node-red-contrib-kafka-manager icon indicating copy to clipboard operation
node-red-contrib-kafka-manager copied to clipboard

network connection increase periodly

Open hnuzhoulin opened this issue 1 year ago • 1 comments

First, thanks this nodered kafka node.I can transfer data from mqtt to kafka or kafka to mqtt easily. I face a problem when using nodered and this kafka node. the nodered dashboard 1880 periodic can not access,if I restart nodered process,it works again.then I debug it.

I noticed that the number of TCP connections had risen from 400 to over 4,000 in the last four hours. All these new TCP connections were to Kafka. Here, Node-RED acts as a Kafka producer, so I couldn't find any useful information on the Kafka side. Looking back at the historical monitoring, I noticed that each time there was a nodered failure, there was this phenomenon of connection counts skyrocketing to over 4,000.

I verified if the data flow in the processes behind Node-RED was interrupted when face this problem. Yes,data forward works well. although the CPU and load on the server hosting Node-RED were high and increase , maybe the load let nodered dashboard not work.

I discuss this in nodered forum,they suggest here for help.

What can I do to find the problem. BTW,it works well six month until last week.yes,we add more device to mqtt.Thanks.

the attached file is my nodered flow which exported. flows202312131652.json

hnuzhoulin avatar Dec 14 '23 02:12 hnuzhoulin

I think I have noticed and fixed the issue with my big rebuild. Basically a bug in check if path available and failing to cleanup test connection properly. It is about finding time to test all. Underlying logic has had a massive restructure, hopefully with greater resilience to outlier states.

peterprib avatar Dec 14 '23 06:12 peterprib