- Newest
- Most votes
- Most comments
I don't think that this is the right approach. The blogs you mention talk about triggering a Lambda function in one account from Kinesis/DDB streams in a different account.
I think that in your case the best would be to have a Lambda function in the first account (DDB) which is triggered from the stream. Let this function run in a VPC in the first account. Peer that VPC to the VPC that is hosting Kafka in the second account. The Lambda function should be able to talk via the peered connection and ingest records into Kafka.
If the two accounts are in the same Organization maybe by using VPC sharing there is no need to do VPC peering. I did not test it so I am not sure it works as Lambda can run in VPCs in the same account. Not sure how Shared VPC are treated.
Just to answer the question about having two lambdas, Lambda function runs by default in an AWS managed VPC. If that Lambda tries to invoke another Lambda, i think it will remain in the same VPC. The second Lambda will be running in the seconds account's VPC, so there will be no internet traffic as far as I can tell.
Relevant content
- Accepted Answerasked a year ago
- Accepted Answerasked 6 months ago
- AWS OFFICIALUpdated a year ago
- AWS OFFICIALUpdated 2 years ago
- AWS OFFICIALUpdated 2 years ago
- AWS OFFICIALUpdated 2 years ago