hey team, n00b pulumi question, when running

pulumi up

to deploy a stack, is there any rules/guidelines to which directory of a repo that cmd needs to be run from? e.g for the below structure, if I want to deploy resources defined in

./resources/s3/index.ts

where do I use

pulumi up?

resources
├── index.ts
└── s3
    └── index.ts

The same directory as your Pulumi.yaml and Pulumi.stack.yaml files.

right gotcha. Main reason this came about was that my outputs weren't getting printed to cli once I had moved

index.ts

from root dir into a deeper dir was wondering why/how the directory restructure would impact the outputs getting printed or not

You need to have your project entrypoint (index.ts) in the same directory as your project (Pulumi.yaml). The entrypoint can call to code anywhere. It could be in the same directory, other directories, NPM packages, whatever.

good call. I initially started with my main definitions in

index.ts

and am now splitting up resources per services, such as

./s3/index.ts
./cognito/index.ts
./lambda/index.ts

I wouldn't do that. You should group them based on deployment cycle. For example, if you need an S3 bucket for you VPC logging, and one for your webapp upload storage area, then every time you destroy an instance of your webapp upload bucket (once per integration test run, probably), then your VPC logging bucket will be destroyed.

argh, my confusion remains as to how best to organise the folders is there a recommended guide I can point to? I have a nested

index.ts

which I want to use to import all my other resource definitions, but unclear of the approach as I seem to be getting it wrong

Yes, Pulumi has published a couple. Here's the most recent: https://www.pulumi.com/blog/iac-recommended-practices-structuring-pulumi-projects/ There is at least one more, but I can't find it right now.