Lot of RAM used

[GiovanniG]

Hi, I'm using the same flow I was using with version 1.06, that time the process used stable about 0,18/0,20 GB of my Raspbian3 RAM even after months.
But after upgrade the system to:
6.14.8
v10.23.0
Node-RED v1.2.2
I'm having (from 5/11/20):

Probably it's caused by an updated module which is now leaking.. but how can I understand which one? There maybe logs increasing inside Node Red process? How can I debug modules?
I still not restarting the process to let us understand more, but soon I will because it's going to rise. Thanks

[krambriw]

How do you expect anyone able to help you? You have the flow, you know what is inside, we know nothing about your configuration. Provide more details what is inside your flow

[Colin]

Start node-red in a terminal and see if there are any unexpected messages there. Post it here (from the start) if you are unsure.
Have you checked in Palette Manager to see whether there are updates available for any of the nodes you use?

[BartButenaers]

@GiovanniG

but how can I understand which one?

Hi Giovanni,
I don't think there is an easy way to determine - based on some kind of analyser - which node is consuming most of the memory. I tried some time ago to build something like that, but it didn't work. I have listed the limitations on my readme page, based on the great feedback I got from the community and Nick.
So therefore I think you indeed need to start reducing your flow, to determine - based on elimination - which node(s) are causing your troubles ...
And of course try first the above tips from Colin and Walter.
Bart

[BartButenaers]

You can also create a heap dump (e.g. via my node-red-contrib-heap-dump node), and then start comparing two heap dumps to see where the memory is being consumed. But it is long ago that I have used that, so don't remember anymore whether that was very helpful to determine which node is the root cause...