pyngrok | A Python wrapper for ngrok | Proxy library

Maybe your port 5000 is already occupied by some other process. Try netstat

I am the developer of pyngrok, and I wanted to reach out to let you know that I've seen reports of this pop up on our issue tracker, several on here, on Colab's issue tracker, etc. a lot in the last week. I reached out to ngrok, as that appeared to be where the issue was, and they just got back to me. They confirmed that they had inadvertently started blocking connections to Colab recently, and they have now fixed this. I just tested this again myself as well, and sure enough, it's fixed!

The reason the script is “stuck” is because pyngrok starts ngrok with a thread to monitor logs, and the Python process can’t exit until all threads have been dealt with. You can stop the monitor thread, as shown here in the documentation, or, if you have no use for it, you can prevent it from starting in the first place:

Per the docs, the DEFAULT_CONFIG_PATH variable is in the conf module, not the ngrok module. So change ngrok.DEFAULT_CONFIG_PATH = to conf.get_default().config_path = .

The thread with the ngrok tunnel will terminate as soon as the Python process terminates. So you are correct, the reason this is happening is because your script is not long lived. The easiest way to accomplish this is by following the example in the documentation.

Another issue is how you're setting the authtoken. Since you're not using the default config_path, you need to set this before setting the authtoken so it gets updated in the correct file (you'd also need to pass it to connect()). There are a couple ways to do this, but the easiest way from the docs is to just update the default config (since that's what will be used if you don't pass a pyngrok_config to any future method calls).

I also see that you're response variable is ssh_url, so you probably want to start a TCP tunnel to a port other than 80 (the default)—perhaps you've configured this in your ngrok.yml, but if not, I've updated the call to connect() to ensure this is the type of tunnel started for you and in case others try to use this same code snippet.

Full disclosure, I am the developer of pyngrok. Here is your code snippet updated with my changes.

After playing around with a few ideas, here is the solution I came other with. Since this question is asked frequently and often opened as a bug against ReadTheDocs on GitHub (which it's not, it just appears to be poorly supported and/or documented), I'll share my workaround here for others to find.

As mentioned above and in the docs, while ReadTheDocs allows you to override the auto-generated robots.txt and publish your own, you can't with sitemap.xml. Unclear why. Regardless, you can simply publish a different sitemap.xml, I named mine sitemap-index.xml, then, tell your robots.txt to point to your custom sitemap.

For my custom sitemap-index.xml, I only put the pages I care about rather then ever generated version (since stable and latest are really what I want search engines to be crawling, not versioned pages):