Open Aashw1n opened 2 weeks ago
Hi @Aashw1n - thanks for reaching out.
The errorERR_SOCKET_CONNECTION_TIMEOUT
means that your application is unable to establish a connection with the Bedrock service within the specified timeout period. This could be due to several reasons including network issues, misconfigured AWS credentials or incorrect service endpoint settings.
BedrockRuntimeClient
instance. For example:
const client = new BedrockRuntimeClient({
credentials: {
....
},
region: "us-west-2",
socketTimeout: 120000 // Increase the timeout to 120 seconds (2 minutes)
});
Hope it helps! Best, John
I'm trying to access the llama 3.1 405B model on AWS bedrock. Here's the code in my Sveltekit server side file: ` import { json } from '@sveltejs/kit';
import {PRIVATE_AWS_ACCESS_KEY, PRIVATE_AWS_SECRET_KEY} from '$env/static/private'
import { BedrockRuntimeClient, InvokeModelCommand } from "@aws-sdk/client-bedrock-runtime";
const client = new BedrockRuntimeClient({ credentials:{ accessKeyId:PRIVATE_AWS_ACCESS_KEY, secretAccessKey: PRIVATE_AWS_SECRET_KEY},region: "us-west-2" });
console.log(client)
const modelId = "meta.llama3-1-405b-instruct-v1:0"; /* @type {import('./$types').RequestHandler} /
/* @type {import('./$types').PageLoad} / export async function load() { try{
const userMessage =
Tell me about donuts
const prompt =
<|begin_of_text|> <|start_header_id|>user<|end_header_id|> ${userMessage} <|eot_id|> <|start_header_id|>assistant<|end_header_id|>
;// Format the request payload using the model's native structure. const request = { prompt, // Optional inference parameters: max_gen_len: 1024, temperature: 0.7, top_p: 0.9, }
const response = await client.send( new InvokeModelCommand({ contentType: "application/json", accept: "application/json", body: JSON.stringify(request), modelId, }), );
// Decode the native response body. /* @type {{ generation: string }} / const nativeResponse = JSON.parse(new TextDecoder().decode(response.body));
// Extract and print the generated text. const responseText = nativeResponse.generation;
console.log(responseText); return {resp:responseText} }catch(err){ console.log(err) return{message: "This isn't working"} }
}`
All I'm able to get is this:
Error [ERR_SOCKET_CONNECTION_TIMEOUT]: Socket connection timeout at new NodeError (node:internal/errors:399:5) at internalConnectMultiple (node:net:1099:20) at Timeout.internalConnectMultipleTimeout (node:net:1638:3) at listOnTimeout (node:internal/timers:575:11) at process.processTimers (node:internal/timers:514:7) { code: 'ERR_SOCKET_CONNECTION_TIMEOUT', '$metadata': { attempts: 1, totalRetryDelay: 0 } }
It does work on a separate, simple typescript project that only calls the model and prints.
Anything, even a clue as to what could be causing this will be a huge help.
Thank you.