Fail HCP datasources for revoked iteration #11854
Conversation
sylviamoss
changed the title
| @@ -96,7 +96,7 @@ require ( | |||
| github.com/hashicorp/packer-plugin-vmware v1.0.7 | |||
| github.com/hashicorp/packer-plugin-vsphere v1.0.5 | |||
| github.com/hashicorp/packer-plugin-yandex v1.1.1 | |||
| github.com/scaleway/packer-plugin-scaleway v1.0.5 | |||
| github.com/scaleway/packer-plugin-scaleway v1.0.4 | |||
There was a problem hiding this comment.
Tag v1.0.5 doesn't exist anymore
|
No, but I should definitely update the documentation. Let me do that and update the changelog! |
internal/registry/service.go
Outdated
| if !revokeAt.IsZero() && revokeAt.Before(time.Now().UTC()) { | ||
| // If RevokeAt is not a zero date and is before NOW, it means this iteration is revoked and should not be used | ||
| // to build new images. | ||
| return nil, fmt.Errorf("the iteration %s is revoked and can not be used on Packer builds", | ||
| resp.Payload.Iteration.ID) |
There was a problem hiding this comment.
I just realized that the error check for revoke_at is in the registry service code and not in the data source code itself. Is there any reason why you went that route as opposed to putting it in the data source?
There was a problem hiding this comment.
I was just reverting the first implementation and haven't thought too much about but I think it makes sense to put the check in the data source instead! So I updated the PR.
|
I'm going to lock this pull request because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues. |
HCP Packer will stop replacing the image id/url with
error_revokedso the hcp datasource will fail when querying revoked iterations. This reverts #11624