6.5.0 is here.
It may be labelled a minor version, but there's nothing minor about this release. 6.5 introduces many new features - from foundational stack-level features that apply to multiple use cases to more solution-specific features that aim to provide more turnkey experiences around specific use cases.
We cover a few feature highlights here. Be sure to dive into the individual product release posts for the full details. So without further ado, here goes.
Infrastructure & Logs UI
Many of our users having been using the Elastic Stack to ingest and explore operational logs and metrics from their IT infrastructure. We are excited to introduce new Kibana apps that make it even easier for users to gain visibility into their operational data. The new Infrastructure solution allows you to easily navigate between the 10,000-foot view of your infrastructure, and the logs and metrics activity on any specific host, pod, or container.
The complementary Logs UI is an advanced log viewer that provides a console-like streaming view of your logs, similar to running
tail -f in a console. Get details on both these apps in the Beats release post
Trace the details in the APM post
- Distributed tracing gives an end-to-end trace on a request as it traverses multiple services
- Agents for Java & Go, both highly awaited languages, are now generally available
- APM monitoring lets you track the health of your Elastic APM deployments from the Kibana Monitoring app
- New NDJSON-based input protocol brings improved memory footprint
The aggregate view is here, but get the full scoop in the detailed post.
- Cross-cluster replication - a new, self-contained mechanism to replicate indices from one cluster to another
- ODBC driver (alpha) - connect Elasticsearch to your favorite 3rd-party tool that supports ODBC
- Java 11 support - Elasticsearch adds support for the newest LTS Java release from Oracle
- G1 Garbage Collector (G1GC) is now supported by Elasticsearch
- Lookup realm - authenticate using one realm (for eg. Kerberos), and authorize using another (for eg. LDAP)
- Improved machine learning methods add the ability to (a) do independent scoring across partitions, and (b) detect anomalies that span multiple buckets
Zoom in on details in Kibana release post, but below is a sample.
- Canvas - create live, pixel-perfect, infographic style presentations of your Elasticsearch data. Canvas, which was available as a separate technology preview, ships with the default distribution of the Elastic Stack starting in version 6.5
- Spaces - organize your Kibana objects (for eg. visualizations and dashboards) into separate "spaces", and use RBAC to control which users have access to which space
- Rollups support in Kibana includes a new management UI to configure and manage your rollup jobs, and the ability to visualize your rolled up indices in Kibana
- Data visualizer for files - understand the structure of the data in your files, in preparation for further analysis
- More sample datasets, dashboards, and Canvas workpads simplifies the getting started experience for all users
Grok the details in the detail post
- SNMP input plugin, which was in beta since 6.2. is now GA
- App Search output plugin enables Logstash to act as a centralized ETL tool for feeding data to Elastic App Search deployments
Get the full beat in the detail post
- Beats Central Management lets you enroll, configure, and manage your Beats deployments from a central place using either the UI or API
- Functionbeat is a new Beat (and a new kind of Beat) that deploys as a function in serverless platform, and lets you stream cloud infrastructure data to Elasticsearch
The reduced form is below, but you can map the full picture in the detail post
- Field type promotion on read where ES-Hadoop will attempt to promote fields to the most common shared type when it encounters a mapping conflict
- Plus a few changes that start the road to 7.0 readiness.