Logging & Streaming
Zeabur provides real-time log streaming for every service, giving you immediate visibility into your application’s behavior directly from the dashboard.
Viewing Logs
To view logs for a service:
- Open your project in the Zeabur dashboard.
- Select the service you want to inspect.
- Click the Logs tab.
The log viewer displays output from your application’s stdout and stderr streams in real time. New log entries appear automatically as they are emitted — no manual refresh is needed.
Build Logs vs. Runtime Logs
Zeabur separates logs into two categories:
- Build logs — Output generated during the build phase (e.g., dependency installation, compilation). These are available under the Deployments tab for each deployment.
- Runtime logs — Output generated while your service is running. These are shown in the Logs tab and stream in real time.
Searching and Filtering
The log viewer supports keyword search to help you locate specific entries. Type a search term in the filter bar at the top of the log viewer to highlight and filter matching lines.
This is particularly useful for:
- Finding error messages or stack traces
- Tracing specific request IDs
- Filtering by log level (e.g.,
ERROR,WARN)
Log Retention
Runtime logs are available as long as the service is running. When a service is restarted or redeployed, logs from the previous instance are no longer accessible in the dashboard.
Zeabur does not provide long-term log storage. If you need to retain logs beyond the current service lifecycle, integrate an external logging service such as Logflare, Better Stack, or Datadog. You can forward logs by configuring your application to send output to these services directly.
Best Practices
- Use structured logging. Output logs in JSON format to make them easier to search and parse, especially if forwarding to external services.
- Include timestamps. While Zeabur adds timestamps to log entries, including them in your application output ensures consistency across environments.
- Log at appropriate levels. Use
INFOfor routine events,WARNfor recoverable issues, andERRORfor failures that need attention. Avoid excessiveDEBUGoutput in production. - Avoid logging sensitive data. Never log passwords, API keys, or personally identifiable information.