High Latency in ICMP Echo Requests and Web Requests When Source IP Belongs to AS476

What is the name of the domain?

example.com

What is the issue you’re encountering

High Latency in ICMP Echo Requests and Web Requests When Source IP Belongs to AS4760

What are the steps to reproduce the issue?

Recently, an unusual issue has been identified: when the source IP belongs to AS4760, ICMP echo requests and web requests exhibit high latency in their responses.
I conducted several tests, and the results are as follows:

mtr to 172.67.209.86, with the source IP range of 218.102.96.0/19, yielded the following results:
<-------------------------------------------------------------------------------
My traceroute [v0.95]
wsl (172.21.107.129) → 172.67.209.86 (172.67.209.86) 2025-01-29T00:57:11+0800
Keys: Help Display mode Restart statistics Order of fields quit
Packets Pings
Host Loss% Snt Last Avg Best Wrst StDev

  1. HIDDEN 0.0% 21 0.2 0.5 0.2 0.8 0.2
  2. HIDDEN 0.0% 20 2.3 2.1 1.4 2.4 0.2
  3. HIDDEN 0.0% 20 3.0 2.8 2.2 3.1 0.2
  4. HIDDEN 0.0% 20 2.4 3.1 2.3 4.9 0.7
  5. cloudflare1-lacp-100g.hkix.net 0.0% 20 4.2 7.9 3.7 23.4 6.2
  6. 103.22.203.75 0.0% 20 5.2 6.1 4.1 15.4 3.3
  7. 172.67.209.86 15.8% 20 419.1 287.7 3.6 545.1 175.9
    ------------------------------------------------------------------------------->

mtr to 104.21.58.225, with the source IP range of 218.102.96.0/19, yielded the following results:
<-------------------------------------------------------------------------------
My traceroute [v0.95]
wsl (172.21.107.129) → 104.21.58.225 (104.21.58.225) 2025-01-29T00:57:41+0800
Keys: Help Display mode Restart statistics Order of fields quit
Packets Pings
Host Loss% Snt Last Avg Best Wrst StDev

  1. HIDDEN 0.0% 21 0.7 0.5 0.2 1.2 0.3
  2. HIDDEN 0.0% 21 2.1 2.0 1.1 2.5 0.4
  3. HIDDEN 0.0% 20 3.1 2.8 2.4 3.2 0.3
  4. HIDDEN 0.0% 20 2.7 3.4 2.7 7.2 1.0
  5. cloudflare1-lacp-100g.hkix.net 0.0% 20 8.3 6.8 4.0 18.4 3.6
  6. 103.22.203.75 0.0% 20 6.2 7.3 3.7 23.6 4.8
  7. 104.21.58.225 0.0% 20 256.4 379.9 160.1 462.5 87.0
    ------------------------------------------------------------------------------->

mtr to 104.21.37.125, with the source IP range of 218.102.96.0/19, yielded the following results:
<-------------------------------------------------------------------------------
My traceroute [v0.95]
wsl (172.21.107.129) → 104.21.37.125 (104.21.37.125) 2025-01-29T00:58:18+0800
Keys: Help Display mode Restart statistics Order of fields quit
Packets Pings
Host Loss% Snt Last Avg Best Wrst StDev

  1. HIDDEN 0.0% 21 0.9 0.6 0.2 0.9 0.3
  2. HIDDEN 0.0% 21 2.2 2.0 1.4 2.5 0.3
  3. HIDDEN 0.0% 20 3.2 2.9 2.4 3.3 0.3
  4. HIDDEN 0.0% 20 2.8 3.4 2.7 4.7 0.5
  5. cloudflare1-lacp-100g.hkix.net 0.0% 20 19.9 7.6 4.0 19.9 4.4
  6. 103.22.203.75 0.0% 20 6.1 6.6 4.3 22.3 4.4
  7. 104.21.37.125 5.0% 20 3.9 243.4 3.7 489.0 214.2
    ------------------------------------------------------------------------------->

The MTR results indicate that the high latency occurs at the final hops of the route, which could suggest that Cloudflare’s servers in the Hong Kong region are experiencing overload or abnormal conditions.

Could you please escalate this issue to the engineering team for further investigation?

This topic was automatically closed after 15 days. New replies are no longer allowed.