cancel
Showing results for 
Search instead for 
Did you mean: 
1

Ask

2

Reply

3

Solution

Twitter, Cloudflare DNS, other routing issues?

Zuccster
4: Newbie

Anyone else seeing a routing blackhole like this:

 

$ tracepath twitter.com
 1?: [LOCALHOST]                      pmtu 1500
 1:  RT-AC86U-60B0.local                                   6.848ms 
 1:  RT-AC86U-60B0.local                                   2.058ms 
 2:  RT-AC86U-60B0.local                                   2.252ms pmtu 1492
 2:  no reply
 3:  63.130.127.213                                        8.214ms asymm  4 
 4:  10.200.18.45                                         12.856ms 
 5:  63.130.104.198                                       11.914ms asymm  6 
 6:  63.130.127.213                                       10.504ms asymm  4 
 7:  10.200.18.45                                         10.930ms asymm  4 
 8:  63.130.104.198                                       11.497ms asymm  6 
 9:  63.130.127.213                                       12.669ms asymm  4 
10:  10.200.18.45                                         12.699ms asymm  4 
11:  63.130.104.198                                       15.672ms asymm  6 
12:  63.130.127.213                                       14.249ms asymm  4 
13:  10.200.18.45                                         14.565ms asymm  4 
14:  63.130.104.198                                       15.385ms asymm  6 
15:  63.130.127.213                                       16.622ms asymm  4 
16:  10.200.18.45                                         14.558ms asymm  4 
17:  63.130.104.198                                       17.474ms asymm  6 
18:  63.130.127.213                                       18.652ms asymm  4 
19:  10.200.18.45                                         17.761ms asymm  4 
20:  63.130.104.198                                       18.485ms asymm  6 
21:  63.130.127.213                                       17.823ms asymm  4 
22:  10.200.18.45                                         19.614ms asymm  4 
23:  63.130.104.198                                       20.963ms asymm  6 
24:  63.130.127.213                                       21.040ms asymm  4 
25:  10.200.18.45                                         20.610ms asymm  4 
26:  63.130.104.198                                       22.114ms asymm  6 
27:  63.130.127.213                                       21.395ms asymm  4 
28:  10.200.18.45                                         22.728ms asymm  4 
29:  63.130.104.198                                       23.665ms asymm  6 
30:  63.130.127.213                                       25.953ms asymm  4 
     Too many hops: pmtu 1492
     Resume: pmtu 1492
44 REPLIES 44

Anonymous
Not applicable

Nope, currently here I get:

 

traceroute twitter.com
traceroute to twitter.com (104.244.42.1), 30 hops max, 60 byte packets
1 rtac86u.broadband (192.168.1.1)     0.163 ms   0.120 ms   0.096 ms
2 host-212-158-250-39.dslgb.com (212.158.250.39) 9.303 ms 9.447 ms 9.464 ms
3 63.130.105.110 (63.130.105.110)                9.366 ms 9.713 ms 9.700 ms
4 10.200.18.53 (10.200.18.53)                   10.304 ms 10.266 ms 10.365 ms
5 63.130.105.134 (63.130.105.134)               15.803 ms 16.647 ms 16.611 ms
6 10.200.18.37 (10.200.18.37)                   17.012 ms 15.894 ms 15.795 ms
7 xe-1-1-0.cr1-lon1.twttr.com (195.66.225.142)  16.783 ms 16.851 ms 16.508 ms
8 104.244.42.1 (104.244.42.1)                   17.130 ms 17.092 ms 17.056 ms

 

Cant explain why you are seeing what you are seeing, but you shouldn't be hitting the router 3 times before it even leaves your own network!  Is there anything odd going on in the routers log files when you do this?

 

 

Thanks for checking. 

