After a few hours of normal operation I've noticed since v4.08 the nodes become unresponsive.
They do not send messages through.
I don't see anything wrong.
I can modify and deploy the software but when I send a command it does not go through which means commands don't get sent through.
Restarting it does make it responsive again but then after a few hours same issue occurs.
Please see the screenshot attached.
I have injected a boolean true but it does not go through to debug.
Are you using charts in your dashboard? If so then work out the total number of points stored in your charts when a few hours has elapsed. If you find that is many thousands then that is probably the problem. For example, a chart with four lines with data every second, would be 14000 points per hour, so if you have specified a time span of 4 hours that would be approaching 60,000 data points which have to shuffled down in the browser every second.
If it is not that then run a system monitor on the machine and see if something is hogging up the processor.
Thank you for the reply.
I am using charts.
Points update every 15minutes.
Node-red was hogging up the CPU.
I disabled all charts the CPU eased off.
Hopefully that was causing it.
Will monitor over the next few days and see the results.
I found I had created a loop which went unnoticed, this was causing node-red to use high CPU upwards of 90%.
This would also make the client machine browser to use high CPU.
It has now stabilized and node-red is using round 5% CPU.
I hope somebody else finds this useful.