The multi-runner example is insufficient and conflicts with the multi-runner code in the main branch

[bob-ashforth-sp]

I have tried to get the multi-runner feature working, but after finding numerous conflicts with no guidance on "what works," I'm abandoning the effort. It is a laudable improvement, but needs a working example. I also think that it should support multiple distinct labels paired with the current option of multiple machine types.

[npalm]

This module is a community effort on best effor. Sorry it not works for you. The multie runner example is the leading one use for running test. So we are not aware of any problems.

[bob-ashforth-sp]

Thanks for the reply, @npalm . If the module is working, more documentation and a full working example would support its use. As it stands, the code in the example does not match the code in main, and if it is supposed to take the place of the runner module, that should be stated. When I tried removing the invocation of the runner module from main.tf, terraform could not complete cleanly.

[npalm]

Did you follow this example? https://github-aws-runners.github.io/terraform-aws-github-runner/examples/multi-runner/

I use this example with a local build (cd lambdas && yarn %% yarn dist). Next run the terraform commands as listed here. It requires to setup you app and update enviroment variables as mentioned in the general docs.

Something like.

export TF_VAR_github_app='{"id": '$APP_ID', "key_base64": "'$APP_PRIVATE_KEY'"}'