-
Notifications
You must be signed in to change notification settings - Fork 876
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
TestAccAwsPyEc2Provisioners test failing #1491
Comments
#1492 pins the Python version to fix the test failure |
Hi @scottslowe thanks so much, forgive me for not finding time to contribute a workaround here. it makes total sense for you to keep this repo with passing tests and pinning the version is 👍 I've created an issue for my team so we can close this one without losing track of the actual breaking change. I don't think we got to a good conclusion of whether the new behavior is correct or needs to be reverted in Pulumi-command. Thanks again and sorry for the disruption. |
What happened?
TestAccAwsPyEc2Provisioners test is failing in CI on master.
I've looked into this and here's the problem. There's two versions of this program, one in TS and one in Python.
With recent changes to pulumi-command, if you are on 0.8.2, the following program:
now marks catConfig.Stdout as a secret output whereas it didn't before.
Now, Node test pins Pulumi-command 0.0.3 or some such, while Python test floats to the latest reference. Therefore Node test passes but Python test fails.
The specific failure is a panic when reading the secret value in ProgramTest Go code. There's an assert that tries to convert catConfigStdout to a string, but it is now a map in Go land. AFAIK Go ProgramTest cannot yet peek inside secret values returned from the stack.
Expected Behavior
Passing test.
Steps to reproduce
See above.
Output of
pulumi about
N/A
Additional context
We had some discussion and dug up this change https:/pulumi/pulumi-go-provider/pull/105/files#diff-4bd4b3f761d7688f63a5a8e766057562d08e3df2f51cfa88e604cb4e353fd042R318 as possible root cause.
Need to decide if this semantic is reasonable for Command and we want to keep it or revert back. If keeping this semantic, we need to adjust the test here to not fail on receiving a secret.
Contributing
Vote on this issue by adding a 👍 reaction.
To contribute a fix for this issue, leave a comment (and link to your pull request, if you've opened one already).
The text was updated successfully, but these errors were encountered: