Connection Problem #29

Closed
opened 2025-11-19 07:12:23 -06:00 by GiteaMirror · 4 comments
Owner

Originally created by @Aalkingkarl on GitHub (May 28, 2025).

Hello,
i rent an IONOS VPS and installed pangolin. I created a site and copy the script to portainer on the client site.
I created a rule vor ICMP.
At portainer a get the error:

INFO: 2025/05/28 09:50:28 Ping attempt 28

INFO: 2025/05/28 09:50:28 Pinging 100.89.128.1

WARN: 2025/05/28 09:50:38 Ping attempt 28 failed: failed to read ICMP packet: i/o timeout

100.89.128.1 is not my IONOS Server.

Image

Originally created by @Aalkingkarl on GitHub (May 28, 2025). Hello, i rent an IONOS VPS and installed pangolin. I created a site and copy the script to portainer on the client site. I created a rule vor ICMP. At portainer a get the error: INFO: 2025/05/28 09:50:28 Ping attempt 28 INFO: 2025/05/28 09:50:28 Pinging 100.89.128.1 WARN: 2025/05/28 09:50:38 Ping attempt 28 failed: failed to read ICMP packet: i/o timeout 100.89.128.1 is not my IONOS Server. ![Image](https://github.com/user-attachments/assets/faabbaa9-2e7f-4aaa-aba6-863960c1e721)
Author
Owner

@unraiders commented on GitHub (May 29, 2025):

Hello, i rent an IONOS VPS and installed pangolin. I created a site and copy the script to portainer on the client site. I created a rule vor ICMP. At portainer a get the error:

INFO: 2025/05/28 09:50:28 Ping attempt 28

INFO: 2025/05/28 09:50:28 Pinging 100.89.128.1

WARN: 2025/05/28 09:50:38 Ping attempt 28 failed: failed to read ICMP packet: i/o timeout

100.89.128.1 is not my IONOS Server.

Image

Have you opened the ports in the IONOS firewall mentioned in the documentation (51820)?

https://docs.fossorial.io/Getting%20Started/quick-install

The 100.89.128.1 doesn't refer to your VPS, but rather to the tunnel that creates the connection.

@unraiders commented on GitHub (May 29, 2025): > Hello, i rent an IONOS VPS and installed pangolin. I created a site and copy the script to portainer on the client site. I created a rule vor ICMP. At portainer a get the error: > > INFO: 2025/05/28 09:50:28 Ping attempt 28 > > INFO: 2025/05/28 09:50:28 Pinging 100.89.128.1 > > WARN: 2025/05/28 09:50:38 Ping attempt 28 failed: failed to read ICMP packet: i/o timeout > > 100.89.128.1 is not my IONOS Server. > > ![Image](https://github.com/user-attachments/assets/faabbaa9-2e7f-4aaa-aba6-863960c1e721) Have you opened the ports in the IONOS firewall mentioned in the documentation (51820)? https://docs.fossorial.io/Getting%20Started/quick-install The 100.89.128.1 doesn't refer to your VPS, but rather to the tunnel that creates the connection.
Author
Owner

@DoctorDrenom commented on GitHub (May 29, 2025):

i have the same issue, even when i just try to setup pangolin on my local server with two vms.

I now that that is the ip/subnet of the gerbil, but somehow they cant ping each other

@DoctorDrenom commented on GitHub (May 29, 2025): i have the same issue, even when i just try to setup pangolin on my local server with two vms. I now that that is the ip/subnet of the gerbil, but somehow they cant ping each other
Author
Owner

@oschwartz10612 commented on GitHub (May 29, 2025):

Usually the ping failures are due to one or both:

  • UDP port 51820 is not open on the server
  • Something is obscuring the public IP of the server such as Cloudflare Proxy being enabled on the DNS records. You can check by making sure what you have entered in gerbil.base_endpoint in the config file resolves to the public IP of the server with dig or nslookup

The next thing I would do is a tcpdump on the server to see if you are getting any traffic to it from newt. Could you do the following command on your pangolin instance and see if you capture any packets when you start the newt connection?

tcpdump -i any -n udp port 51820

If you are not getting packets it is likely something is blocking traffic. You can also try to run echo "poke" | nc -u <base_endpoint> 51820 from the machine running newt to see if you see the packets. If not it is even more likely something is blocking it.

@oschwartz10612 commented on GitHub (May 29, 2025): Usually the ping failures are due to one or both: - UDP port 51820 is not open on the server - Something is obscuring the public IP of the server such as Cloudflare Proxy being enabled on the DNS records. You can check by making sure what you have entered in gerbil.base_endpoint in the config file resolves to the public IP of the server with dig or nslookup The next thing I would do is a tcpdump on the server to see if you are getting any traffic to it from newt. Could you do the following command on your pangolin instance and see if you capture any packets when you start the newt connection? tcpdump -i any -n udp port 51820 If you are not getting packets it is likely something is blocking traffic. You can also try to run echo "poke" | nc -u <base_endpoint> 51820 from the machine running newt to see if you see the packets. If not it is even more likely something is blocking it.
Author
Owner

@Aalkingkarl commented on GitHub (Jun 2, 2025):

Yhea, i'm very stupid. i forgot the udp Port 51820. Thanks for your help

@Aalkingkarl commented on GitHub (Jun 2, 2025): Yhea, i'm very stupid. i forgot the udp Port 51820. Thanks for your help
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/newt#29