I have data consisting of the following three fields: jobId, subJobId, and outcome, where each jobId has several subJobs, each with one outcome.
I want to show a graph indicating how many jobs had at least one failure outcome, and how many had only successes.
If I do a split on outcome, I get the proper information for how many subJobs passed/failed. If I add a unique count on jobId, I get closer, but a job that had both a failure and success shows up twice. A job that only had failures or only successes shows up once.
Is there an easy way to do this in Kibana?
Sample Data
jobId=1, subJobId=a, outcome=pass
jobId=1, subJobId=b, outcome=pass
jobId=2, subJobId=c, outcome=pass
jobId=2, subJobId=d, outcome=fail
jobId=3, subJobId=e, outcome=fail
jobId=3, subJobId=f, outcome=fail
Desired graph: (doesn't matter exactly how, let's say a vertical bar chart with a split series. Whatever)
number of exclusively successes: 1
number of one or more failures: 2
If you store every record as an individual document then finding the number of one of more failures will be easy but finding exclusive successes might not be possible.
Thank you! I can't change the reporting format, but these queries look perfect!
If you don't mind, one more question. Can I put these explicit ES queries into a Kibana graph, or are they only suitable for running directly against ES? I.e., if I wanted to do a vertical bar chart, split on the X Axis as a date histogram and with two bars, one for each query, can that be done?
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.