Originally I created a single field called roles which has all the role names / ids / counts grouped together but ES complains that objects are not well supported and that’s why I turn them into 3 arrays instead.
That said, I don’t know how I might show the value of a role over time. What I need: the count changes overtime for each role_name. My work around right now is that I dynamically created a ton of fields:
and now I am able to track them overtime / use it in the Metric visualization. But as you can see this gets pretty bad as the number of fields I have just increased exponentially.
I recommend adjusting your data ingestion so it separates this information into individual fields and documents. If you can index each tuple as a separate document, with fields role_name, role_id, role_count and timestamp, this then becomes easy to visualize, and you can filter by role_name:Battle-Bay to narrow down to a single role.
these then get sent to logstash-* — I don’t actually know this well enough to create documents.
This is the library I am using btw: https://pypi.python.org/pypi/python-logstash — I need to use Python to send data via logstash to ES. Or are you suggesting that I communicate directly to ES? If so, what would you suggest that I do? Please point me to the necessary documentation if possible.
You know what, never mind… I was wondering why it previously complained and it won‘t let me model things the way I wanted to. But I just did a test run and this seems to be working right now:
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.