 52 byte packets

1 192.168.1.1 (192.168.1.1) 20.535 ms 98.985 ms 1.112 ms

2 * * *

3 192.168.255.252 (192.168.255.252) 91.758 ms 2.854 ms 2.862 ms

4 192.168.255.253 (192.168.255.253) 2.718 ms 2.132 ms 2.850 ms

5 default-rdns.vocus.co.nz (101.98.5.248) 2.615 ms 2.407 ms 2.758 ms

6 167.82.128.173 (167.82.128.173) 11.304 ms 252.732 ms 11.312 ms

7 * * *

8 * * *

It just keeps going with the "* * *" steps from here to hop 64

DNS - Yes I've tried using multiple different public DNS servers and the result is the same.

freitasm

BDFL - Memuneh
79250 posts

Uber Geek

Administrator

ID Verified

Trusted

Geekzone

Lifetime subscriber

#3123288 3-Sep-2023 14:08

What router are you using? What's the MTU size?

acsi

11 posts

Geek

#3123385 3-Sep-2023 18:09

freitasm: What router are you using? What's the MTU size?

I've got the 2degrees provided Fritzbox 7590 which was I had been using for the past few weeks (I've been swapping things around trying to isolate the problem). Currently I'm using a D-Link DIR-X3260.

MTU size is set to 1500 but I have also tried it set to 1492.

sir1963

3260 posts

Uber Geek

Subscriber

#3123407 3-Sep-2023 20:27

acsi:

Hi

I've got a problem where a webpage will come up with "Cannot connect to server". It will eventually connect if the page is refreshed (sometimes up to around 40-50 times). This also affects the sending and receiving of emails.

This has been on and off for the last year and a bit but is now pretty much constant. It's the reason I swapped provider from Slingshot to 2Degrees (Who were promptly then brought out by Slingshot).

Multiple calls to the support desk seem to get it working again for a week or so (Its becoming more constant). Turning the ONT off and back on sometimes also gets it working again. Sometimes it will work fine for a month or a week but it now is constantly there.

I've swapped cables, routers and tried disconnecting devices from the network to no avail, and I don't think I can handle another call to tech support.

Once mine or my missus work VPN connects they do not drop out at all even though on other devices which aren't connected via the VPN you'll get server not found. So i'm now sure it's the ONT or further down the fibre as it's the same behaviour on different routers (I'm back to using the provided Fritzbox).

I'm now at the point of seriously considering swapping to hyper fibre just because it'll replace the ONT (I think).

Is there anyway to check the ONT before I go down the rabbit hole?

Same issue as I had, the site I had issues with was 2degrees.

Changed my DNS to Cloudflare and the problems went away

acsi

11 posts

Geek

#3123411 3-Sep-2023 21:04

sir1963: Same issue as I had, the site I had issues with was 2degrees.

Changed my DNS to Cloudflare and the problems went away

I just tried that one. Unfortunately the result is the same.

Interestingly whilst Speedtest.net was refusing to be found I tried Pining it with the below result.

PING speedtest.net (151.101.194.219): 56 data bytes

64 bytes from 151.101.194.219: icmp_seq=0 ttl=57 time=289.553 ms

64 bytes from 151.101.194.219: icmp_seq=1 ttl=57 time=202.217 ms

64 bytes from 151.101.194.219: icmp_seq=2 ttl=57 time=212.296 ms

64 bytes from 151.101.194.219: icmp_seq=3 ttl=57 time=202.318 ms

64 bytes from 151.101.194.219: icmp_seq=4 ttl=57 time=215.035 ms

64 bytes from 151.101.194.219: icmp_seq=5 ttl=57 time=227.998 ms

Request timeout for icmp_seq 6

64 bytes from 151.101.194.219: icmp_seq=7 ttl=57 time=715.910 ms

64 bytes from 151.101.194.219: icmp_seq=8 ttl=57 time=46.903 ms

64 bytes from 151.101.194.219: icmp_seq=9 ttl=57 time=94.034 ms

Once it did resolve (after about 20 refresh's later) it looks like this

PING speedtest.net (151.101.194.219): 56 data bytes

64 bytes from 151.101.194.219: icmp_seq=0 ttl=57 time=382.741 ms

64 bytes from 151.101.194.219: icmp_seq=1 ttl=57 time=229.275 ms

64 bytes from 151.101.194.219: icmp_seq=2 ttl=57 time=11.052 ms

64 bytes from 151.101.194.219: icmp_seq=3 ttl=57 time=10.797 ms

64 bytes from 151.101.194.219: icmp_seq=4 ttl=57 time=35.482 ms

64 bytes from 151.101.194.219: icmp_seq=5 ttl=57 time=11.005 ms

64 bytes from 151.101.194.219: icmp_seq=6 ttl=57 time=10.482 ms

64 bytes from 151.101.194.219: icmp_seq=7 ttl=57 time=11.878 ms

64 bytes from 151.101.194.219: icmp_seq=8 ttl=57 time=58.210 ms

64 bytes from 151.101.194.219: icmp_seq=9 ttl=57 time=193.417 ms

64 bytes from 151.101.194.219: icmp_seq=10 ttl=57 time=241.441 ms

fe31nz

1228 posts

Uber Geek

#3123418 4-Sep-2023 00:19

The 151.101.194.219 address for speedtest.net appears to be on the 2Degrees network - I get consistently low pings to it most of the time (from 2Degrees static IP in Palmerston North):

