Getting 403 on chat ui config for aws sagemaker endpoint

[nauts]

Hi All,

Looking into configuring chat ui with aws sagemaker endpoint and getting following error:

DOTENV_LOCAL was found in the ENV variables. Creating .env.local file.
{"level":30,"time":1727231014113,"pid":23,"hostname":"fbe21dc3ad38","msg":"Starting server..."}
{"level":30,"time":1727231014147,"pid":23,"hostname":"fbe21dc3ad38","msg":"[MIGRATIONS] Begin check..."}
{"level":30,"time":1727231014175,"pid":23,"hostname":"fbe21dc3ad38","msg":"[MIGRATIONS] \"Update search assistants\" already applied. Skipping..."}
Listening on 0.0.0.0:3000
{"level":30,"time":1727231014175,"pid":23,"hostname":"fbe21dc3ad38","msg":"[MIGRATIONS] \"Update deprecated models in assistants with the default model\" should not be applied for this run. Skipping..."}
{"level":30,"time":1727231014175,"pid":23,"hostname":"fbe21dc3ad38","msg":"[MIGRATIONS] \"Add empty 'tools' record in settings\" already applied. Skipping..."}
{"level":30,"time":1727231014175,"pid":23,"hostname":"fbe21dc3ad38","msg":"[MIGRATIONS] \"Convert message updates to the new schema\" already applied. Skipping..."}
{"level":30,"time":1727231014175,"pid":23,"hostname":"fbe21dc3ad38","msg":"[MIGRATIONS] \"Convert message files to the new schema\" already applied. Skipping..."}
{"level":30,"time":1727231014175,"pid":23,"hostname":"fbe21dc3ad38","msg":"[MIGRATIONS] \"Trim message updates to reduce stored size\" already applied. Skipping..."}
{"level":30,"time":1727231014175,"pid":23,"hostname":"fbe21dc3ad38","msg":"[MIGRATIONS] \"Reset tools to empty\" already applied. Skipping..."}
{"level":30,"time":1727231014175,"pid":23,"hostname":"fbe21dc3ad38","msg":"[MIGRATIONS] All migrations applied. Releasing lock"}
{"level":30,"time":1727231014207,"pid":23,"hostname":"fbe21dc3ad38","minDate":"2024-09-25T00:00:00.000Z","dateField":"createdAt","span":"day","type":"conversation","msg":"Computing conversation stats"}
{"level":30,"time":1727231014216,"pid":23,"hostname":"fbe21dc3ad38","minDate":"2024-09-25T00:00:00.000Z","dateField":"updatedAt","span":"day","type":"conversation","msg":"Computing conversation stats"}
{"level":30,"time":1727231014219,"pid":23,"hostname":"fbe21dc3ad38","minDate":"2024-09-25T00:00:00.000Z","dateField":"createdAt","span":"day","type":"message","msg":"Computing conversation stats"}
{"level":30,"time":1727231014220,"pid":23,"hostname":"fbe21dc3ad38","minDate":"2024-09-22T00:00:00.000Z","dateField":"updatedAt","span":"week","type":"conversation","msg":"Computing conversation stats"}
{"level":30,"time":1727231014224,"pid":23,"hostname":"fbe21dc3ad38","minDate":"2024-09-22T00:00:00.000Z","dateField":"createdAt","span":"week","type":"conversation","msg":"Computing conversation stats"}
{"level":30,"time":1727231014227,"pid":23,"hostname":"fbe21dc3ad38","minDate":"2024-09-01T00:00:00.000Z","dateField":"createdAt","span":"month","type":"message","msg":"Computing conversation stats"}
{"level":30,"time":1727231014229,"pid":23,"hostname":"fbe21dc3ad38","minDate":"2024-09-25T00:00:00.000Z","dateField":"createdAt","span":"day","type":"conversation","msg":"Computed conversation stats"}
{"level":30,"time":1727231014229,"pid":23,"hostname":"fbe21dc3ad38","minDate":"2024-09-25T00:00:00.000Z","dateField":"updatedAt","span":"day","type":"conversation","msg":"Computed conversation stats"}
{"level":30,"time":1727231014230,"pid":23,"hostname":"fbe21dc3ad38","minDate":"2024-09-25T00:00:00.000Z","dateField":"createdAt","span":"day","type":"message","msg":"Computed conversation stats"}
{"level":30,"time":1727231014230,"pid":23,"hostname":"fbe21dc3ad38","minDate":"2024-09-22T00:00:00.000Z","dateField":"updatedAt","span":"week","type":"conversation","msg":"Computed conversation stats"}
{"level":30,"time":1727231014231,"pid":23,"hostname":"fbe21dc3ad38","minDate":"2024-09-22T00:00:00.000Z","dateField":"createdAt","span":"week","type":"message","msg":"Computing conversation stats"}
{"level":30,"time":1727231014235,"pid":23,"hostname":"fbe21dc3ad38","minDate":"2024-09-01T00:00:00.000Z","dateField":"createdAt","span":"month","type":"message","msg":"Computed conversation stats"}
{"level":30,"time":1727231014236,"pid":23,"hostname":"fbe21dc3ad38","minDate":"2024-09-22T00:00:00.000Z","dateField":"createdAt","span":"week","type":"conversation","msg":"Computed conversation stats"}
