You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It would be beneficial to the developer experience if in-progress deployments were kept track of, and were able to be resumed at a given point if a step fails.
Right now, if you have a script that deploys many interlinked contracts and one of the deployments / actions fails, the script fails entirely with dangling contracts in an unknown state. With the deployment observer, the latest step and state would be known and could be resumed before the failure.
This would likely use the observer pattern where each step executing an action using the Boot ContractInstance would increment the step counter and save the state at that step. If a deployment "resumes" (via flag or otherwise) it would see that the step counter is > 0 and resume after the latest successful step.
The text was updated successfully, but these errors were encountered:
It would be beneficial to the developer experience if in-progress deployments were kept track of, and were able to be resumed at a given point if a step fails.
Right now, if you have a script that deploys many interlinked contracts and one of the deployments / actions fails, the script fails entirely with dangling contracts in an unknown state. With the deployment observer, the latest step and state would be known and could be resumed before the failure.
This would likely use the observer pattern where each step executing an action using the Boot
ContractInstance
would increment the step counter and save the state at that step. If a deployment "resumes" (via flag or otherwise) it would see that the step counter is > 0 and resume after the latest successful step.The text was updated successfully, but these errors were encountered: