After attaching a persistent volume and creating a snapshot in Koyeb, my worker service (Telegram bot) stopped showing any stdout/stderr logs in the dashboard.
The bot itself works correctly — it responds in Telegram without issues — and the instance status shows as “Healthy”. However, no application logs appear in the Logs tab (stdout/stderr are empty). Only system events such as “Instance created” and “Instance is healthy” are visible.
I have already tried:
-
Rolling back to a previously stable deployment
-
Enabling unbuffered Python output (
PYTHONUNBUFFERED=1andpython -u) -
Verifying that logging is configured to write to stdout
-
Checking different instances and regions in the logs view
None of these steps restored the log output.
The only infrastructure change made before this issue appeared was adding a volume snapshot.
Is there any known interaction between volume snapshots and log streaming for worker services? Could attaching or restoring a volume affect how stdout/stderr logs are captured?
