I am using Kibana to visualise a POC on monitoring total transactions occurred and compares it with moving average of window 10, to raise alerts.
My alerting code is in Python which queries elastic search (using index name) and redirects users to Kibana to visualise the change in trend. I see that Moving average values calculated in Kibana differs from the moving average values I from elastic search queries. As a consequence, our alerting system alerts for an anomaly but visualisation isn't plotting it as errors.
Can someone please explain how I can sync both moving average values !?
The other thought I just had come to mind is what is the interval you're using in Kibana vs. Elasticsearch? Window will be the number of buckets the interval creates. So a window of 10 on a 1 hour chart with 10 second interval will give you a different result then a window of 10 on a 24 hour chart with a 1 hour interval.
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.