You could also have 2 Flows under 1 group. First flow gets the token and caches it .. second retrieves cached token to use where needed. ... View more

You need to implement some kind of check to see which db is running.  A small select would do on db1 from nifi - on success set an attribute (pre-added to the file) to db1 on failure update to db2 then the attribute will always have the value of the running db. Both would lead to UpdateAttribute processors then they can lead to the single update using the attribute for the db to connect to. FYI .. can be a good idea to rename the queues from success and failure to show DB1 and DB2 or whatever makes sense to you. Hope this helps 🙂 ... View more

Fetch needs a filename to be able to fetch the file.  If known you can set it as an attribute on the file or a parameter context to pass in as the filename. If you don't know what it will be then you'll need to use a list processor. ... View more