Hi!
I read about the new Rollup API coming with ES 6.3.1 here: https://www.elastic.co/guide/en/elasticsearch/reference/6.3/rollup-apis.html
And I tried to understand it. But currently I'm not able to managed and understand it based on the APIs doc only.
Can you provide some examples and explain them?
Maybe an example that shows how to aggregate my data from a day based index containing all the raw data to a daily, weekly, monthly aggregated index.
Some questions the came up after my first tries:
How many jobs for an example like the only above? how often called?
What do the interval and delay params do? What's the page_size for? What to specify under groups and metrics? What will happen with my aggregated data?
Or even some more basic questions: Do I have to specify the rollup index in addition to my regular one or does Elasticsearch know that when searching with my regular index-pattern that it has to search in my rollup index too?
BTW: how to address unique count in aggregated data?
Thank you @Lukas!
This is the documentation on why someone should use this feature and from a very high point of view but not that much on how exactly do it.
You can read there "Details about setting up and configuring Rollup are covered in Create Job API". But to me, I still found no answer to my above questions.
E.g. you can read at https://www.elastic.co/guide/en/elasticsearch/reference/6.3/_rollup_uses_standard_query_dsl.html that I have to redirect my dashboards. But how do I do that?
Hi @TimV! That sounds promising
I will give it a try and give feedback.
After a very quick read thru I have not seen if I have to make or update my existing visualizations to reflect rollup data.
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.