If the AWS NAT Gateway fails, is there a way to force replacement of that instance with Pulumi?
06/29/2022, 12:54 AM
That's AWS' problem, not yours. AWS' principle of shared responsibility states that you're responsible for the load / traffic that you create, and they're responsible for the services they provide. If their NAT fails, then it's on them to detect it and replace it.
If you provide your own NAT (e.g. on an EC2 instance in a public subnet in your VPC), then that's your responsibility to replace on failure.
06/29/2022, 3:16 PM
@little-cartoon-10569 This is an inquiry about whether Pulumi has a similar function to taint and untaint.
@little-cartoon-10569 In addition to failures, if the public IP is sometimes blocked when a client connects, a new NAT gateway may need to be created.
06/29/2022, 9:00 PM
Not sure what taint is. If you want to force a recreate of a resource, you can change its (Pulumi) name.
That's force-recreating a resource. An easy way to achieve this is to change the first parameter to the resource constructor. This will make Pulumi think that the old resource needs to be deleted, and a new resource needs to be created. Which is the same outcome.