We have the similar requirements. We had small POC which worked very well. When we started assessing NFR, we stuck with this bottle neck.   Issue is: 1. We have a clustered environment with three nodes. 2. To check what we did, we had set of processors. And ran a flow. Then we stopped one of them from primary node. All information got reflected correctly on all nodes. 3. We scaled down primary node from where we ran the flow. 4. Earlier we were able to see replicated stuck/queued message on all non-primary nodes.  As soon as, primary node was down, other nodes does not show that queued message. 5. When we started back earlier primary node, we can see everything good.   Is there any plan for support for this HA scenario for apache ni-fi? https://cwiki.apache.org/confluence/display/NIFI/High+Availability+Processing   Please suggest. ... View more

Hi @Matt Clarke, Does the table you shared above hold true today as well? Apache Nifi Crash Course video on https://www.youtube.com/watch?v=fblkgr1PJ0o mentions the same that a cluster should preferably have a single digit number but if really needed you can rather have 2 separate clusters with 10 nodes each and establish a sync between them. All I am trying to understand is with the latest version it still holds true and 10 nodes are still good to hold hundreds of thousands of events per second? Thanks in advance! ... View more