Re: How to report on a full queue in Apache NiFi (Version 1.27.0)?

[Jens M. Kofoed]

I would love to share. But I’m currently not at work and don’t have access to the system at the moment. So after the easter holidays I will look at it.  Kind regards  Jens Den 11. apr. 2025 kl. 15.45 skrev stephane kouajip <stephanekoua...@gmail.com>:  Hi Jens, How did you configured the site2sitereporting service ?  I tried that approach but my site2site never worked, it was failing saying  <A4C8F9E1-2108-481F-9354-94811E545247.jpeg> Could you please share how you configure your nifi proprieties?  Regards, Stephane On Fri, Apr 11, 2025 at 3:23 AM Jens M. Kofoed <jmkofoed....@gmail.com> wrote: Dear Stephane I’m using a site2sitereporting service. Created a flow where I filter on connections, and calculate how many procent each connection is full. Both for flow files and for bytes (amount in queue vs. Max back pressure allowed). Later I filter connections with queues reach a threshold and send info to another system, which give me some alarms. I also create some parent group names, so I know in which group there might be a problem.  Kind regards  Jens Den 9. apr. 2025 kl. 20.35 skrev stephane kouajip <stephanekoua...@gmail.com>:  I will take a look. Thanks  On Wed, Apr 9, 2025 at 1:25 PM Daniel Chaffelson <chaffel...@gmail.com> wrote: Hi Stephane, I have code in NiPyAPI[1] for purging all connections in a process group, which was created to easily reset test flows. I expect you could use it as an example to instead inspect connection status to determine capacity issues using some scheduled fetch and analyse process. I am not aware of a native method inside NiFi for your requirement, but other wiser minds may offer further insight. [1] https://nipyapi.readthedocs.io/en/latest/nipyapi-docs/nipyapi.html#nipyapi.canvas.list_all_connections  On Wed, Apr 9, 2025 at 1:29 PM stephane kouajip <stephanekoua...@gmail.com> wrote: Hi NiFi Community, I'm working with Apache NiFi version 1.27.0, and I need help figuring out how to report or monitor when a queue is full. I want to be able to track and get alerts if any queue in my NiFi instance reaches its full capacity. I’ve looked through the NiFi documentation and community discussions, but I haven't found a clear approach to monitor the queues effectively. Can anyone provide advice or best practices for reporting on a full queue? Specifically, I'm looking for a way to: Monitor when a queue reaches its maximum capacity. Set up alerts or notifications based on queue status. Log or report this event for monitoring or troubleshooting purposes. Any help or pointers would be greatly appreciated. Thanks in advance!