Skip to content

About error handling when using a custom connector as a destination #68

@yumaishida

Description

@yumaishida

We are currently implementing a custom connector.
Please tell me about error handling when using a custom connector as a destination.

In the "Configure flow" step of AppFlow, following configuration items related to error handling would appear on the screen (Use S3 as the source) .

image

(Question)
There are two options "Stop the current flow run" and "Ignore and continue the flow run", are these options handled by the AppFlow infrastructure and not something we need to be aware of in the custom connector?
Also, if it is handled on the AppFlow infrastructure side, could you please tell me exactly how each option works?

(Background)

It appears that com.amazonaws.appflow.custom.connector.connector.model.writeWriteDataRequest does not hold that configuration item, and I believe there is no way for the custom connector implementation to know which one the user has chosen. Therefore, I did not think it was a setting item to be handled on the custom connector side.


I tried both options and could not understand the difference in behavior. Both seemed to have the "Ignore and continue the flow run" behavior (In both cases, when returning a WriteDataResponse with isSucecss false and when throwing an Exception, the process seemed to stop).

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions