Error When Deploying Project

drees · 23 May 2023 14:31

I am getting the follow error when deploying one of our projects. I’ve included screenshots of the error below:

clare.martin · 23 May 2023 15:41

Good afternoon @drees,

Thank you for contacting Octopus Support and welcome to the forums!

I am sorry you are seeing this error when deploying a release, it does look like you may be running into this known bug -

github.com/OctopusDeploy/Issues

Automatic deployments can partially fail, causing a React error when viewing the release

opened 11:10PM - 31 May 22 UTC

closed 05:55AM - 16 Jun 22 UTC

susanpann

kind/bug state/committed

### Team - [X] I've assigned a team label to this issue ### Severity _N…o response_ ### Version reported on 2022.2.4955, 2022.2.4850 and occurs on latest ### What happened? When a release is automatically deployed via lifecycle rules it attempts a deployment immediately. This can fail on certain conditions: #### Prompted variables If a required prompted variable is part of the deployment, this automatic deployment will always fail. See how [tenanted deployments work with automatic lifecycle promotions](https://octopus.com/docs/releases/lifecycles#phases-with-environments). #### Untenanted and tenanted If the deployment is going to more than one tenant but deployment targets don't exist for a certain tenant then this will fail. The problem is that unlike creating a deployment via `POST /deployments`, the transaction doesn't get rolled back fully. This causes some issues on the UI: - If you try to open the server task from the release details page, you get `The resource 'Deployments-X' was not found.` ![image](https://user-images.githubusercontent.com/11970672/171763039-a4059040-c3ba-454c-8ff8-d4ea4c796332.png) - If you try to open the server task from the Tasks page, you get an empty page - If you create a successful deployment, the release details page then tries to read the task from each deployment on the release. Since the failed deployment has no task id the page blows up ### Reproduction - Create a lifecycle with one phase set to auto-deploy - Switch the project lifecycle to the newly created lifecycle - Add a variable to the project variable set - set it to prompt for a value and required - Create a release - The deployment will fail due to not having a value for the prompted variable - Deploy the release manually and supply a value for the prompted variable - Go back to the release details page - 💥 ### Error and Stacktrace ```shell An unexpected error occurred in Octopus v2022.2.4955: "TypeError: Cannot read properties of undefined (reading 'StartTime')" in DeploymentChildRow An unexpected error occurred in Octopus v2022.2.4850: "TypeError: Cannot read properties of undefined (reading 'State')" in PermissionCheck in LifecycleProgression ``` ### More Information Only happening in recent releases - older releases are ok. Other pages that show the release details are also fine (Tasks, Releases Overview) Internal Link: https://octopus.zendesk.com/agent/tickets/89742 Internal Link: https://octopus.zendesk.com/agent/tickets/89829 Internal Link: https://octopusdeploy.slack.com/archives/CNHBHV2BX/p1653526001934529 ### Workaround #### To avoid releases ending up in this state: - Switch to a manual deployment phase lifecycle - Check project tenanted deployment mode settings - if you only need tenanted deployments, don't select both untenanted and tenanted #### To clean up any releases that have been affected: via the DB: 1. Remove any server tasks related to bad deployments ``` DELETE FROM [dbo].[ServerTask] WHERE Id IN ( SELECT Id FROM [dbo].[ServerTask] WHERE JSON_VALUE([JSON], '$.Arguments.DeploymentId') IN ( SELECT Id from [dbo].[Deployment] WHERE TaskId is null) ) ``` 2. Remove any bad deployments ``` DELETE FROM [dbo].[Deployment] WHERE Id in (SELECT Id from [dbo].[Deployment] WHERE TaskId is null) ``` Alternatively, delete the deployment via the API: ``` Find all the deployments for the broken release GET api/{SpaceID}/releases/{ReleaseID}/deployments Delete any deployments which have a TaskId of null DELETE api/{SpaceID}/deployments/{DeploymentID}

Does this look like what you are experiencing?

If so there are a few workarounds in that GitHub issue to rectify this but you will need to delete that deployment unfortunately, my latest comment on that issue is the best way to fix this via the API and not the Database:

Alternatively, delete the deployment via the API:

Find all the deployments for the broken release
GET api/{SpaceID}/releases/{ReleaseID}/deployments

Delete any deployments which have a TaskId of null
DELETE api/{SpaceID}/deployments/{DeploymentID}

Let me know if that GitHub issue is not related to this though and we can take another look for you, the issue itself was fixed in 2022.3.2568 but you will still experience it if you upgrade, as you would need to delete that deployment.

Also of note in that issue if it is related:

To avoid releases ending up in this state:

Switch to a manual deployment phase lifecycle

Check project tenanted deployment mode settings - if you only need tenanted deployments, don’t select both untenanted and tenanted

Hopefully that helps but let us know if the issue is unrelated and we can take another look,
Kind Regards,
Clare

system · 23 June 2023 15:42

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