issue_comments: 1255603780
This data as json
| html_url | issue_url | id | node_id | user | created_at | updated_at | author_association | body | reactions | issue | performed_via_github_app |
|---|---|---|---|---|---|---|---|---|---|---|---|
| https://github.com/simonw/datasette/issues/1415#issuecomment-1255603780 | https://api.github.com/repos/simonw/datasette/issues/1415 | 1255603780 | IC_kwDOBm6k_c5K1v5E | 17532695 | 2022-09-22T22:06:10Z | 2022-09-22T22:06:10Z | NONE | This would be great! I just went through the process of figuring out the minimum permissions for a service account to run `datasette publish cloudrun` for [PUDL](https://github.com/catalyst-cooperative/pudl)'s [datasette deployment](https://data.catalyst.coop/). These are the roles I gave the service account (disclaim: I'm not sure these are the minimum permissions): - Cloud Build Service Account: The SA needs this role to publish the build on Cloud Build. - Cloud Run Admin for the Cloud Run datasette service so the SA can deploy the build. - I gave the SA the Storage Admin role on the bucket Cloud Build creates to store the build tar files. - The Viewer Role is [required for storing build logs in the default bucket](https://cloud.google.com/build/docs/running-builds/submit-build-via-cli-api#permissions). More on this below! The Viewer Role is a Basic IAM role that [Google does not recommend using](https://cloud.google.com/build/docs/running-builds/submit-build-via-cli-api#permissions): > Caution: Basic roles include thousands of permissions across all Google Cloud services. In production environments, do not grant basic roles unless there is no alternative. Instead, grant the most limited [predefined roles](https://cloud.google.com/iam/docs/understanding-roles#predefined_roles) or [custom roles](https://cloud.google.com/iam/docs/understanding-custom-roles) that meet your needs. If you don't grant the Viewer role the `gcloud builds submit` command will successfully create a build but returns exit code 1, preventing the script from getting to the cloud run step: ``` ERROR: (gcloud.builds.submit) The build is running, and logs are being written to the default logs bucket. This tool can only stream logs if you are Viewer/Owner of the project and, if applicable, allowed by your VPC-SC security policy. The default logs bucket is always outside any VPC-SC security perimeter. If you want your logs saved inside your VPC-SC perimeter, use your own bucket. See https://cloud.google.com/build/docs/securing-builds/store-manage-build-logs. ``` long stack trace... ``` CalledProcessError: Command 'gcloud builds submit --tag gcr.io/catalyst-cooperative-pudl/datasette' returned non-zero exit status 1. ``` You can store Cloud Build logs in a [user-created bucket](https://cloud.google.com/build/docs/securing-builds/store-manage-build-logs#store-custom-bucket) which only requires the Storage Admin role. However, you have to pass a config file to `gcloud builds submit`, which isn't possible with the current options for `datasette publish cloudrun`. I propose we add an additional CLI option to `datasette publish cloudrun` called `--build-config` that allows users to pass a [config file](https://cloud.google.com/build/docs/running-builds/submit-build-via-cli-api#running_builds) specifying a user create Cloud Build log bucket. | {"total_count": 1, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 1, "eyes": 0} | 959137143 |