Kendra Sync Indexing taking more than 2 hours and still not finished

Question

I have few files in S3 about 5-10 files with total size of about 15 MB. My kendra data source started sync and after 2 hours it si still in indexing state and not completing. How much time is it expected to take? Is it normal. I have added administrator access permission to my role in order to have kendra data source indexing permissions. I am following this blog:

https://aws.amazon.com/blogs/machine-learning/quickly-build-high-accuracy-generative-ai-applications-on-enterprise-data-using-amazon-kendra-langchain-and-large-language-models/

[ https://aws.amazon.com/blogs/machine-learning/quickly-build-high-accuracy-generative-ai-applications-on-enterprise-data-using-amazon-kendra-langchain-and-large-language-models/]()

ronaldtfaws · Answer

According to https://docs.aws.amazon.com/kendra/latest/dg/troubleshooting-data-sources.html:

```
If there are no updates to documents, sync time for a Amazon Kendra index increases in linear proportion to the number of documents. For example, 1,000 documents without any updates would take about five minutes to sync and 2,000 documents without any updates will take about 10 minutes. If there are any updates to the documents, then the sync time will increase based on the number of documents updated.
```

So, it looks that there are some problems with the syncing (maybe, with the access to the sources). You can have a look at https://docs.aws.amazon.com/kendra/latest/dg/iam-roles.html if you need help in creating/updating the policies to grant access to S3 from Kendra.

Kendra Sync Indexing taking more than 2 hours and still not finished

답변 추가

관련 콘텐츠