Enhancements and bug fixesedit

The following changes are included in this release.

Featuresedit

Autoscaling for trial deployments. You can now turn on autoscaling on trial deployments. The autoscaling will be limited to a few levels while the deployment is in trial. Once you subscribe to Elastic Cloud, all autoscaling options will then become available.

Enhancementsedit

Add uid to direct trusts. Direct trusts now have a uid field. This field is generated automatically upon creation and allows unique resource addressing without relying on index as well as distinguishing between updates and inserts.

Disable and enable applications from the deployment Edit page only. You can no longer terminate and then restart a terminated non-Elasticsearch application (Enterprise Search, Integrations Server, Kibana, APM & Fleet) from the application page when administering a deployment. Instead, you can go to the Edit page of the deployment to disable or re-enable these same applications.

No sso_url generated for Heroku deployments. Kibana to Elasticsearch Service SSO links are no longer returned from the Elasticsearch Service API and UI for deployments hosted in Heroku (https://elements.heroku.com/addons/foundelasticsearch), as they are not supported in this environment.

Update Kibana reporting settings. Some reporting settings have been renamed in Kibana. You may refer to these settings by their new name. The change is backwards compatible so that either path may be used.

xpack.reporting.capture.timeoutsxpack.screenshotting.capture.timeouts

Allow Kibana elasticsearch.compression setting. For version 8.3 and higher of the stack, it is now possible to enable compression for communications between Kibana and Elasticsearch by using the elasticsearch.compression setting of the Kibana configuration.

Update deployment hardware profile from the UI. You can now migrate a deployment to a different hardware profile from the Edit page of a deployment.

Bug fixesedit

Display upgrade link for deployments with older versions. Fixes an issue where deployments with End of Life stack versions did not show any upgrade link on the deployment management page. With this link, you can now upgrade more easily to more recent versions.

Fetch frozen tier node stats by instance name instead of service ID. Display the searchable object storage on frozen tier nodes more reliably in the deployment overview.

Don’t show active deployments as deleted in trust management settings. Fixes an issue where trusting more than one specific deployment would show those deployments as (deleted) in the deployment trust UI.

Add trafficFilter bypass token to all proxied requests. Fixes an issue where deployments with traffic filters configured were unable to label their master node on the UI.

Remove legacy attributes when updating Heroku deployment. Fixes an issue where legacy attributes could cause some Heroku deployment configuration changes to be rejected.

Added forbid_closed_indices parameter to the Elasticsearch Index stats API. Fixes the index_closed_exception issue generated during a vacate operation when a closed index had an alias pointing to it.

Summary sidebar squashed when scrolling. Fixes an issue where the summary sidebar could get squashed when scrolling down the Edit page of a deployment.

Docsedit

Add troubleshooting content on unavailable shards. New troubleshooting content on unavailable shards. Adds a list of [common symptoms and troubleshooting steps](https://www.elastic.co/guide/en/cloud/current/ec-scenario_why_are_shards_unavailable.html) on how to resolve issues when shards become unhealthy or unavailable.