# Bedrock Knowledge Base + Pinecone fails after vectors inserted

**URL:** <https://community.pinecone.io/t/bedrock-knowledge-base-pinecone-fails-after-vectors-inserted/8388>\
**Category:** Support\
**Tags:** vector-database\
**Created:** [September 12, 2025, 4:36pm UTC](https://community.pinecone.io/t/bedrock-knowledge-base-pinecone-fails-after-vectors-inserted/8388 "2025-09-12T16:36:39Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![armandofabians](https://sea2.discourse-cdn.com/flex020/user_avatar/community.pinecone.io/armandofabians/32/4409_2.png) [@armandofabians](https://community.pinecone.io/u/armandofabians)\
**Post date:** [September 12, 2025, 4:36pm UTC](https://community.pinecone.io/t/bedrock-knowledge-base-pinecone-fails-after-vectors-inserted/8388/1 "2025-09-12T16:36:39Z")

</div>

I’m integrating **AWS Bedrock Knowledge Bases (VECTOR)** with **Pinecone Serverless** as the vector store. Embeddings are **Amazon Titan Text Embeddings v2** (1024 dims, cosine). When my Pinecone index is **empty** , I can create/connect the KB. As soon as I **insert vectors** and try to retrieve, Bedrock returns an error — and if I try to re-create the KB pointing to a non-empty index/namespace, I get a storage configuration validation error.

**Pinecone index details**

- Type: **Serverless (Dense)**

- Cloud/Region: **aws / us-west-2**

- Dimensions: **1024**

- Metric: **cosine**

- Namespaces tested:

I can upsert and fetch vectors directly via Pinecone HTTP API.

1. **Bedrock retrieve** (index has data)

```auto
aws bedrock-agent-runtime retrieve \
  --region us-west-2 \
  --knowledge-base-id 123 \
  --retrieval-query '{ "text": "comprar una propiedad en México notario preventa Zapopan" }' \
  --retrieval-configuration '{"vectorSearchConfiguration":{"numberOfResults":5}}'

```

Returns:

```auto
An error occurred (DependencyFailedException) when calling the Retrieve operation: Internal Server Exception

```

1. **Bedrock create-knowledge-base** (pointing to a non-empty index / namespace)

- If I include `"namespace":"vitacasa"` (or any non-empty), I get:

```auto
ValidationException: The knowledge base storage configuration provided is invalid, please check your configuration.

```

- If I omit the namespace (so default/empty), KB creation succeeds **only when the index is empty** ; once the index has data, retrieval later fails with the same `DependencyFailedException`.

1. Could Bedrock be calling Pinecone endpoints like `/namespaces` **without** the `X-Pinecone-API-Version` header during validation or retrieval, causing Pinecone to return an error and Bedrock to surface `ValidationException` / `DependencyFailedException`?
2. Are there **constraints on metadata** that could cause Pinecone to reject/400 something Bedrock sends (e.g., large `text` metadata, field names like `text` / `url`, stringified numbers, etc.)? Any reserved keys?

---

<div class="post-metadata">

**Author:** ![jenna](https://sea2.discourse-cdn.com/flex020/user_avatar/community.pinecone.io/jenna/32/3998_2.png) [@jenna](https://community.pinecone.io/u/jenna)\
**Post date:** [September 22, 2025, 2:53am UTC](https://community.pinecone.io/t/bedrock-knowledge-base-pinecone-fails-after-vectors-inserted/8388/2 "2025-09-22T02:53:12Z")

</div>

Hi @armandofabians - Thanks for joining the Pinecone forums!

Unfortunately, the current integration between Amazon Bedrock Knowledge Base and Pinecone does not support existing (non-empty) indexes/namespaces. You’ll need to do the data ingestion through Bedrock as there is metadata created by Bedrock integration during data loading that is required for retrieval.

Is this a capability you’ll be needing in the future? I’d like to make sure we capture it with our product team and the AWS team we work with, if so.

Also, this is not well documented on our end, but I’ve passed this along to the team to get the docs updated.