Build a visualization based on gerenralization of data

[Carlos_Zottmann]

Hi, Everyone!

Sorry for the newbie question!

I have data of an inventory of my virtual servers. So far it has been a piece of cake building a visualization of my servers grouped by Operating Systems, so I´ve got a nice graph showing how many RHEL 5, RHEL 6, RHEL 7, RHEL 8, Windows 2008, Windows 2012, Windows 2012, Windows 2016, an so on.

But I would like to build a visualization where I aggregate my servers under the general categories "Windows", "Red Hat Linux", "Other Linux", and so on ...

How could I do this?

Thanks in advance,
Carlos.

[bhavyarm]

Hello Carlos,

All questions are awesome. Thanks for posting.

I think there are two possible ways of getting this done:

1. At the data ingestion - add another field which will be a sorta super field and make sure its values are what you would like and reindex the data
2. Make a scripted field under index pattern management in Kibana. https://www.elastic.co/blog/using-painless-kibana-scripted-fields
   And get it to show up as a composite of the values you would like to see.

Thanks,
Bhavya

[Carlos_Zottmann]

Hi! Thanks for your prompt reply!

I´ve read the link you´ve mentioned and it seems to me that it can be done by using the method in " Match a string using regex, and take action on a match".

But it also seems that it would be better to use your first option, as it would be available at the ElasticSearch Index itself, and not only to Kibana. Could you point me to instructions on how to add another field at the data ingestion? Also, I´ve loaded the data using the Machine Learning option in Kibana to load them from a csv file. Can I create this extra field while using this option?

Thanks again?
Carlos.