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
There are certain situations that we can catch prior to deployment that would result in a Flow not behaving as expected. In particular, if a user tries to deploy a Flow which:
has a task with retry settings
that task also has a keyed upstream edge
the upstream task has no result handler
Then retries will fail to work in Cloud. Because this situation can be inferred apriori we should raise an informative error during the Docker health check.
Similarly, if a user has tasks with checkpoint=True that don't have corresponding result handlers, we should raise an error.
Note that if the Flow itself has a result handler (but the Task does not), then all of this is OK because the Flow's should be the fallback.
The text was updated successfully, but these errors were encountered:
There are certain situations that we can catch prior to deployment that would result in a Flow not behaving as expected. In particular, if a user tries to deploy a Flow which:
Then retries will fail to work in Cloud. Because this situation can be inferred apriori we should raise an informative error during the Docker health check.
Similarly, if a user has tasks with
checkpoint=True
that don't have corresponding result handlers, we should raise an error.Note that if the Flow itself has a result handler (but the Task does not), then all of this is OK because the Flow's should be the fallback.
The text was updated successfully, but these errors were encountered: