New for Elastic Cloud Enterprise 2.3.0:
Role-Based Access Control for the ECE console is generally available (GA). Previously in beta, the RBAC functionality for the ECE console is now officially released and supported. In the UI, you can access your RBAC user profile from the Users page.
RBAC now also includes Active Directory support for platform authentication, both in the UI and the RESTful API, alongside SAML and LDAP.
To help you configure role-based access control, we have also updated our documentation again. Check Configure role-based access control.
- Support for API examples in the UI. When creating a deployment template, you can now select API Example to get the equivalent RESTful API call. These API examples can help you in your automation journey or if you want to integrate ECE with existing tooling.
- Improved Activity page for deployments. When looking at the Activity page for a deployment, you now have more filtering and details available. Find the info you want, more easily.
- Support for Elasticsearch heap dumps. In the Operations page for a deployment, you can now create an Elasticsearch heap dump for troubleshooting and further analysis.
- Elasticsearch and Kibana 7.2.0 and 6.8.1. This version of Elastic Cloud Enterprise includes more recent Elastic Stack versions. Previous versions of ECE can download these versions and add them as Elastic Stack packs. To learn more about best practices when upgrading, check Upgrade versions.
- Removed Ubuntu 14.04 LTS (Trusty Tahr) documentation. As announced in our 2.2.0 release notes, the official end-of-life (EOL) from Canonical for Ubuntu 14.04 LTS (Trusty Tahr) was April 2019. We have removed the related Ubuntu Trusty documentation for ECE and strongly recommend that you upgrade to a fully supported version, such as Ubuntu 16.04 LTS (Xenial Xerus). You can either perform host maintenance to upgrade your hosts or prepare new hosts and reinstall ECE on them.
APM and default ILM policies. For 7.3.0 hot-warm deployments, APM now creates four initial index lifecycle policies. The hot policy is configured by default, but you’ll need to select the
data:warmnode attribute to make sure ECE knows what to do with your warm data. Check Configure index management.
- Disk usage estimate replaced with actual disk usage. The disk usage for Elasticsearch nodes are now based on the actual disk usage rather an estimate based on shard sizes. This ensures that all files are accounted for - including closed indices, broken shards, and transaction log files. Installations running without working XFS quotas will report the disk usage of the underlying host as that is then the real limit. A similar scenario can also happen if the administrator overcommits on disk usage by creating larger quotas than what is available on the host. In sum, this change makes the disk usage more relevant as a node and cluster health metric.
Release date: July 25, 2019