[Edit: router logs look normal]

I think the three internal hops is an artifact  of using tracepath rather than traceroute, here's the same route examined with traceroute:

 

$ sudo traceroute 104.244.42.1
traceroute to 104.244.42.1 (104.244.42.1), 30 hops max, 60 byte packets
 1  RT-AC86U-60B0.local (192.168.1.1)  2.220 ms  2.204 ms  2.248 ms
 2  * * *
 3  63.130.127.213 (63.130.127.213)  9.649 ms  9.993 ms  9.991 ms
 4  10.200.18.45 (10.200.18.45)  9.617 ms  9.971 ms  10.242 ms
 5  63.130.104.198 (63.130.104.198)  11.026 ms  11.331 ms  11.336 ms
 6  63.130.127.213 (63.130.127.213)  11.076 ms  9.314 ms  9.244 ms
 7  10.200.18.45 (10.200.18.45)  9.404 ms  11.744 ms  11.132 ms
 8  63.130.104.198 (63.130.104.198)  12.704 ms  12.032 ms  12.692 ms
 9  63.130.127.213 (63.130.127.213)  11.767 ms  12.044 ms  12.182 ms
10  10.200.18.45 (10.200.18.45)  11.782 ms  11.935 ms  11.096 ms
11  63.130.104.198 (63.130.104.198)  12.712 ms  12.566 ms  14.793 ms
12  63.130.127.213 (63.130.127.213)  14.262 ms  13.678 ms  13.998 ms
13  10.200.18.45 (10.200.18.45)  14.014 ms  13.347 ms  13.568 ms
14  63.130.104.198 (63.130.104.198)  14.114 ms  13.828 ms  13.822 ms
15  63.130.127.213 (63.130.127.213)  14.639 ms  15.617 ms  14.723 ms
16  10.200.18.45 (10.200.18.45)  17.400 ms  16.603 ms  17.393 ms
17  63.130.104.198 (63.130.104.198)  18.005 ms  17.335 ms  17.330 ms
18  63.130.127.213 (63.130.127.213)  16.888 ms  16.315 ms  16.725 ms
19  10.200.18.45 (10.200.18.45)  16.773 ms  16.787 ms  15.955 ms

etc. etc. round and round util it gives up at 30 hops.

I can see from your IP that you're on a different subnet, and thus your routing obviously differs.

For info, I'm gigafast, and the VF dhcpd has give my router 84.65.0.1 as the default gateway.

 

I'm hitting the same problem, pings to twitter.com fails with 'Time to live exceeded', and a traceroute shows:

 

traceroute to www.twitter.com (104.244.42.65), 60 hops max, 60 byte packets
 1  vodafone.lan (192.168.1.1)  0.798 ms  1.152 ms  1.537 ms
 2  host-212-158-250-40.dslgb.com (212.158.250.40)  9.638 ms  9.741 ms  9.834 ms
 3  63.130.127.221 (63.130.127.221)  11.330 ms  11.443 ms  13.053 ms
 4  10.200.18.45 (10.200.18.45)  13.191 ms  15.380 ms  13.215 ms
 5  63.130.104.198 (63.130.104.198)  15.389 ms  15.487 ms  15.848 ms
 6  63.130.127.213 (63.130.127.213)  15.588 ms  14.400 ms  14.462 ms
 7  10.200.18.45 (10.200.18.45)  14.445 ms  10.187 ms  10.120 ms
 8  63.130.104.198 (63.130.104.198)  12.783 ms  12.380 ms  12.514 ms
 9  63.130.127.213 (63.130.127.213)  12.104 ms  11.499 ms  11.153 ms
10  10.200.18.45 (10.200.18.45)  11.451 ms  11.841 ms  11.778 ms
11  63.130.104.198 (63.130.104.198)  13.472 ms  13.708 ms  13.654 ms
12  63.130.127.213 (63.130.127.213)  13.021 ms  13.382 ms  13.317 ms
13  10.200.18.45 (10.200.18.45)  14.001 ms  13.655 ms  13.153 ms
14  63.130.104.198 (63.130.104.198)  15.573 ms  15.976 ms  15.804 ms
15  63.130.127.213 (63.130.127.213)  14.215 ms  14.662 ms  14.089 ms

... (the last three IPs loop continually)

 

Do Vodafone have any network status page, or a place to report problems? The only status page I could find just shows their phone coverage.

