This section outlines new features and enhancements introduced in the Cluster release 5.19.0.
MCR and MKE versions update¶
Updated the Mirantis Container Runtime (MCR) version from 20.10.5 to 20.10.6 and Mirantis Kubernetes Engine version from 3.3.6 to 3.3.12 for the Container Cloud management, regional, and managed clusters. Also, added support for attachment of existing MKE clusters 3.3.7-3.3.12 and 3.4.1-3.4.5.
For the MCR release highlights and components versions, see MCR documentation: MCR release notes and MKE documentation: MKE release notes.
Ceph maintenance improvement¶
Integrated the Ceph maintenance to the common upgrade procedure. Now, the
maintenance flag function is set up programmatically and the flag itself is
Ceph RADOS Gateway tolerations¶
Implemented the capability to specify RADOS Gateway tolerations through the
KaaSCephCluster spec using the native Rook way for setting resource
requirements for Ceph daemons.
Short names for Kubernetes nodes in Grafana dashboards¶
Enhanced the Grafana dashboards to display user-friendly short names for
Kubernetes nodes, for example,
master-0, instead of long
This feature provides for consistency with Kubernetes nodes naming in the
Container Cloud web UI.
All Grafana dashboards that present node data now have an additional Node identifier drop-down menu. By default, it is set to machine to display short names for Kubernetes nodes. To display Kubernetes node name labels as previously, change this option to node.
Improvements to StackLight alerting¶
Implemented the following improvements to StackLight alerting:
Enhanced the alert inhibition rules.
Reworked a number of alerts to improve alerting efficiency and reduce alert flooding.
Removed the inefficient
matcherskey to the routes configuration. Deprecated the
Logs-based metrics in StackLight¶
Implemented the capability to create custom logs-based metrics that you can use to configure StackLight notifications.