How to Troubleshoot a Vertex AI HTTP Connection Error

[derikfernandes]

I am trying to connect via HTTP request to Google Vertex AI and can successfully connect for a while. However, after a few hours (more than 12 hours), the following error appears:

{ "error": { "code": 401, "message": "Request is missing required authentication credential. Expected OAuth 2 access token, login cookie or other valid authentication credential. See https://developers.google.com/identity/sign-in/web/devconsole-project.", "status": "UNAUTHENTICATED", "details": [ { "@type": "type.googleapis.com/google.rpc.ErrorInfo", "reason": "CREDENTIALS_MISSING", "domain": "googleapis.com", "metadata": { "method": "google.cloud.aiplatform.v1.PredictionService.GenerateContent", "service": "aiplatform.googleapis.com" } } ] } }

oauth { "error": "invalid_grant", "error_description": "reauth related error (invalid_rapt)", "error_uri": "https://support.google.com/a/answer/9368756", "error_subtype": "invalid_rapt" } oauth@access_token

The method I am using involves creating credentials within the console. I select the Vertex AI API and create credentials with the following characteristics:

(https://i.sstatic.net/YFVaaIsx.png)

Then, I go to OAuth Playground: https://developers.google.com/oauthplayground/

I select everything that is listed under Vertex AI API.

(https://i.sstatic.net/pKZbbgfg.png)

I choose the following items in the settings:

(https://i.sstatic.net/4baTrcLj.png)

• I input the OAuth Client ID and OAuth Client Secret.

• I click on "Exchange authorization code for tokens."

• I save the Refresh Token and the Access Token.

Within the platform where I will make the HTTP request, I perform the following actions:

1º Create the following HTTP request:

Method: POST URL: https://accounts.google.com/o/oauth2/token

JSON (BODY)

{ "client_secret": "XXXXXXXXXXX", "grant_type": "refresh_token", "refresh_token": "XXXXXXXXXXX", "client_id": "XXXXXXXXXXXXXXX" } Variable for the response body: oauth

2º Create the second HTTP request:

Method: Post

URL: https://us-central1-aiplatform.googleapis.com/v1/projects/crias-mvp/locations/us-central1/publishers/google/models/gemini-1.5-pro-preview-0409:streamGenerateContent

HEADERS:: Authorization Bearer {{oauth@access_token}} Content-Type application/json

JSON(BODY): {{payload}}

Name response variable: JSON

Afterward, I simply extract the response from the JSON and process it.

As I mentioned earlier, it works initially, but after about 12 hours, it stops working.