Error When Creating Team via API When Omitting ExternalSecurityGroups

Hi all,

We’ve recently upgraded our Octopus Server version from 2022.3.10664 to 2023.2.13113 and are experiencing some issues with creating new teams via our automation. We heavily utilize the terraform provider and prior to this upgrade we had no issues excluding the “external_security_groups” parameter for teams. Post upgrade we are seeing the following error:

Error: octopus deploy api returned an error on endpoint /api/teams - [The ExternalSecurityGroups field is required.]

Our understanding is that this is an optional field; however, the terraform provider results in an error when we do not specify any groups.

Based on our investigation, the API is expecting the ExternalSecurityGroups field to be in the request, but be an empty array when no groups are specified. The terraform provider behavior is to omit the field from the request entirely.

Versions:
Octopus Server: 2023.2.13113
Terraform Provider: 0.7.61 (We can’t upgrade this because it breaks everything)

Any help is appreciated!

Regards,
Evan Cleary

Hi @evan.cleary,

Thanks for reaching out, welcome to the Octopus community!

Cheers for all the details provided, I’ve managed to reproduce the issue using the same versions specified and have confirmed the issue doesn’t exist on the latest version of our Terraform Provider so I believe you are running into this closed issue: team resource failing · Issue #191 · OctopusDeployLabs/terraform-provider-octopusdeploy · GitHub

Updating to latest versions is always recommended as it brings the latest features/bug fixes, but I understand not wanting to modify working processes. Are you able to please share the issues that arise when using a newer TF provider version? Are you seeing any specific errors or loss of functionality?

Feel free to reach out with any questions at all!

Best Regards,

Hey @finnian.dempsey,

We’ve managed to get the one part of the configuration that manages the teams updated to the latest provider while leaving our other projects untouched, so we are unblocked for now.

The biggest issues we’ve come into with updating the provider pertains to loss of some functionality on certain deployment process actions as they’ve gained explicit blocks. One example:

Other problems we’ve run into is having certain resources generate continuous diffs which makes reviewing our plans overly cumbersome, example:

We’ve found workarounds for some of these over time. We haven’t attempted an upgrade in some time, so a bunch of what we’ve experienced in the past may be resolved at this point. With the success we had with this one portion, we’ll probably set some time aside to try and get the rest of the system up to speed.

2 Likes

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.