I want to plot a specific line chart with Kibana Visualize. My data looks like this
Timestamp / Value / Serie
15000 / 5.0 / 1
15500 / 12.5 / 1
16000 / 23 / 1
16500 / 16.2 / 2
17000 / 14.9 / 2
18000 / 4.6 / 3
...
On X axis, it will be the timestamp and on the Y axis, only the max value of each serie (here 23, 16.2 and 4.6). How can I plot this chart with Kibana?
You should able to do it with a Line chart in Kibana.
For the metric, select Max for the Value field.
For the X-axis select: date histogram on the timestamp field.
Then click on Add Buckets and select Split Series, using the terms aggregation on the Serie field.
This is an example that I just created. It shows max bytes for web requests, over time, on of the extensions listed in the legend.
As you can see, there are several points for 1 serie (the serie 200,174,459 has 4 points for example). I want only the max value for the serie (only the point 126 for the serie 200,174,459). Is it possible ?
You should change the Interval in the Date Histogram to cover all the data in there. If you don't need the timestamp at all in the chart, I think just having a bar chart with each bar representing the max of each series would be more useful to you.
I want to have the timestamp in the chart. By changing the Interval in the Date Histogram, I have only the max value of the serie plotted. OK
But this max is not with the right timestamp: for example my maximum for the serie 200,174,459 is at 07h; by changing the Interval to 12h, this max will be plotted maybe at 0h => I want it to be plotted at 07h.
There is no way to change where it's plotted in the date histogram as the dot is put right in the middle of the bucket from the x-axis. But I think you should still be able to see the timestamp for max if you mouseover the point.
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.