Multiple merge and vectorize code:500 errors

Support
1

Description

Getting a ton of these errors (and similar):

ERROR: [{“message”:“msg:merge and vectorize code:500 err:update vector: send POST request: Post "https://api.openai.com/v1/embeddings": context deadline exceeded (Client.Timeout exceeded while awaiting headers)”}]

Despite a 120000 millisecond timeout. First time I’ve ever seen this.

ERROR: [{“message”:“msg:merge and vectorize code:500 err:update vector: connection to: OpenAI API failed with status: 503 error: Service Unavailable.”}]

ERROR: [{“message”:“msg:merge and vectorize code:500 err:update vector: connection to: OpenAI API failed with status: 500 error: The server had an error while processing your request. Sorry about that!”}]

Server Setup Information

  • Weaviate Server Version: 1.23.7 (WCS)
  • Deployment Method:
  • Multi Node? Number of Running Nodes:
  • Client Language and Version:

Any additional Information

1 Like
2

Hi!

Looks like this is an issue with OpenAI, as the message you have is from OpenAI directly.

I usually see some errors like thos one, but they will eventually go away :man_shrugging:

Are you still facing it?

Thanks!

3

These errors persisted for a few hours last night. Finally got the embedding done, though. Will let you know if it happens again.

1 Like
4

I’m getting this as well with Cohere, they’re not down and I’m sure my API key is right…
{‘error’: [{‘message’: ‘send POST request: Post “https://api.cohere.ai/v1/embed”: context deadline exceeded (Client.Timeout exceeded while awaiting headers)’}]}

I’m using weaviate client V3 (I want to move to V4 but cannot when self-hosted through railway as they only export 1 port), latest image self-hosted, I create an instance and start loading to it immediately.

5

Don’t quote me on this, but I believe this issue for me involved the limitations of my cluster. I had data loss, and it appears that happened because my cluster size did not expand as my data expanded.

You should contact support and ask them if there is an issue with your cluster.

6

@Baptiste_Cumin, I believe this error is indeed from Cohere not answering the post request sent from Weaviate on a timely manner.

Maybe the content are too big and is taking time to be vectorized?