@tbragin We spoke today at the meet up. For the "other" pie-slice issue, I was asking if you could use the "sum_other_doc_count" field in the response json. Here's a small snip:
Kibana 3.
The most cause is the incompatibility between K3 and K4, I cannot recreate all my dashboard.
I understand the 'developer' motivations, but from the point of view as product it was the worst path.
The other are
@tinle I would love to see your fixes pulled into the fork of Kibana 3!
Here's the github repo that we are tracking: https://github.com/kibana-community/kibana3
I pulled in the latest code from the official branch and some community submitted improvements like csv exports.
We can share ownership of that repo with you to make it easier for you to commit your changes.
@Sesha_Narahari indeed, it appears that as a result of recent enhancements, Elasticsearch now has ability to calculate both "other" and "missing" values. We updated the ticket accordingly and tentatively targeted to one of the 4.x releases:
It is also annoying that you need a server to run kibana 4. Kibana 3 was very easy to install...
"Need a server"âcould you elaborate? K4 requires you to run a lightweight server-side process with hardly any dependencies while K3 requires a web server that serves static files. Both methods require you to deploy a tarball on a server so the only difference seems to be that K4 requires an init script since we can't piggy-back on the web server's. Is that the problem or am I missing something?
K4 requires you to run a lightweight server-side process
We have deployed the kibana 3 static files on our server and allow admin users to access Elasticsearch. With kibana 4 we now have to manage another process on the server and reverse proxy it in some way. Or am I missing something here?
No, you're right except that a reverse proxy configuration is optional. I just don't think that spinning up a service is significantly more difficult than configuring a web server and have it serve a specific set of static files.
For such a comparison to be fair you have to start from zero in both cases. Obviously, if you already have automatic deployment of static files to a web site in place then K3 will be easier to deploy. If your existing infrastructure offers other features things play out differently. At my previous job I had Ansible configurations to install and maintain both K3 and K4 and there was no difference in complexity.
We are using kibana3 and 4, both in an production environment.
The reason for kibana3 is that it is easier to use for user without any knowledge on ES/Lucene.
Although with https://github.com/christian-marie/kibana3_auth we are able to implement field/type based access rules - something that isn´t avaiable for kibana4(maybe with shield, but AFAIK you can only do index based ACLs).
Still kibana4 is great for its interactive dashboards where you can correlate many of data in 1 place and we are glad to have it.
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.