Off the top of my head, based on what you are describing, I would consider just using a funnel.
Your flow would start somehow, then maybe slit based on some logic, and if you wanted everything to come back together, just create a funnel that leads back to your final processor / set of processors. Craig Craig S. Connell CTO & Senior VP of Engineering csconn...@staq.com 443-789-4842 On Wed, Aug 19, 2020 at 10:20 AM bsavard <bob.sav...@aifoundry.com> wrote: > Hi, > > I have a flow that does a bunch of stuff and along the way, there's > potential for error. When an error happens, no further processing of the > flow should execute, except that a certain processor or process group > /must/ > execute as a clean-up step. Pretty much like how a "finally" works with > Java's try/catch. > > So I'm wondering if there's a 'best practice' for how to /always/ route to > a > certain processor or process group, especially when you have multiple > nested > layers of process groups. > > TIA > > > > > -- > Sent from: http://apache-nifi-users-list.2361937.n4.nabble.com/ >
