scality/metalk8s 2.6.0

MetalK8s 2.6.0

on GitHub

2.6.0 embeds Kubernetes 1.17.14 which will be EOL in a few months or weeks. It is recommended to upgrade to 2.7.0, using 2.6.0 as an intermediate upgrade step since it is not possible to upgrade directly from 2.5.0 or 2.5.1 to 2.7.0

Documentation

Hosted on ReadTheDocs
Sources available here

Upgrade Notes

Please follow the upgrade instructions here

Downgrade Notes

Note that downgrading from 2.6.0 to 2.5.1 or 2.5.0 is not supported because of some breaking changes in the etcd cluster. Downgrading would require manual backup/restore of the etcd cluster.

Changelog since 2.5.1

Full list of closed issues is available here

What's new

MetalK8s 2.6.0 is now based on Kubernetes 1.17.14. It includes:

• a new log centralization service, aggregating Pod and journal logs into Loki. Logs can be browsed from Grafana, through new dedicated dashboards or using the Explore tab.
• the ability to configure Alertmanager and existing Prometheus rules. All details are available in Service Configurations Customization documentation
• a refined UI bringing more details about Nodes and Volumes, as well as contextual links to Metrics or Logs Grafana dashboards
• some changes to the Solution framework:
  • provide your Operator (Cluster)Role(s) in a single role.yaml file
  • use the scality/metalk8s/go/solution-operator-lib Go module to read the configuration file for your Operator
  • Admin UIs are now expected to be deployed by your Operator