No matter how you like to participate in developer communities, Pulumi wants to meet you there. If you want to meet other Pulumi users to share use-cases and best practices, contribute code or documentation, see us at an event, or just tell a story about something cool you did with Pulumi, you are part of our community.

Pulumi Community

Are you using the default AWS provider, or creating one? If you're creating one, are you passing in the correct region?

Ok then we're into the weird-side-cases part of the investigation :slightly_smiling_face: Are you running pulumi only from your local CLI? In particular, are you running it from a test framework, from a script or other wrapper code, or from a cloud pipeline or similar?

I'm running it from a pipeline (where the deployment failed) and trying to destroy from my local cli before rerunning the pipeline. I'm using the s3 backend

(specifically amazon codepipeline with codebuild to run pulumi)

And is the region set from your stack config file? (Pulumi.&lt;stack&gt;.yaml)? If it is, have you verified that the file in the pipeline looks the way you expect? There wouldn't be a line termination issue, for example?

Yep, looks correct in both places. It has config: aws:region: us-west-2, and the line endings are LF everywhere

What's weird is that I only see this when destroying from my local box--the deploy failed for unrelated reasons, and I'd already destroyed and re-created this stack once before

We're in really early stages of bringing up our CI/CD

Bit confused: earlier you said "I'm running it from a pipeline (where the deployment failed)", but you've just said "I only see this when destroying from my local box". Is the pipeline all good, wrt this specific problem?

Hey <@U01KT4S8TNG> are you sure that the stack you are destroying and the config file are 1:1 in their name convention?

Yeah, the pipeline is all good, my local cli is where I'm seeingthis

I wonder if you need to Pulumi stack select the correct stack name

<@UHTP242EB> yes, I'm sure. They're both the same file name.

And the currently selected stack is the same name

All three are {companyname}-platform-staging

Can you run “pulumi stack select” and make sure you can see the stack

And when I run just pulumi stack, I see the resources that failed to delete

Looks to me like you've thought of everything. I'm stumped. Maybe clone to a fresh directory and try again? Or remove ~/.pulumi and see if that helps?

Just to close the loop here, I was eventually able to just roll forward and the problem took care of itself. Still, seems like a scary time.

Roll forward to a newer version of Pulumi?

Roll forward with a newer version of my infrastructure using pulumi up. The resources that couldn't delete were still tracked, so it all worked.