Troubleshoot Pipeline Deployment
It may happen that deploying your pipeline fails or not all your files are indexed. Learn how to resolve this.
You must be an Admin to perform this task.
Pipeline Status: Partially Indexed
- Description: All of the files were processed, but at least one of the files failed to be indexed.
- Solution: Check the following issues:
- From an NLP standpoint, is your pipeline OK?
- Are your files OK? Are any of your files corrupted?
- Is there a node in your pipeline that may be throwing exceptions for certain files because of encoding or some other node settings?
- Use the get pipeline files endpoint to see the status of the files.
- Delete your files and upload them again to trigger indexing for these files. For more information, see Upload Files.
Pipeline Status: Failed to Deploy
- Description: Your pipeline wasn't deployed, and the files weren't indexed.
- Run your pipeline locally and check if there are any errors. You can do it by opening it in JupyterLab and running it there. For more information, see Update a Pipeline and Use SDK to Work with Pipelines.
- Contact your deepset Cloud representative.
Updated 2 months ago