Hi, Are you running nifi docker images with kubernetes, I assume the image has zookeeper, as that is required for teh cluster to work. nifi logs are in the nifi-app files, please check that as well. ... View more

Someone has an InfluxDB Writer https://github.com/fsauer65/NiFi-Extensions/tree/master/nifi-influxdb-bundle/nifi-influxdb-processors/src/main/java/org/apache/nifi/influxdb ... View more

If you restart one of the nodes, what errors/warnings are show in nifi-app.log? ... View more

https://github.com/fsauer65/NiFi-Extensions/tree/master/nifi-influxdb-bundle/nifi-influxdb-processors/src/main/java/org/apache/nifi/influxdb This processor should be able to write to Inlfux. See this for docker ports: https://forums.docker.com/t/accessing-udp-for-check-dhcp-from-a-container/8851 Docs on Listen: https://nifi.apache.org/docs/nifi-docs/components/org.apache.nifi/nifi-standard-nar/1.4.0/org.apache.nifi.processors.standard.ListenUDP/index.html Collecting Logs Article: https://bryanbende.com/development/2015/05/17/collecting-logs-with-apache-nifi ... View more

@cloud car Step one will not work. Not only do all components need to be in a stopped state (with no active threads), but NiFi will not let you delete a connection if it has data queued in it. When you stop a processor any running threads will still run to completion. "stopping" a processor simply stops the processor from triggering again. You can issue a rest-api call against the Process group to stop all components inside that process group. You can issue a rest-api call against every connection to "empty queue" to make sure they are empty before trying to delete. Thanks, Matt ... View more