I am running into a bug, I hope someone can help me here. here is a scenario: I have a lambda function which is running on AWS for 4 mins, with a trigger rate set to 1 min. so at any given time i have 4 lambda functions running, now I was asked to increase this to 16 functions at any given time, since the lowest rate I can set is 1 min, and not in seconds, I used the name property of function to deploy the same function again with a different name. function is successfully deployed and triggered, however the cloudFormation stack is stuck with the message ‘CloudFormation is waiting for NetworkInterfaces associated with the Lambda Function to be cleaned up.’ The entire process is wrapped in jenkins job, so the jenkins job is also stuck in executing state.
Sounds like you’re using VPC-based Lambdas?
There have been issues with cleaning up the ENIs that are allocated to VPC-based Lambdas in the past - it used to just not do it if any traffic had been sent through the ENI in last 12-24 hours . That error message suggests that they might now clean themselves up!
If possible, I would recommend waiting a day and seeing what happens/trying to clean it up again.
If you can’t wait that long, you should log a support request with AWS.
Other weird part is the existing lambdas are getting deleted. so lets say if I deployed lambda with function name ‘sqs2kafka-set-0-functionName’ and if I redploy the function again with name ‘sqs2kafka-set-1-functionname’ the previous one gets deleted.
Hmm, not sure why that would be - Lambda function names need to be unique. Might be something to do with the generated CloudFormation from SLS.