20230731 DCE 5.0 Enterprise Released¶

This page explains the latest features, optimizations, bug fixes, and more included in 20230731 DCE 5.0 Enterprise. Each product module in DCE 5.0 is independently decoupled. The versions mentioned in this release are included by default in the DCE 5.0 installer. If you need to upgrade or replace individual modules separately, please contact Daocloud's customer support personnel to download the proper version and perform the upgrade. Please also refer to the DCE 5.0 Enterprise Deployment Requirements.

Included Modules¶

New Features and Optimizations¶

Below are some of the new features and optimizations included in the latest release:

Installer v0.10.0¶

• Added support for Oracle Linux R8-U7 operating system
• Added import-artifact subcommand to import offline resources based on the external services defined in clusterConfig.yaml configuration file
• Improved the detection layer of clusterConfig by refactoring and abstracting it
• Improved error messages in the pre-dependency installation script
• Improved the installation process to allow installation to continue when Elasticsearch health status is yellow during minimal installation

Microservice Engine v0.26.2¶

• Added support for API-level global rate limiting
• Added support for domain-level global rate limiting
• Added support for batch operations (online/offline/delete) on API lists
• Added support for extending capabilities of cloud native microservice governance through plugins, including JWT plugin, Auth plugin, Wasm plugin, global rate limiting plugin, etc.

Insight v0.19.1¶

• Added Kubernetes event analysis and query functionality
• Added built-in alert strategies with alert repair recommendations
• Added support for customizing log export fields and formats
• Improved logic for retrieving Elasticsearch component status in system components
• Improved support for importing default message templates in installation parameters configuration

Service Mesh v0.18.0¶

• Added recommended versions in the Istio version list
• Added service diagnostics with automatic repair and manual repair guides
• Added support for high availability deployment configuration of mesh control plane across multiple clusters
• Improved default filtering of system namespaces in namespace sidecar
• Improved mesh permissions to adapt to workspace updates

Multicloud Management v0.11.1¶

• Added support for viewing event information in resource details
• Upgraded Operator to the community version of Karmada

Container Management v0.20.0¶

• Added support for viewing Helm operation logs in the Helm application interface
• Added support for heterogeneous node integration in workspace clusters
• Added support for batch importing nodes when creating clusters
• Added support for creating NFS type data volumes in storage
• Added support for vGPU, including automatic detection of node CPU and support for adding negative CPU quotas as workload configuration
• Improved cluster integration logic - when integrating a cluster for the second time with a new management platform, it is necessary to clean up redundant data from the old management platform before integration. For more details on cluster integration, refer to Uninstall/Remove Cluster
• Improved upgraded Clusterpedia to v0.7.0
• Improved page interactions based on permissions - users without permission will be unable to access pages without resource permissions
• Improved support for configuring advanced parameters such as kernel tuning in integrated nodes
• Improved installation detection mechanism for Insight components
• Fixed an issue where Helm tasks were stuck in "Installing" or "Uninstalling" state
• Fixed an issue with checking the kernel version during cluster node creation
• Fixed an issue with customizing namespaces when creating cluster plugins
• Fixed an issue with adding ca.crt data by default when updating secrets

Container Network v0.8.0¶

• Added nodeName and multusName fields in Spiderpool CR to support node topology and configure networks as needed
• Added Spiderpool SpiderMultusConfig CR to simplify writing JSON-formatted Multus CNI configurations and automatically manage Multus NetworkAttachmentDefinition CR
• Added Coordinator plugin in Spiderpool to solve issues such as Underlay Pod unable to access ClusterIP, routing of tuned Pods, detection of IP conflicts, and reachability of Pod gateways. Refer to Coordinator Documentation.
• Added deep support for IPVlan, suitable for any public cloud environment
• Added support for multiple default IP pools, simplifying usage
• Added CNI plugin Ifacer , used to automatically create sub-interfaces. Refer to Ifacer Documentation
• Added specifying default route network interface through Pod annotation
• Added support for enabling/disabling automatic pool recycling, allowing customization of whether the automatic pool should be deleted
• Improved support for elastic IP of cluster subnet, effectively solving the issue of old Pods not being deleted and new Pods lacking available IPs during rolling updates.

