Slow query response times

Support
1

Hi,

I am using Weaviate v4 I have created my collection as below:

collection = await wvt_client.collections.create(
    name="test",
    properties=collection_properties,
    vectorizer_config=Configure.Vectorizer.none()
)

I have 9 collections with a total of 3.8M objects, one collections contains the majority of 2.7M, I am doing a hybrid search as shown below:

response = await collection.query.hybrid(
    query="what is the capital of france?",
    vector=get_embeddings("what is the capital of france?"),
    alpha=0.5,  
    query_properties=["text", "file_name", "metadata", "tags"],
    limit=20
)

Current server config:

  • 12 gb ram
  • 4 cores

I am getting the below error when I try to query with the 2.7M objects collection, I can query other collections but I don’t see any issues.

 Unexpected error during Weaviate db search: Query call with protocol GRPC search failed with message <AioRpcError of RPC that terminated with:
        status = StatusCode.DEADLINE_EXCEEDED
        details = "Deadline Exceeded"
        debug_error_string = "UNKNOWN:Error received from peer  {created_time:"2025-06-17T06:01:47.589064253+00:00", grpc_status:4, grpc_message:"Deadline Exceeded"}

Thanks!

2

Hello @srik,

Welcome to our community — it’s lovely to have you here with us! We’re looking forward to supporting you and having you on board :partying_face:

The DEADLINE_EXCEEDED error usually means a timeout (latency/networking slowness) - the query is taking too long to complete, and the timeout configured on the client side has been exceeded. Essentially, the client didn’t receive a response from the server within the expected time.

I recommend increasing the timeout settings in your client configuration. Here’s an example:

# Connect to Weaviate
client = weaviate.connect_to_weaviate_cloud(
    cluster_url=CLUSTER_URL,
    auth_credentials=weaviate.auth.AuthApiKey(API_KEY),
    headers={
        "X-OpenAI-Api-Key": OPENAI_API_KEY,
        "X-Cohere-Api-Key": COHERE_API_KEY
    },
    additional_config=AdditionalConfig(
        timeout=Timeout(init=60, query=240, insert=240)
    )
)

Best regards,

Mohamed Shahin
Weaviate Support Engineer
(Ireland, UTC±00:00/+01:00)

3

Hi @Mohamed_Shahin Thank you for the response., after increasing the timeout it works. But, I do not want the query response times to be more than 1-2sec, how to achieve this?

More info about deployment:
image: 'cr.weaviate.io/semitechnologies/weaviate:1.30.0
resources:
requests:
cpu: “2000m”
memory: “8Gi”
ephemeral-storage: “8Gi”
limits:
cpu: “4000m”
memory: “12Gi”
ephemeral-storage: “12Gi”

We are deploying through helm charts, Single Node deployment in EKS cluster.