Kibana X-Pack 6.2.0 Release Notes

edit

Kibana X-Pack 6.2.0 Release Notes

edit

New Features

edit
APM
  • The Elastic APM is generally available as part of the 6.2.0 release of the Elastic Stack. For more information about the APM functionality in Kibana, see APM.
Machine Learning
Monitoring
  • Added monitoring information for Beats in Kibana. See Beats Metrics.

Enhancements

edit
Machine Learning
  • Enhanced the creation of custom URLs for existing jobs. When you edit jobs, there is a new Custom URLs tab, which provides options for creating a link to a Kibana dashboard or a different URL.
Management
  • Renamed the Save button Save & Deploy. This change makes it clearer that when you click the button the pipeline is saved to an Elasticsearch index and deployed to any Logstash instances that are configured to run it.
  • Added the ability to clone pipelines in Kibana. When you click the Clone button, data from the original pipeline is used to pre-populate the new pipeline form.
  • Enabled auto-sizing of the pipeline configuration text area. The text area automatically grows or shrinks as you type out a pipeline definition.
Monitoring
  • Added resolved alerts to the list of alerts that are visible during a selected time frame in Kibana.
  • Added the interval size for aggregated metrics in the tooltip for each chart.
  • Improved error handling such that generic 500 errors provide better messages.
  • Redesigned the Logstash Pipelines view to provide more actionable information in a tabular format.
  • Added a basic detail drawer to the Logstash Pipeline Viewer. When you click on any vertex in the Pipeline Viewer, the drawer slides open. It shows basic information about the vertex that was clicked and any metrics, if available.
Reporting
  • By default, weekly .reporting indices created after upgrading to 6.2 use a single primary shard and have the auto_expand_replicas: 0-1 setting rather than a fixed setting of 1 replica. As a result, these indices can have a green status on single node clusters. There is no impact in multi-node clusters.

Bug fixes

edit
Machine Learning
  • The job selection control now remembers the state of the Also apply time range checkbox, such that the previous selection is restored when the control is re-opened.
  • Fixed issues where lines were drawn over the Y-axis in the event rate charts for multimetric and population jobs.
  • Removed scripted fields from job wizards.
  • Added type-ahead filtering to the job group and influencer selection controls in the job wizards.
  • Updated the data preview to use the index pattern or the list of indices that were selected.
  • Disabled the start button for jobs that do not have datafeeds.
  • Updated the actual and typical values for anomalies in Kibana, such that they are formatted according to their magnitude.
  • Fixed the handling of anomalies when field values contain backslashes.
  • Ensured Anomaly Explorer swimlane rows are sorted by score.
  • Disabled the Forecast button for jobs that were created before 6.1.0 and for jobs that contain an over_field_name property.
  • Added support for field names with non-alphanumeric characters in the Data Visualizer.
  • Fixed problems creating jobs from saved searches that have exists filters.
Monitoring
  • Added the ability to navigate between the pipeline versions in the Logstash Pipeline Viewer.
  • Updated the Logstash Pipeline Viewer such that you can change the refresh interval but you cannot select the time range.
  • Added Logstash version information back in the node list.
  • Improved logging when the Kibana monitoring collectors stop or resume and downgraded the log level for these log messages to info.
Reporting
  • Resolved issue with horizontal metric visualizations being displayed incorrectly in PDF reports.
  • Fixed bug where the title of the Dashboard panels was hidden when creating PDFs with the preserve layout option.
  • Fixed issue with certain characters in the job parameters which caused CSV reports to fail.
  • Resolved issue with preserve layout PDF reports intermittently having blank visualizations.