Kubean v0.8.0¶

• Added configuration for retaining a specified number of cluster operation records
• Added Apache-2.0 open-source license

Workbenchv0.19.3¶

• Added the ability to navigate to Jira from the Issues interface to view details
• Added support for editing YAML, adding components, editing components, and editing operational features in OAM applications
• Added support for Sonarqube configuration and code quality result steps in pipelines

Middleware¶

• Added access whitelist configuration for Mcamel-Redis
• Added configuration capability for PVC containing the log directory in Mcamel-MongoDB
• Added anti-affinity capability for workloads in Mcamel-MongoDB and Mcamel-RabbitMQ
• Added backup management capability for Mcamel-PostgreSQL
• Added access restrictions for UI interfaces of all middleware instances under Data Services
• Improved create instance dialog of Mcamel-Redis by adding default anti-affinity label value to simplify configuration process
• Improved data recovery interface of Mcamel-Redis
• Improved data recovery interface of Mcamel-MySQL
• Improved create instance dialog of Mcamel-MongoDB by adding default anti-affinity label value and allowing creation in non-operator namespaces
• Improved display of frontend interface permissions related information in Mcamel-MongoDB
• Improved monitoring charts of Mcamel-PostgreSQL by removing distracting elements and adding time range selection
• Fixed failure to disable node affinity in Mcamel-Redis
• Fixed an issue with some panels not displaying in the monitoring charts of Mcamel-PostgreSQL

Container Registry v0.10.0¶

• Added migration/backup/restore solutions for container registries, verified through migration of the Release-ci repository
• Added best practices for logging into non-secure container registries
• Added support for using internal middleware MinIo when creating hosted Harbor
• Added support for People's Congress Gold Warehouse PostgreSQL mode
• Improved format validation of PostgreSQL and Redis addresses when creating hosted Harbor
• Improved grayed out + prompt for no permission
• Improved handling of special cases after unbinding clusters

Global Management v0.19.2¶

• Added support for binding Mesh/Mesh-namespace resources to workspace-resource groups
• Added tips prompt for custom role permissions points
• Added ability to limit multiple concurrent sessions for a single user in platform settings-security policies
• Added ability to limit the maximum number of concurrent session connections for the system in platform settings-security policies
• Added validation to check if there are still resources in the Container Registry when deleting workspaces
• Improved graying out of buttons and menus without permissions in Ghippo, with a prompt for no permission on hover

Virtual Machine v0.1.0¶

• Added support for displaying virtual machine lists through clusters
• Added support for creating virtual machines from container images
• Added support for powering off/starting up and deleting virtual machines
• Added support for accessing virtual machines via console (VNC)
• Added support for viewing virtual machine details

Upgrade Considerations¶

• Microservice Engine: When upgrading from a version greater than v0.24.2 of Microservice Engine, there is an incompatible update regarding the gateway due to changes in the open-source component repository. Therefore, before upgrading, manually delete the old gateway-api-admission-xxxxx Job and then proceed with the normal upgrade process.
• Global Management: When upgrading from v0.15.x (or lower) to v0.19.0 (or higher) of Global Management, the database connection Helm parameter format needs to be modified. For details, refer to Offline Upgrade of Global Management.

Compatibility¶

The supported versions of each module for Global Management v0.19.0 are as follows:

• Container Management v0.18.1 and above
• Insight v0.17.2 and above
• Workbenchv0.17.3 and above
• Multicloud Management v0.9.1 and above
• Microservice Engine v0.22.0 and above
• Service Mesh v0.16.2 and above
• Middleware RabbitMQ v0.11.0 and above
• Middleware mysql v0.9.0 and above
• Middleware elasticsearch v0.8.0 and above
• Middleware redis v0.8.0 and above
• Middleware MinIO v0.6.0 and above
• Middleware Kafka v0.6.0 and above
• Middleware PostgreSQL v0.2.0 and above
• Container Registry v0.8.0 and above

Known Issues¶

Known issues with Installer v0.10.0:

• After restarting kubelet service in Redhat v8.6 operating system with kind runtime, the service fails to start with the error: failed to initialize top level QOS containers: root container [kubelet kubepods] doesn't exist
• When installing a cluster based on TencentOS 3.1, the package manager cannot be identified correctly. If TencentOS 3.1 is required, please use Installer v0.9.0.

Installation¶

• DCE 5.0 Enterprise Installation
• DCE Community Installation

Download Center¶

• DCE 5.0 Download Center

Comments