多次执行 AWS Lambda function 的 API 会引发“消息”：“端点请求超时”。 如何解决这个问题？

Question

Lambda Specs: Python Version - 3.7 || Memory - 10240MB (10GB)

The Synchronous API timeout limit is 30 seconds. My code executes for 4 seconds. The below code is a sample one that sleeps for 4 seconds. I'm using EFS mounted in Ubuntu for storing the packages as the limit of lambda deployment is 250MB.

import time
import json
import os
import sys 
sys.path.append("/mnt/access")
import math
import pandas as pd
import statsmodels.api as sm
import numpy as np
from datetime import datetime, timedelta
import sqlalchemy
from statsmodels.stats.outliers_influence import variance_inflation_factor

def lambda_handler(event, context):
    time.sleep(4)

1. This is a simple code integrated with API Gateway. It works fine when executed one or two times but when executed more than 10 times it runs into a timeout error.
2. I tried to use Provisioned Concurrency but it is so expensive. What to do to make it work all the time? Any Alternatives?

Answer 1

What is the AWS Lambda function timeout set at?

It sounds like the lambda function timeout is set too low.

https://console.aws.amazon.com/lambda/home - open the function and set the timeout under basic settings.