Understanding AWS lambda retry mechanism

Question

My use case : From the spring-boot application, I am publishing a payload to AWS SNS, this SNS is triggering the Lambda function.

If the lambda function fails, is there any configuration available on AWS lambda where we can specify the number of retries and the duration after which each retry happens?

Answer 1

https://docs.aws.amazon.com/sns/latest/dg/sns-message-delivery-retries.html specifies the retry policies for Lambda in case the Lambda service itself is unavailable.

Immediate retry (no delay) phase	Pre-backoff phase	Backoff phase	Post-backoff phase	Total attempts
3 times, without delay	2 times, 1 second apart	10 times, with exponential backoff, from 1 second to 20 seconds	100,000 times, 20 seconds apart	100,015 times, over 23 days

And also notes

With the exception of HTTP/S, you can't change Amazon SNS-defined delivery policies. Only HTTP/S supports custom policies. See Creating an HTTP/S delivery policy.

Therefore no, you cannot change the retry behaviour of SNS for Lambda invocations.

There are however two things you can do:

• you can subscribe an SQS queue to SNS and then let you lambda get triggered by that queue. On SQS you can configure the retry behaviour to some extent: https://docs.aws.amazon.com/lambda/latest/operatorguide/sqs-retries.html .
• you can change the async invocation configuration on the lambda and set the number of retries between 0 and 2. See https://docs.aws.amazon.com/lambda/latest/dg/invocation-async.html#invocation-async-errors

Answer 2

Sadly, you can't control retry policies as explained in the docs:

With the exception of HTTP/S, you can't change Amazon SNS-defined delivery policies.