We need to use kibana (with Logstash and Elasticsearch) through an additional server for security within our intranet applications.
Everything works fine, but our 'Guardian' Server seems to have problem with some special characters in the intermediate url, generated by Kibana for communicating with elasticsearch. For Example : http://IP/elasticsearch/.kibana/_mapping/*/field/source?=1485776369021
Is there a way to modify these URLs (to avoid or encode the special characters) generated while Kibana is loaded?
What I tried so far : As we cannot make the guardian server allow some specific characters, we are trying to find some solution at the kibana side. I tried to use alternate URLs :
e.g., replace * with specific names that are actually used - index-pattern,search,config so the url becomes /elasticsearch/.kibana/_mapping/index-pattern,search,config/field/_source?
through apache mod proxy but there seems to be two issues -
While other reverse proxy configs successfully work on kibana, this one with * character doesn't seems to have any effect
The alternate URL itself contains , character (which might again cause issue, though we're still in process of confirming the restricted characters on our guardian server)
I also tried to use encoded URL and nocanon flag with reverse proxy, but no success.
Is there any configuration or alternate way to avoid internal URLs from using these special characters (at kibana or elasticsearch end) ?
Kibana Version : 4.5.1
Elasticsearch Version : 2.4.0
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.