Right, I'll log it with support and point them here.

Suggest others do the same.

Yes, for twitter.com and 1.1.1.1 (cloudflare DNS)

 

Appears to be an Vodafone internal l3 routing loop given I'm not seeing the packets leave Vodafone/c&w nor am I seeing the same routing from other ISPs.

 

tac2@MacBook-Pro  ~  sudo /usr/local/sbin/mtr twitter.com --report --report-wide
Password:
Start: 2020-06-18T22:30:55+0100
HOST: MacBook-Pro.3ke.uk Loss% Snt Last Avg Best Wrst StDev
1.|-- router.3ke.uk 0.0% 10 1.2 1.3 1.1 1.5 0.1
2.|-- host-212-158-250-38.dslgb.com 0.0% 10 7.1 14.0 7.0 45.7 11.8
3.|-- 63.130.127.213 0.0% 10 7.7 10.3 7.2 35.3 8.8
4.|-- 10.200.18.45 0.0% 10 9.2 8.0 7.3 9.2 0.6
5.|-- 63.130.104.198 0.0% 10 9.7 10.4 8.5 21.0 3.7
6.|-- 63.130.127.213 0.0% 10 9.6 14.2 8.8 39.5 10.8
7.|-- 10.200.18.45 0.0% 10 10.0 10.7 9.2 21.4 3.8
8.|-- 63.130.104.198 0.0% 10 11.5 16.4 10.3 43.4 11.3
9.|-- 63.130.127.213 0.0% 10 10.6 10.8 10.3 12.2 0.5
10.|-- 10.200.18.45 0.0% 10 11.9 11.7 10.8 14.9 1.2
11.|-- 63.130.104.198 0.0% 10 13.1 12.6 12.1 14.1 0.6
12.|-- 63.130.127.213 0.0% 10 13.5 14.1 12.2 23.1 3.5
13.|-- 10.200.18.45 0.0% 10 13.4 13.3 12.3 15.1 0.8
14.|-- 63.130.104.198 0.0% 10 14.5 17.7 13.6 42.3 8.8
15.|-- 63.130.127.213 0.0% 10 14.8 15.0 13.1 19.0 2.0
16.|-- 10.200.18.45 0.0% 10 14.3 14.8 13.8 19.7 1.7
17.|-- 63.130.104.198 0.0% 10 15.7 17.3 15.1 32.5 5.3
18.|-- 63.130.127.213 0.0% 10 15.2 16.0 15.2 17.4 0.8
19.|-- 10.200.18.45 0.0% 10 15.6 15.9 15.3 16.4 0.3
20.|-- 63.130.104.198 0.0% 10 17.3 20.2 16.6 47.1 9.5
21.|-- 63.130.127.213 0.0% 10 16.6 17.3 16.6 18.4 0.5
22.|-- 10.200.18.45 0.0% 10 17.4 18.4 17.0 26.7 2.9
23.|-- 63.130.104.198 0.0% 10 18.4 21.8 18.4 44.4 8.0
24.|-- 63.130.127.213 0.0% 10 18.5 18.9 18.4 19.8 0.4
25.|-- 10.200.18.45 0.0% 10 18.8 18.9 18.4 19.7 0.4
26.|-- 63.130.104.198 0.0% 10 22.4 23.3 19.5 43.2 7.1
27.|-- 63.130.127.213 0.0% 10 20.8 20.2 19.5 20.8 0.3
28.|-- 10.200.18.45 0.0% 10 20.3 20.7 19.9 22.1 0.6
29.|-- 63.130.104.198 0.0% 10 27.6 24.7 21.4 42.5 6.5
30.|-- 63.130.127.213 0.0% 10 23.3 22.1 21.0 23.4 0.7

