I have a problem with Atlas Taxonomy. Any operation on taxonomies are taking a long time to run (both using WebUI and API). I am using HBase + Solr Cloud. Simple GET Atlas Taxonomy takes 50 seconds. There are only some INFOs (no ERROR or WARN) in application.log, solr.log and kafka logs. Operations on tags are working good.
Ranger plugin is disabled for Atlas and Kafka. Cluster is secured (kerberos). Atlas is configured to authenticate via AD, but I am working as atlas user (generated kerb ticket for atlas user for API). HBase table 'atlas_titan' has 380k records.
I am using HDP 2.6.3. with Ambari 2.6.0. I restarted HBase, ZooKeeper, Sqoop, Oozie, Ambari Infra and Kafka as I found it as a solution on this community.
Atlas Taxonomy have known to have performance problems. For now, this has been disabled by default on branch-0.8 (see commit). While there is a plan to have this feature within re-introduced, there is no firm ETA on it.