Addressed issues

The following issues have been addressed in the Mirantis Container Cloud release 2.3.0 and the Cluster releases 5.10.0 and 6.10.0:

  • [8869] Upgraded kind from version 0.3.0 to 0.9.0 and the kindest/node image version from 1.14.2 to 1.18.8 to enhance the Container Cloud performance and prevent compatibility issues.

  • [8220] Fixed the issue with failure to switch the default label from one BareMetalHostProfile to another.

  • [7255] Fixed the issue with slow creation of the OpenStack clients and pools by redesigning as well as increasing efficiency and speed of ceph-controller.

  • [8618] Fixed the issue with missing pools during a Ceph cluster deployment.

  • [8111] Fixed the issue with a Ceph cluster being available after deleting it using the Container Cloud web UI or deleting the KaaSCephCluster object from the Kubernetes namespace using CLI.

  • [8409, 3836] Refactored and stabilized the upgrade procedure to prevent locks during the upgrade operations.

  • [8925] Fixed improper handling of errors in lcm-controller that may lead to its panic.

  • [8361] Fixed the issue with admission-controller allowing addition of duplicated node labels per machine.

  • [8402] Fixed the issue with the AWS provider failing during node labeling with the Observed a panic: “invalid memory address or nil pointer dereference” error if privateIP is not set for a machine.

  • [7673] Moved logs collection of the bootstrap cluster to the /bootstrap subdirectory to prevent unintentional erasure of the management and regional cluster logs.