This message was deleted.

In theory, you could export the old stack, update the dependencies, and import it. But the risk of being affected by an incompatible API is very high given how far apart the versions are, so I wouldn't do this.

"just use another computer" yeah is great but what a pain, i don't want to have a trail of different computers/ec2/images for old deploys. is there no way to bring a stack onto a newer aws provider? just not sure how to upgrade a dev env/stack to use a newer

pulumi:providers:aws

- even with access to the old intel stack and installing all the latest pulumi versions, the stack is still pinned to use

3_30_1

instead of latest

5_10_0

.

@square-ability-48831 you can start a docker container with the right arch:

docker run --entrypoint bash --rm -it --platform linux/amd64 pulumi/pulumi

Then update the provider version in the stack

npm update @pulumi/aws

You’ll need to run a successful deploy with the old and new version to update the provider version

pulumi up

From then on, you can go back to your machine

thanks for the insight, guess i'll just go through the pain of destroying and recreating, glad i have the luxury for this project!

you can update aws provider without create/destroy

Wouldn't that require the current binaries to run? But they don't, because they're not built for the M1. At least, I thought that was the problem...

I would think you’d be able to use the older Intel binaries on the M1 (via Rosetta). Not the same, but I was accidentally running the Intel version of Go on my M1 for several months before I realized. 🤦🏻

I also made this mistake 😂