AWS Lambda: 'MemorySize' value failed to satisfy constraint

Question

I have a really weird problem while trying to change Lambda's memory size. It's clearly stated in the UI that we can Set memory to between 128 MB and 10240 MB, but when I set 8192 value and click Apply, I am getting the following error message:

'MemorySize' value failed to satisfy constraint: Member must have value less than or equal to 3008

Already checked the Quotas and it's all good there. Honestly, I am stuck

Thanks

Answer 1

It seems new accounts have restricted memory and concurrency quotas. From the docs:

"Important New AWS accounts have reduced concurrency and memory quotas. AWS raises these quotas automatically based on your usage. You can also request a quota increase."

Answer 2

I got an answer from AWS support:

Lambda functions with memory configuration greater than 3GB are currently unavailable for first time use in multiple regions. We are working on restoring this feature and mitigations are in progress. If you urgently require to use your function with memory greater than 3GB, please provide your account and region details so we can expedite access internally.

Answer 3

I do have the same problem here. I am using a Python 3.8 runtime in Frankfurt (eu-central-1). I also checked with other runtimes and London (eu-west-2) - same error.

However if I create a function in Ireland (eu-west-1) or North Virginia (us-east-1) the limit is at the expected limit of 10240 MB. I assume that there are (temporary?) provisioning issues with the other regions.

Answer 4

3008 MB is the previous memory limit. Check out this announcement. Note that not all AWS regions support this 10 GB of memory yet. For example, I tried setting memory to 10240 MB in Asia Pacific (Hong Kong) region, I got the same error.