I have a field that contains a string representation of a state. Some states are normal which means that there are many documents with these states. Other states represents errors. Usually, the number of documents containing these states are much lower than the normal states. I would like to view a count of the states as a time series. Since the count of different states are so different, I need to have a different scale for each state. Is that possible in kibana? I have tried to create a line graph with date histogram and then split chart based on terms. But the scale on Y axis is the same for all charts which means that some charts just have a line close to 0 even if they are not 0.
I think you can get your chart with Lens. You need to create one layer with the vertical axis on the left for one state, using the Filters option in the Break down by section to only render that data. Then you add another layer and filter by the other data and set your vertical axis on the right.
Check this screenshot. These are flight positions. On the left you have all positions that are on the ground based on the filter onGround: true and rendered as bars. Then on the right, with the same horizontal axis and vertical axis parameters, but changing the Filters to be onGround: false and rendered as a line. See the right axis shows a much larger value.
Thanks! but I don't think it is exactly what I want. I have about 25 states that have values in different scales. Some have a count of thousands and other maybe have a couple of hundreds or in the single digits. I would like to have one chart per state but without having to create 25 individual charts. The split chart functionality almost did what I wanted, but all charts shared the same scale on the y-axis...
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.