 

tac2@MacBook-Pro  ~  sudo /usr/local/sbin/mtr 1.1.1.1 --report --report-wide
Start: 2020-06-18T21:58:53+0100
HOST: MacBook-Pro.3ke.uk Loss% Snt Last Avg Best Wrst StDev
1.|-- router.3ke.uk 0.0% 10 1.2 1.6 1.1 2.7 0.5
2.|-- host-212-158-250-38.dslgb.com 0.0% 10 7.3 8.8 7.0 16.1 2.9
3.|-- 63.130.127.213 0.0% 10 7.2 10.3 7.2 34.3 8.4
4.|-- 10.200.18.45 0.0% 10 8.8 12.4 7.6 49.1 12.9
5.|-- 63.130.104.198 0.0% 10 8.7 9.9 8.7 15.8 2.1
6.|-- 63.130.127.213 0.0% 10 9.0 11.6 8.8 30.1 6.6
7.|-- 10.200.18.45 0.0% 10 9.2 9.6 8.9 10.1 0.4
8.|-- 63.130.104.198 0.0% 10 10.8 11.1 10.4 12.9 0.7
9.|-- 63.130.127.213 0.0% 10 10.9 13.4 10.3 38.5 8.8
10.|-- 10.200.18.45 0.0% 10 11.9 11.8 10.7 14.5 1.3
11.|-- 63.130.104.198 0.0% 10 12.3 12.3 11.7 13.2 0.4
12.|-- 63.130.127.213 0.0% 10 12.1 12.4 11.8 13.1 0.4
13.|-- 10.200.18.45 0.0% 10 12.5 14.3 12.1 27.8 4.8
14.|-- 63.130.104.198 0.0% 10 14.3 14.3 13.8 16.0 0.6
15.|-- 63.130.127.213 0.0% 10 14.4 16.3 13.6 36.8 7.2
16.|-- 10.200.18.45 0.0% 10 14.3 14.4 14.0 15.1 0.4
17.|-- 63.130.104.198 0.0% 10 16.5 15.9 15.3 16.8 0.5
18.|-- 63.130.127.213 0.0% 10 15.5 18.1 15.1 40.8 8.0
19.|-- 10.200.18.45 0.0% 10 15.9 16.3 15.5 19.2 1.1
20.|-- 63.130.104.198 0.0% 10 17.5 17.3 16.7 18.1 0.4
21.|-- 63.130.127.213 0.0% 10 17.6 19.0 16.5 31.2 4.5
22.|-- 10.200.18.45 0.0% 10 17.5 17.4 16.8 17.7 0.2
23.|-- 63.130.104.198 0.0% 10 18.5 21.2 18.5 43.0 7.7
24.|-- 63.130.127.213 0.0% 10 19.5 19.3 17.9 24.7 2.0
25.|-- 10.200.18.45 0.0% 10 18.7 21.3 18.7 40.6 6.8
26.|-- 63.130.104.198 0.0% 10 21.0 21.7 20.0 33.5 4.1
27.|-- 63.130.127.213 0.0% 10 20.6 22.5 20.0 43.1 7.2
28.|-- 10.200.18.45 0.0% 10 20.4 22.5 20.4 37.6 5.3
29.|-- 63.130.104.198 0.0% 10 30.5 25.9 21.5 49.0 8.6
30.|-- 63.130.127.213 0.0% 10 22.1 22.1 21.2 23.5 0.7

Anonymous
Not applicable

The test would be to see where the routing problem is.

 

Probably want to start looking for that by checking if:

 

NsLookup

Server: 35.178.60.174
Address 1: 35.178.60.174 ec2-35-178-60-174.eu-west-2.compute.amazonaws.com

Name: twitter.com
Address 1: 104.244.42.129
Address 2: 104.244.42.193

 

Just check that we are all in the same ballpark regards where we are looking.

*I don't have any linux machines here that use the specific "tracepath! instruction.

It seems to be a fairly straightforward upstream router configuration cockup.

I've just spent 30 mins on chat with a fantastically nice, but totally out of his depth, first line support  guy.

I was told to call the Gigafast support no. (08080-034-515) in the morning.

I'm having the same problem with cloudflare (1.1.1.1), traffic looping between the above IP's and not leaving Vodafone. 

Escalated to third line support just now.