PING 151.101.194.219 (151.101.194.219) 56(84) bytes of data.
64 bytes from 151.101.194.219: icmp_seq=1 ttl=57 time=11.9 ms
64 bytes from 151.101.194.219: icmp_seq=2 ttl=57 time=17.8 ms
64 bytes from 151.101.194.219: icmp_seq=3 ttl=57 time=12.8 ms
64 bytes from 151.101.194.219: icmp_seq=4 ttl=57 time=13.4 ms
64 bytes from 151.101.194.219: icmp_seq=5 ttl=57 time=12.1 ms
64 bytes from 151.101.194.219: icmp_seq=6 ttl=57 time=12.4 ms
64 bytes from 151.101.194.219: icmp_seq=7 ttl=57 time=12.0 ms
64 bytes from 151.101.194.219: icmp_seq=8 ttl=57 time=11.7 ms
64 bytes from 151.101.194.219: icmp_seq=9 ttl=57 time=12.3 ms
64 bytes from 151.101.194.219: icmp_seq=10 ttl=57 time=11.9 ms
64 bytes from 151.101.194.219: icmp_seq=11 ttl=57 time=13.4 ms
64 bytes from 151.101.194.219: icmp_seq=12 ttl=57 time=12.3 ms
64 bytes from 151.101.194.219: icmp_seq=13 ttl=57 time=12.3 ms

Tracing route to 151.101.194.219 over a maximum of 30 hops

1 <1 ms <1 ms <1 ms er4.jsw.gen.nz [10.0.1.251]
2 1 ms 2 ms 1 ms 108.7.69.111.static.snap.net.nz [111.69.7.108]
3 4 ms 2 ms 4 ms default-rdns.vocus.co.nz [101.98.5.213]
4 14 ms 21 ms 8 ms default-rdns.vocus.co.nz [101.98.5.212]
5 11 ms 13 ms 12 ms 167.82.128.173
6 * * * Request timed out.
7 11 ms 12 ms 11 ms 151.101.194.219

You should not be getting any large ping times to it at all, so that is clearly a problem (unless you are badly overloading your connection while doing your tests).

This line:

5 192.168.255.253 (192.168.255.253) 2.706 ms !X 2.118 ms !X 2.532 ms !X

from your TradeMe trace is interesting. The !X means you received an ICMP reply packet that said "administratively prohibited". In other words, a router firewall was deliberately dropping the packets, but was set up so that it actually bothered to reply to your traffic with an ICMP reply packet to tell you about the problem. The address of the router doing that is 192.168.255.253, which seems to be your Internet router. There are lots of reasons why a router might reply like that, but the obvious one would be that it was currently disconnected from the upstream router and had nowhere to send the packets. So if you check the router logs, is it disconnecting and reconnecting frequently?

Equinox IT

Cloud spending continues to surge globally, but most organisations haven’t made the changes necessary to maximise the value and cost-efficiency benefits of their cloud investments. Download the whitepaper From Overspend to Advantage now.

fe31nz

1228 posts

Uber Geek

#3123419 4-Sep-2023 00:32

Also, I am surprised at the long ping times for the parts of the traces inside your network - ping times to all the devices on your own Ethernet network should be less than 1 millisecond, unless there is heavy traffic. You are getting much larger times from just the first hop to 192.168.1.1.

And there appear to be four hops to get out of your network:

192.168.1.1

(unknown address - it does not reply to traces)

192.168.255.252

192.168.255.253

That is very unusual for a home network. Usually, home network devices will be directly connected to the Internet router, as you see on my trace, where my Internet router is 10.0.1.251 and the next hop is the router it connects to on the 2Degrees network. So what are all those hops? What are the devices with those addresses?

yitz

2074 posts

Uber Geek

#3123420 4-Sep-2023 00:38

192.168.255.x is 2degrees CG-NAT.

fe31nz

1228 posts

Uber Geek

#3123422 4-Sep-2023 00:54

yitz:

192.168.255.x is 2degrees CG-NAT.

That would be very non-standard. The CG-NAT addresses are supposed to be 100.64.0.0/10. The 192.168.0.0/24 addresses are normally used by internal networks behind NAT (they are non-routeable), and if 2Degrees was using them for their CG-NAT addresses there would likely be address assignment clashes at times.

yitz

2074 posts

Uber Geek

#3123423 4-Sep-2023 01:02

fe31nz:

That would be very non-standard. The CG-NAT addresses are supposed to be 100.64.0.0/10. The 192.168.0.0/24 addresses are normally used by internal networks behind NAT (they are non-routeable), and if 2Degrees was using them for their CG-NAT addresses there would likely be address assignment clashes at times.

