This section outlines new features implemented in the Cluster release 8.6.0 that is introduced in the Container Cloud release 2.16.1.
MKE version update from 3.4.6 to 3.4.7¶
Updated the Mirantis Kubernetes Engine (MKE) major version from 3.4.6 to 3.4.7 for the Container Cloud management, regional, and managed clusters. Also, added support for attachment of existing MKE 3.4.7 clusters.
Improvements to StackLight alerting¶
KubePodsRegularLongTermRestarts alert that raises in case of a
long-term periodic restart of containers.
Elasticsearch retention time per index¶
Implemented the capability to configure the Elasticsearch retention time per
index using the
elasticsearch.retentionTime parameter in the StackLight
Helm chart. Now, you can configure different retention periods for different
indices: logs, events, and notifications.
elasticsearch.logstashRetentionTime parameter is now deprecated.
Prometheus Blackbox Exporter configuration¶
Implemented the capability to configure Prometheus Blackbox Exporter, including
timeoutOffset, through the StackLight Helm chart.
Custom Prometheus scrape configurations¶
Implemented the capability to define custom Prometheus scrape configurations.
Elasticsearch switch to OpenSearch¶
Due to licensing changes for Elasticsearch, Mirantis Container Cloud has switched from using Elasticsearch to OpenSearch and Kibana has switched to OpenSearch Dashboards. OpenSearch is a fork of Elasticsearch under the open-source Apache License with development led by Amazon Web Services.
For new deployments with the logging stack enabled, OpenSearch is now deployed by default. For existing deployments, migration to OpenSearch is performed automatically during clusters update. However, the entire Elasticsearch cluster may go down for up to 15 minutes.