I am following below blueprint for creating openai embedding connector with opensearch 2.11
# OpenAI connector blueprint example for embedding model
## 1. Create connector for OpenAI embedding model:
Refer to OpenAI [official doc](https://platform.openai.com/docs/guides/embeddings).
If you are using self-managed Opensearch, you should supply OpenAI API key:
```json
POST /_plugins/_ml/connectors/_create
{
"name": "<YOUR CONNECTOR NAME>",
"description": "<YOUR CONNECTOR DESCRIPTION>",
"version": "<YOUR CONNECTOR VERSION>",
"protocol": "http",
"parameters": {
"model": "text-embedding-ada-002"
},
"credential": {
"openAI_key": "<PLEASE ADD YOUR OPENAI API KEY HERE>"
show original
when I execute below request it error out with “Request failed to get to the server (status code: 502)”
POST /_plugins/_ml/connectors/_create
{
"name": "text-embedding-3-small",
"description": "text-embedding-3-small",
"version": "1",
"protocol": "https",
"parameters": {
"model": "text-embedding-3-small",
"endpoint": "api.openai.com"
},
"credential": {
"openAI_key": "api-key"
},
"actions": [
{
"action_type": "predict",
"method": "POST",
"url": "${parameters.endpoint}/v1/embeddings",
"headers": {
"Authorization": "Bearer ${credential.openAI_key}"
},
"request_body": "{ \"input\": ${parameters.input}, \"model\": \"${parameters.model}\" }",
"pre_process_function": "connector.pre_process.openai.embedding",
"post_process_function": "connector.post_process.openai.embedding"
}
]
}
It seems that in case of AWS Opensearch Service we have to provide role and secretto api-key
Now payload has been changed to
{
"name": "text-embedding-3-small",
"description": "text-embedding-3-small",
"version": "1",
"protocol": "https",
"parameters": {
"model": "text-embedding-3-small"
},
"credential": {
"secretArn": "arn:aws:secretsmanager:ap-northeast-1:xxxx:secret:openapikey-bRLaNi",
"roleArn": "arn:aws:iam::xxxx:role/opensearch-secretmanager-role"
},
"actions": [
{
"action_type": "predict",
"method": "POST",
"url": "https://api.openai.com/v1/embeddings",
"headers": {
"Authorization": "Bearer ${credential.secretArn.key}"
},
"request_body": "{ \"input\": ${parameters.input}, \"model\": \"${parameters.model}\" }",
"pre_process_function": "connector.pre_process.openai.embedding",
"post_process_function": "connector.post_process.openai.embedding"
}
]
}
But now it is erroring out with
no permissions for [cluster:admin/opensearch/ml/create_connector] and User [name=arn:aws:iam::xxxx:user/opensearch-user, backend_roles=