← All platforms

Tornado error tracking

The I/O loop keeps moving. Don’t lose the exception along the way. The Sentry Tornado integration sends your error reports to Telebugs on your server.

~/hoofnotes  pip install sentry-sdk
~/hoofnotes  nvim app.py
import sentry_sdk
sentry_sdk.init(
# Copy the complete Telebugs DSN, key and path included.
dsn="YOUR_TELEBUGS_DSN",
)
from hoofnotes.server import main
~/hoofnotes  python -m hoofnotes.server

Space pauses or resumes. R replays. Escape shows the complete setup.

Hang on, I already use Sentry

Keep sentry-sdk and change its destination to your Telebugs DSN. Your error reports stay on your server. You don’t need a Sentry account.

Will I see the handler that failed?

The Tornado integration reports request handler exceptions with their Python backtraces. You can also report an exception yourself when your code catches it and carries on.

What about an error I handle myself?

It can still be worth investigating. Send it with sentry_sdk.capture_exception and keep its backtrace alongside the errors the integration captures.