Hi, I believe it is a bug,
I've a recursive periodic injection every 0,2 seconds and I needed to debug a value, so I generated a msg retur and connected to the funzion the msg debug. After few seconds NR gets' frozen, maybe to don't discard debug objects it pretends to catch all and go to frozen. 0,2 seconds is not much, my PC isn't the latest but I suppose it can manage to receive a message esvery 0,2, seconds..
Only with patience I was able to delete the flow between function and node and deploy again. I needed also to reload the browser (firefox) to gain the normal fast conditions.
I think there is a slow java script here that should be optimizated to avoid this. There is also needed to introduce a switch somewhere to stop the flow working if something like ths get wrong, to keep again the control of web interface. If I stop node red I can't have more the web interface
There are a lot of reasons that debug output has the performance it does. It is discussed in other threads. You can try outputing to the log rather than the admin ui or, perhaps better if all you want is to monitor the current value, put that out to the status display instead.
If you think you can further optimise the javascript code I’m sure the project would accept a pull request.
We introduced the --safe command-line parameter in 0.20 for exactly this.
I'll try to update to .20 tomorrow, thanks
May you kindly write exaclty the line command with --safe?
Thank you
The command is
node-red --safe
however, if you have node-red to automatically run on boot you will have to stop it first.
1 Like
