I have a brand new OS large installation v3.63 and just getting my feet wet on node-red. SignalK is running flawlessly, but when I go to node-red, it just keeps losing connection to the server. Eventually, it will load the page and let me do some “programming”. It will reconnect by itself again, but loose its connection quickly after.
Watching the console, I can see the following error:
Firefox can’t establish a connection to the server at wss://venus.local:1881/comm
when red.min.js wants to establish a websocket connection, it is refused by the server with http error 502 - bad gateway.
Yes, almost the same. Again, SignalK connects fine through VRM.
Connecting to Node-red initially yields:
Error: remote installation does not seem to have a running Node Red service.
After a few refreshes it appeared to load and then fail:
I have never had an issue. I use Chrome on my PC and set up a 24/7 dashboard on an Android POS via LAN that is programmed to turn on when the camera senses motion with Hermit. I have never seen that message.
I have another system on my boat, and I really hope that one doesn’t have this issue. Will know tomorrow.
I have tried multiple browsers, but as you can see, it really is a Cerbo server issue.
As I use SignalK a lot now, I have decided to turn off node-red and use the signalK node-red plugin instead. This way, the signalK nodes work right away. Works well sofar. I haven’t tried this yet on my cerbo that has the connection issue, though.
Yesterday again permanently “lost connection” but at the same time there was quality problem (high packet lost) on the side off the internet service provider.
I don’t really see there should a link between those 2 as I was working on venus.local where there is 0 packet lost.
I have the same problem only when I connect to Node_Red via “Node_Red local”
However, and this is just a thought and correct me if I’m wrong, the service issue could be related to the “lost connection” because we are still going through VRM in order to access Node-Red
BTW, no one has answered my very same question as yet.
I’m facing the same issue. I connect directly to the IP via https://<IP>:1881 to manage my programs locally instead of using the remote connection. However, around 10 to 20 minutes after a successful start, the connection always drops. To continue working, I have to disable Node-RED and restart it.
Does anyone have the link to the local dashboard? I couldn’t find any reference to it.
I’ve also been trying to monitor the Node-RED server by connecting via SSH to Venus OS. Has anyone found a good solution for managing it this way?
I didn’t find any issues in the log files, but it’s possible I was checking them before the system actually crashed — which might explain why nothing seemed out of place at the time.
Monitoring system resources shows that my scripts are causing 60–70% CPU usage, mostly because values are pulled from Victron nodes which update every second. I’ll try reducing the number of values to avoid exhausting the system.
Any tips or suggestions would be greatly appreciated!
