troubleshooting-dbt-job-errors

To troubleshoot failing dbt Cloud jobs, use a systematic diagnostic workflow to classify failures into Infrastructure, Code/Compilation, or Data/Test categories using logs and run_results.json. This identifies root causes for unclear errors.

Intermittent dbt Cloud job failures are diagnosed by gathering logs, run_results, and contextual artifacts to classify the failure type. This systematic approach isolates root causes that are difficult to reproduce during local development.

Debugging dbt Cloud job errors requires access to MCP tools, the dbt CLI, logs, and the ability to create investigation docs and PRs. These components gather run_results and implement fixes for identified root causes.

No, this diagnostic approach is specifically for failed dbt Cloud or platform jobs. It does not apply to local dbt development errors, which require a different debugging process outside the MCP Admin API workflow.

The best way to investigate unknown dbt Cloud job failures is gathering logs, run_results, and recent changes to classify the failure type. If the root cause remains undetermined, generate documentation and handoff artifacts.