Chart plotting live data hangs due to browser high cpu/gpu usage

1159 points at this time...
graph_array

How is the node configured?
How old are the oldest timestamps in the data?
Are the message counts stable or increasing?
What version of the dashboard is in use?

The configuration is shown in posting 18.

8 hours

They vary depending on system loading. Total point count was 1419 when I checked a few minutes ago - that's up 260 points compared to this morning.

dashboard is v2.28.1 node-red is v1.29
Everything is working fine. CPU loading from Chrome peaks at 20% on 5 second intervals. I expect this to flatten out after I change sequentially trigger charts and gauges from global data rather than all at once on receipt of MQTT data.

I just noticed that Series 2 data is showing 1081 points and that data seems to be properly charted, yet debug is allowing access to the first 1000 points. FWIW...
graph_array2

So that is consistent with my understanding. It is limiting to 8 hours or 2000 points on any one line. None of the lines have reached 2000 so they are going up to 8 hours. If you were to set the limit to 1000 then I would expect some of the lines to be truncated.

Maybe not really 100% on topic, but not 100% off topic either: I feed my gathered data to an InfluxDB and query it from the DB when updating the chart. Depending on the time domain I want to see in the chart I use some aggregation or timed moving average. This reduces the number of data points while querying the DB. This way not too many points are displayed in the chart.

The debug node does truncate data in order not limit the amount of data sent to the browser. You can rely on the array length shown being correct.

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.