{"level":30,"time":1727231014236,"pid":23,"hostname":"fbe21dc3ad38","minDate":"2024-09-22T00:00:00.000Z","dateField":"createdAt","span":"week","type":"message","msg":"Computed conversation stats"}
{"level":30,"time":1727231014238,"pid":23,"hostname":"fbe21dc3ad38","minDate":"2024-09-01T00:00:00.000Z","dateField":"createdAt","span":"month","type":"conversation","msg":"Computing conversation stats"}
{"level":30,"time":1727231014239,"pid":23,"hostname":"fbe21dc3ad38","minDate":"2024-09-01T00:00:00.000Z","dateField":"updatedAt","span":"month","type":"conversation","msg":"Computing conversation stats"}
{"level":30,"time":1727231014242,"pid":23,"hostname":"fbe21dc3ad38","minDate":"2024-09-01T00:00:00.000Z","dateField":"createdAt","span":"month","type":"conversation","msg":"Computed conversation stats"}
{"level":30,"time":1727231014243,"pid":23,"hostname":"fbe21dc3ad38","minDate":"2024-09-01T00:00:00.000Z","dateField":"updatedAt","span":"month","type":"conversation","msg":"Computed conversation stats"}
{"level":50,"time":1727231020662,"pid":23,"hostname":"fbe21dc3ad38","err":{"type":"TypeError","message":"fetch failed: getaddrinfo ENOTFOUND raw.githubusercontent.com","stack":"TypeError: fetch failed\n    at node:internal/deps/undici/undici:13178:13\n    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)\n    atasync Promise.all (index 1)\n    at async Promise.all (index 0)\n    at async file:///app/build/server/chunks/index3-BP6IVcLY.js:174:63\ncaused by: Error: getaddrinfo ENOTFOUND raw.githubusercontent.com\n    at GetAddrInfoReqWrap.onlookupall [as oncomplete] (node:dns:120:26)"},"msg":"Failed to initialize PlaywrightBlocker from prebuilt lists"}
Using a model URL is deprecated, please use the `endpointUrl` parameter instead
Using a model URL is deprecated, please use the `endpointUrl` parameter instead
{"level":50,"time":1727231032591,"pid":23,"hostname":"fbe21dc3ad38","err":{"type":"Error","message":"Server response contains error: 403","stack":"Error: Server response contains error: 403\n    at streamingRequest (file:///app/node_modules/@huggingface/inference/dist/index.js:334:11)\n    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)\n    at async textGenerationStream (file:///app/node_modules/@huggingface/inference/dist/index.js:715:3)\n    at async generate (file:///app/build/server/chunks/_server.ts-D8UXwvVE.js:438:20)\n    at async textGenerationWithoutTitle (file:///app/build/server/chunks/_server.ts-D8UXwvVE.js:503:3)"},"msg":"Server response contains error: 403"}
{"level":50,"time":1727231032595,"pid":23,"hostname":"fbe21dc3ad38","err":{"type":"Error","message":"Server response contains error: 403","stack":"Error: Server response contains error: 403\n    at streamingRequest (file:///app/node_modules/@huggingface/inference/dist/index.js:334:11)\n    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)\n    at async textGenerationStream (file:///app/node_modules/@huggingface/inference/dist/index.js:715:3)\n    at async generateFromDefaultEndpoint (file:///app/build/server/chunks/index3-BP6IVcLY.js:1047:20)\n    at async generateTitle (file:///app/build/server/chunks/_server.ts-D8UXwvVE.js:198:10)\n    at async generateTitleForConversation (file:///app/build/server/chunks/_server.ts-D8UXwvVE.js:162:19)"},"msg":"Server response contains error: 403"}

The .env.local file provide is following:

MONGODB_URL=mongodb://mongodb:27017
PLAYWRIGHT_ADBLOCK=false
MODELS=`[{
  "name":"llama-modelname",
  "displayName":"llama-modelname",
  "description":"test",
  "parameters":{
     "max_new_tokens":100
  },
  "endpoints":[
    {
      "type":"aws",
      "service":"sagemaker",
      "url":"https://runtime.sagemaker.ap-southeast-2.amazonaws.com/endpoints/llama-modelname/invocations",
      "weight":1
    }
  ]
}]`

As there is no accesskey and id provided in .env file , assumption is the it would resolve the access using the ec2 instance profile attached to ec2 where above chat-ui container is running. The ec2 instance have the right role to involve the endpoint mentioned in above config.

Image used : huggingface/chat-ui :latest

Could you please suggest if this configuration is good to perform dynamic credentials resolution? As mentioned here - #1419

Thank you in advance!