When using Deploy, sometimes tasks end up in Failed or Unknown state and pollute the system. They also prevent us from doing an upgrade.
It's possible to manually abort and delete those tasks, but that is a tedious manual process when there are a lot of them.
It would be great if there was a built-in cleanup mechanism for these 'ghost tasks'
by: DutchCustomerDay ㉔. | 2 months ago | Deployments
Comments
We have some follow up questions regarding this request:
1/ Would you like Deploy to automatically delete tasks in the Unknown or Failed state permanently? If so, could you clarify why and when this should happen (e.g., immediately after they occur, or after a specific period)?
2/ Should all Failed tasks be handled in the same way, or do you see value in differentiating between types of failures?
3/ Would you prefer this cleanup mechanism to focus solely on Unknown or Failed tasks, or would you consider extending it to include all tasks? (Note: We currently offer archiving and purging policy configurations.)
4/ Can you explain why keeping these tasks (including Unknown and Failed) in the Archive Database is not a viable solution for your use case?
5/ Could you describe the specific pain points caused by these tasks? For instance, how often do you encounter them, and what is the typical volume of Failed or Unknown tasks you handle daily or weekly?
1.) no, after 2 days. We need to give the developer time to determine and solve the problem he/she has during the deployment
2.) yes , all in the same way
3.) No, also on task with the state : Pending
4.) Once they are in archiving nobody will see the urge to solve it and they will kept tehere for years
5.) we run into this on daily base.
Our company is : UWV netherlands