Common Telegram Bot Hosting Errors, and Fixes

A running log of the most common issues when deploying a Telegram bot, and the actual fix for each.

"ModuleNotFoundError" / "No module named X"

A Python package your code imports isn't installed. Add it to your requirements.txt (the Requirements file startup variable points at this) and restart — the server runs pip install against it automatically on every boot. If you need something quick without touching the file, use the Additional Python packages variable instead.

Bot Doesn't Respond to Any Messages

• Check the console for a confirmation your bot actually started polling/connected — if that line never shows up, the process likely crashed before fully initializing. Scroll up for the real traceback.
• Double-check your bot token is correct and hasn't been revoked — see our token safety guide.
• If using webhooks instead of polling, confirm the webhook URL is actually reachable from Telegram's servers — a webhook pointed at an unreachable address fails silently from your bot's perspective.

"Conflict: terminated by other getUpdates request"

This means two instances of your bot are running at once using the same token — Telegram only allows one active polling connection per bot token. Usually caused by an old process that didn't fully stop before you started a new one, or accidentally running the bot both locally and on the server simultaneously. Stop everything, confirm only one process is running, then start again.

Bot Works, Then Stops Responding After a While

• Check if the process actually crashed (server shows stopped) vs. is still running but stuck — the console tells you which.
• A silent hang usually means an unhandled exception in a background task, or a network call that never times out. Add explicit timeouts to any HTTP/API calls your bot makes.
• See our 24/7 uptime guide for building real resilience against this class of issue.

"Wrong file executed" / Bot Doesn't Start At All

Check the App py file variable on the Startup tab matches your actual entry point's exact filename — a mismatch here means the server is trying to run a file that doesn't exist or isn't your real bot.

Git Pull Fails on Startup

• For a private repository, confirm Git Username and Git Access Token are both set correctly — a token, not your actual account password, is what's expected here.
• Confirm Git Branch matches an actual branch name in your repo if you're not using the default.

Still Stuck?

Open a support ticket with your service ID and the exact console output around the failure — a pasted traceback is almost always enough for us to point you at the real fix immediately.