Skip to content

Add resiliency to the GET operation of AWS resource deployments #7352

Closed
@willdavsmith

Description

Overview of feature request

Sometimes in the functional tests (and potentially for users), we see a failed AWS deployment due to a GET call returning 404. We should investigate why the call fails and potentially add retries to resolve this.

Example: https://github.com/radius-project/radius/actions/runs/8311733036/job/22745893415
It looks like the deployment fails because it cannot find the bucket after it is created. Checking the console will show that the bucket is created. (it's left hanging)

My hypothesis is that AWS takes a minute to register the bucket as created sometimes.

Acceptance criteria

  • Retries for the GET call have been added

Additional context

Related issue: #5963

Would you like to support us?

  • Yes, I would like to support you

AB#11554

Metadata

Assignees

No one assigned

    Labels

    featureThis issue describes a feature request in RadiustriagedThis issue has been reviewed and triaged

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions