- Newest
- Most votes
- Most comments
Hello,
The error message "LAMBDA_RUNTIME Failed to get next invocation. No Response from endpoint" suggests a potential problem with AWS Lambda service in the US-east-1 region.
-
Inconsistently failing Lambda in US-east-1 likely has a regional issue. Check AWS Service Health (https://docs.aws.amazon.com/health/latest/ug/aws-health-dashboard-status.html) for Lambda problems in US-east-1.
-
If no health issues are reported, consider IAM permission differences between your Lambda functions in ap-south-1 and US-east-1.
Permissions: Ensure your Lambda function in US-east-1 has the necessary permissions to perform its tasks. Double-check IAM policies for any discrepancies between regions.
Are any calls made in the handler method or the function's initialisation code (code executed without the handler method being called) to an AWS or external service, particularly to a regional AWS service located in ap-south-1 or elsewhere outside us-east-1, such as an S3 bucket, Secrets Manager secret, DynamoDB table, or STS service endpoint (such as sts.ap-south-1.amazonaws.com) for assuming a role?
Hi, We are not executing any code outside the handler. There aren't any calls being made to different region (in this case ap-south-1) as well. The lambda function role is also regional and is different and new for each regions.
Does the function access any external resources, AWS services, DNS names, or anything other than the raw input being passed to the function? You mentioned that you're seeing the Lambda function timing out, so it's almost certain that there is a dependency on something outside the function and that's where the problem also likely lies. It could also be a local resource, if it's configured differently or has different data contents between the region where the function is working and the other region where it isn't.
Hi Leo, There is no timeout log as well.
It only logs INIT_START message and there is nothing after that.
We have moved to us-east-2 and the lambda streaming seems to be working without any issues in that region.
Hi,
Do you use a Custom Runtime (i.e. a container image that you built) to run your Lambda?
If yes, check that the Lambda in second region has proper access to ECR registry where image is stored.
Best,
Didier
We are using Node.js 20.x runtime. (No custom docker runtime). We are not using ECR as well as this is non docker version.
We have all the IAM permission same in both region's lambda functions. (In short same serverless.yml file with only difference is value of region)
Relevant content
- AWS OFFICIALUpdated 3 years ago
Thanks for the reply. We are using Node.js 20.x runtime. (No custom docker runtime). We are not using ECR as well as this is non docker version.
We have all the IAM permission same in both region's lambda functions. (In short same serverless.yml file with only difference is value of region)