0

Wanted to doublecheck some behaviour I'm seeing with pyngrok and ngrok.

Recently I think my companies IT team utilises a DNS service blocking the use of ngrok (reverse tunnel is a security issue) and my pyngrok script suddenly stopped working.

I noticed when I manually run the same configuration + yaml file, it fails the initial connection, same as pyngrok, then it retries the connection utilising Google DNS instead of my local DNS provider and the tunnel forms just fine.

Can I get pyngrok to allow the retry connection and utilise the Google DNS response to build the tunnel with pyngrok as well? I've looked at the API Documentation and there are some hints as to possibly how it could be done but I wasn't sure if it encapsulated ngrok's method of retries or some other way.

I have a video showing exactly what happened which I can try to find a way to upload and share.

vladbypass
  • 1
  • 2
  • [video link here](https://youtu.be/XGhAj00x3oA) – vladbypass Aug 13 '21 at 03:53
  • Hello, I am the developer of `pyngrok`. In theory, `pyngrok` should support this, as it effectively just acts as a pass thru to `ngrok` and utilizes all the same commands and configs. But it does also try to parse `ngrok`'s logs as its running the binary, so perhaps something there is throwing an exception and causing `pyngrok` to shut `ngrok` down? Could you share your failing script and its stacktrace and log an issue on [GitHub](https://github.com/alexdlaird/pyngrok/issues)? – alexdlaird Aug 13 '21 at 17:27
  • Also, in your video, could you execute the same command with the flag `--log=stdout` and share that output in the GitHub issue? This is the log format `pyngrok` is parsing, so it will help me investigate. – alexdlaird Aug 13 '21 at 17:28

1 Answers1

0

Fix is discussed on Github here:

https://github.com/alexdlaird/pyngrok/issues/88

vladbypass
  • 1
  • 2