Intermediate hops 2 and 3 are link net IPs between BNG and CG-NAT boxes and 4 and 5 are CG-NAT boxes to border gateway. Last CG-NAT connection I came across the WAN IP was in 100.64.0.0/10 and had similar intermediate hops (same ranges but differ based on geography/aggregation). Keep in mind traceroute outputs may show artifacts and the communication administratively prohibited might just be a red herring as a symptom to how UDP traceroute traffic is being treated and seeing how there is no issue with straight ICMP echo replies.

timmmay

20574 posts

Uber Geek

Trusted

Lifetime subscriber

#3123439 4-Sep-2023 07:54

As a temporary test, try turning off IPv6 on your computer. Don't leave it turned off, this is simply a troubleshooting tip. Just a random idea, probably won't help, but will only take a few minutes to try.

acsi

11 posts

Geek

#3123629 4-Sep-2023 18:19

timmmay:

As a temporary test, try turning off IPv6 on your computer. Don't leave it turned off, this is simply a troubleshooting tip. Just a random idea, probably won't help, but will only take a few minutes to try.

Tried this. No change unfortunately.

fe31nz:

Also, I am surprised at the long ping times for the parts of the traces inside your network - ping times to all the devices on your own Ethernet network should be less than 1 millisecond, unless there is heavy traffic. You are getting much larger times from just the first hop to 192.168.1.1.

And there appear to be four hops to get out of your network:

192.168.1.1

(unknown address - it does not reply to traces)

192.168.255.252

192.168.255.253

That is very unusual for a home network. Usually, home network devices will be directly connected to the Internet router, as you see on my trace, where my Internet router is 10.0.1.251 and the next hop is the router it connects to on the 2Degrees network. So what are all those hops? What are the devices with those addresses?

My network is set to only use the 192.168.1.??? range so don't believe they are part of my network.

I have tracked the variability in the ping timings down to the wifi. Connected via cable they are stable.

PING trademe.co.nz (151.101.130.137): 56 data bytes

64 bytes from 151.101.130.137: icmp_seq=0 ttl=57 time=14.275 ms

64 bytes from 151.101.130.137: icmp_seq=1 ttl=57 time=26.377 ms

64 bytes from 151.101.130.137: icmp_seq=2 ttl=57 time=14.926 ms

64 bytes from 151.101.130.137: icmp_seq=3 ttl=57 time=14.060 ms

64 bytes from 151.101.130.137: icmp_seq=4 ttl=57 time=13.710 ms

64 bytes from 151.101.130.137: icmp_seq=5 ttl=57 time=14.515 ms

64 bytes from 151.101.130.137: icmp_seq=6 ttl=57 time=15.596 ms

64 bytes from 151.101.130.137: icmp_seq=7 ttl=57 time=14.470 ms

64 bytes from 151.101.130.137: icmp_seq=8 ttl=57 time=13.590 ms

64 bytes from 151.101.130.137: icmp_seq=9 ttl=57 time=14.415 ms

64 bytes from 151.101.130.137: icmp_seq=10 ttl=57 time=15.059 ms

And trace route

traceroute: Warning: trademe.co.nz has multiple addresses; using 151.101.130.137

traceroute to trademe.co.nz (151.101.130.137), 64 hops max, 52 byte packets

1 192.168.1.1 (192.168.1.1) 8.171 ms 4.067 ms 4.022 ms

2 * * *

3 192.168.255.252 (192.168.255.252) 5.885 ms 5.952 ms 7.086 ms

4 192.168.255.253 (192.168.255.253) 6.647 ms 6.845 ms 5.335 ms

5 192.168.255.253 (192.168.255.253) 5.802 ms !X 5.599 ms !X 8.017 ms !X

And its still constantly not finding pages without multiple refreshes.

acsi

11 posts

Geek

#3124022 6-Sep-2023 07:06

So after reading about the Fritzbox 7.57 update on another thread I swapped back to the Fritzbox 7590 this morning and updated it.

So far so good and Traceroute seems to be going further

traceroute to nzpost.co.nz (103.241.84.82), 64 hops max, 52 byte packets

1 192.168.1.1 (192.168.1.1) 2.191 ms 2.377 ms 1.048 ms

2 * * *

3 192.168.255.252 (192.168.255.252) 296.710 ms 166.753 ms 357.234 ms

4 192.168.255.253 (192.168.255.253) 7.607 ms 7.663 ms 8.639 ms

5 default-rdns.vocus.co.nz (101.98.5.248) 8.722 ms 134.257 ms 422.835 ms

6 default-rdns.vocus.co.nz (101.98.5.212) 8.050 ms 7.532 ms 8.488 ms

7 default-rdns.vocus.co.nz (101.98.5.213) 8.154 ms 7.507 ms 8.403 ms

8 169.14.69.111.static.snap.net.nz (111.69.14.169) 16.321 ms 16.079 ms 18.989 ms

9 as134433.auckland.megaport.com (43.243.22.106) 348.736 ms 169.658 ms 354.293 ms

10 103.241.84.82 (103.241.84.82) 14.468 ms 14.826 ms 106.640 ms

Linux

11391 posts

Uber Geek

Trusted

Lifetime subscriber

#3124024 6-Sep-2023 07:10

That is interesting for sure

1 | 2