X-Azure-Api-Key doesn't work in headers. Asking openai api key instead. while calling withNearText search.

[mihir-kanzariya]

How to reproduce this bug?

I am using azure openai, i created class and client .

const client = weaviate.client({ scheme: 'https', host: 'XXXXX-vv0royyl.weaviate.network', headers: { "X-Azure-Api-Key": PROCESS.ENV.AZURE_OPENAI_KEY } });

I get below error while calling Below function

Function: result = await client.graphql .get() .withClassName(model) .withNearText({ concepts: [question], targetVectors: ['content'], }) .withWhere({ path: ['trainingDataType'], operator: 'Equal', valueText: trainingDataType, }) .withLimit(limit) .withFields('content _additional { distance }') .do();

Error
{ "response": { "data": { "Get": { "Custcity": null } }, "errors": [ { "locations": [ { "column": 6, "line": 1 } ], "message": "explorer: get class: vectorize params: vectorize params: vectorize params: vectorize keywords: remote client vectorize: API Key: no api key found neither in request header: X-Openai-Api-Key nor in environment variable under OPENAI_APIKEY", "path": [ "Get", "Custcity" ] } ], "status": 200, "headers": {} }, "request": { "query": "{Get{Custcity(where:{operator:Equal,valueText:\"SCHEMA\",path:[\"trainingDataType\"]},nearText:{concepts:[\"Analyze the correlation between the number of activities (tasks and events) and opportunity win rates.\"],targetVectors:[\"content\"]},limit:8){content _additional { distance }}}}" } }

What is the expected behavior?

It should consider azure open ai key

What is the actual behavior?

It is not working with azure openai key

Slack conversation about issue: https://weaviate.slack.com/archives/C02RRQP23K3/p1712830332189929

Supporting information

No response

Server Version

1.24.8

Code of Conduct

• I have read and agree to the Weaviate's Contributor Guide and Code of Conduct

[andrewisplinghoff]

Probably not related to the actual problem, but I do not think that you can pass baseURL, deploymentId and resourceName as query parameters: https://weaviate.io/developers/weaviate/modules/retriever-vectorizer-modules/text2vec-openai#query-time-parameters

These settings related to Azure OpenAI are configured on the collection level:
https://weaviate.io/developers/weaviate/modules/retriever-vectorizer-modules/text2vec-openai#api-settings-azure-openai

[mihir-kanzariya]

in any case, it didn't work, I tried many combinations. Please follow slack conversation for better understanding of issue.

I guess there is a bug.

[rg1918]

Is there any update on this ?

[antas-marcin]

@mihir-kanzariya I think that I know why are you getting this error. I don't know how your class is configured but it looks like that Weaviate doesn't recognize it that it's an Azure configuration, did you define resourceName and deploymentId in your classes config?