[Bug]: delete index in graphrag

[ajain85]

Do you need to file an issue?

• I have searched the existing issues and this bug is not already filed.
• My model is hosted on OpenAI or Azure. If not, please look at the "model providers" issue and don't file a new one here.
• I believe this is a legitimate bug, not just a question. If this is a question, please use the Discussions area.

Describe the bug

I am facing index deletion issue in azure ai search and in local , I have deleted the file in blob input folder and After running update command it doesn't delete the indexes in the azure search ai. I am running below command .

cli - python -m graphrag update --config .\cli_graphrag\settings.yaml

Steps to reproduce

No response

Expected Behavior

No response

GraphRAG Config Used

# Paste your config here

Logs and screenshots

No response

Additional Information

• GraphRAG Version:
• Operating System:
• Python Version:
• Related Issues:

[natoverse]

This is correct - the update command appends new data, but does not remove data. That's a much more complicated task because the summarization process is lossy.

If you retain your cache, subsequent runs can avoid re-invoking the LLM for things like graph extraction. This makes it possible to re-run indexing with different mixes of document content. So you could remove the documents in question and then re-run the regular indexing. Graph extraction should be "free" in that it just uses the cache for all the existing text units. Depending on how the removed documents affect the community structure, community report generation could be the same cost as a usual run, or cheaper if some do not change and therefore also use the cache.

[natoverse]

I should mention: when the embeddings are generated during normal indexing, they overwrite the existing vector store index, so your old entries should go away. Because the update command only adds new content, the old entries would still be there.