> ## Documentation Index
> Fetch the complete documentation index at: https://webscraping.titannet.io/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Monitoring and troubleshooting

> Titan separates task definition, execution control, worker execution, ingestion, and result access.

That separation is useful for troubleshooting because it narrows down where a problem most likely lives.

## If a task cannot be created

Start with:

* Authentication state
* Bearer token scopes
* Task payload validity
* Output schema shape

Most create-time issues are control-plane issues, not worker issues.

## If an execution never seems to start

Check:

* Whether the execution was created successfully
* Whether the execution remains queued
* Whether worker capacity is available
* Whether the relevant worker path is healthy

This usually points to orchestration or worker availability rather than result access.

## If an execution fails

Treat failure analysis in this order:

1. Execution state and logs
2. Worker-side runtime behavior
3. Ingestion acceptance of returned output

This keeps runtime failures distinct from post-execution data handling issues.

## If results are missing

Check:

* Whether the execution actually reached completion
* Whether ingestion accepted and finalized the output
* Whether you are calling result endpoints rather than execution control endpoints

Missing results often indicate an incomplete handoff between runtime completion and user-facing result publication.

## If media URLs do not behave as expected

Check:

* Whether the execution output includes platform-managed media URLs
* Whether the caller has the right access context
* Whether you are using the correct result export or media download path

Media access issues are usually access-path issues rather than scraping issues.

## Recommended operating model

For operational clarity, debug in this order:

1. Auth and access
2. Task definition
3. Execution state
4. Worker path
5. Ingestion and result access

That order matches the actual structure of the platform and reduces confusion.

## Next steps

* [Architecture overview](/about-platform/architecture-overview)
* [Monitor and control executions](/use-the-platform/monitor-and-control-executions)
* [How workers return data to the platform](/worker-guides/how-workers-return-data-to-the-platform)