How to use node redis with AWS Lambda ? Max number of clients reached

[TheUncharted]

Hi
I m using Redis cloud with the free tier with an AWS lambda that run a ft.search

I initialized my client like this at the start of the lambda file as a global variable

const redis = require("redis");
const client = createClient({
url: redis://${Config.REDIS_USER}:${Config.REDIS_PASSWORD}@${Config.REDIS_HOST},
});
client.on("error", (err) => console.log("Redis Client Error", err));
await client.connect();

When I was testing my app I noticed that I got an error :

Redis Client Error [ErrorReply: ERR max number of clients reached]

I tried multiple things like this article
https://dev.to/afrazkhan/reuse-redis-connecitons-across-aws-lambda-invocations-4kk

then I tried to use the isolationPoolOptions

const client = createClient({ url: redis://${Config.REDIS_USER}:${Config.REDIS_PASSWORD}@${Config.REDIS_HOST}, isolationPoolOptions: { min: 5, max: 50, idleTimeoutMillis: 120 * 1000, }, });

But when I try to load test this lambda with a performance test with postman with a ramp up scenario of 20 user: I reach the max connection in maybe one minute

Do you think I m using the right initialization code for redis ? or is it because of the free tiers limitation of 256 connection ?
In my mind, the client should reuse the connection with the isolation pool and since I m calling the lambda several time
the global variable client should be reused

Any idea on how to manage this ?

[TheUncharted]

I answering to my own question but redis cloud free tiers has only 30 connections
When I load testing the lambda will scale and create a pool for each lambda and if I put a max too big, the max connection of the free tiers will be reached :

N Lambdas x M max connection in the pool > 30 max connections limit