Skip to main content

Hi,

We have FME Server 2017.1 with lots of scheduled tasks and sometimes, for several reasons, some of them fails.

Do you know if it's possible to tell FME Server to run the process (for some workbench only) a second time if the first time it failed?

Exemple : I have a scheduled task supposed to run at midnight but for a reason it fails. FME could try to run automatically an hour later the same process with success, or not. If not, it could stop there, or try an hour later a third time or indefinitely, until it succeed.

So, is that possible or I'm dreaming?

In the scheduler definition you can set a notification topic in case the workspace fails. You could create a subscription that launches a second workspace whenever that topic is called, and then let that workspace check to see if a re-launch of the original workspace is in order.

You'll probably have to persist an incrementing retry counter somewhere, however, so that you don't restart the workspace indefinitely if there's a more serious problem.


Reply