KubeSphere Enterprise v3.3.1 Release Notes
Starting from v3.3.1, the previous QingCloud KubeSphere Container Platform (QKCP) has been renamed to KubeSphere Enterprise.
Observability
Features
- Add Whizard v0.5.0 as the Prometheus long-term storage for multi-cluster monitoring & alerting.
- Add the Observability Center, which supports multi-cluster monitoring and alerting.
- Introduce three new CRDs
GlobalRuleGroups
,ClusterRuleGroups
, andRuleGroups
for multi-cluster, cluster, and project alerting respectively. Alert rules are now defined inGlobalRuleGroups
,ClusterRuleGroups
orRuleGroups
, which will be converted to groups inPrometheusRules
. - Add a controller to sync
PrometheusRules
to the host cluster. - Add an admission webhook and controllers for
RuleGroups
,ClusterRuleGroups
, andGlobalRuleGroups
. - Built-in alert rules can be modified, enabled, disabled, and reset now.
- Add support for alerting of KubeSphere core components.
- Add support for cleaning up a cluster's notification settings when removing this cluster.
- Add the Feishu notification channel.
- Add support for silencing notifications.
- Add support for changing the language of notifications.
- Add support for adding a user account's email address to this user's notification settings.
- Use OpenSearch as built-in log storage instead of Elasticsearch.
- Add the OpenSearch curator.
- Add the OpenSearch Dashboard.
- Add OpenSearch as the log receiver.
- Add support for searching logs from OpenSearch in the KubeSphere console.
- Add process and thread monitoring metrics for containers.
- Add disk monitoring metrics, for example, disk usage data.
- Add support for importing Grafana templates to create a custom monitoring dashboard in a namespace.
- Add support for defining data retention periods for container logs, Kubernetes event logs, and audit logs respectively.
Enhancements & Updates
- Improve the cluster overview page.
- Improve workspace and project monitoring.
- Upgrade Prometheus to v2.39.1, Thanos to v0.29.5 (customized), and kube-state-metrics to v2.6.0.
- Upgrade fluentbit-operator to v0.14.0.
- Upgrade Notification Manager to v2.1.0.
- Upgrade Alertmanager from v0.21.0 to v0.23.0.
- Upgrade Grafana from v7.4.3 to v8.3.3.
- Upgrade node-exporter from v0.18.1 to v1.3.1.
- Upgrade Prometheus Operator from v0.43.2 to v0.55.1.
- Upgrade kube-rbac-proxy from v0.8.0 to v0.11.0.
- Upgrade configmap-reload from v0.3.0 to v0.5.0.
- Upgrade kube-events from v0.3.0 to v0.4.0.
- Upgrade Fluent Bit from v1.8.3 to v1.9.4.
Bug Fixes
- Fix the style of log details in the toolbox.
- Fix inaccurate time unit of the monitoring metrics.
- Fix the "No Cluster Available" issue during log search.
DevOps
Features
- Add the Continuous Deployment feature, which supports GitOps and uses Argo CD as the backend.
- Optimize pipeline templates loaded from APIs.
- Add the allowlist feature on the Basic Information page of a DevOps project.
- Add the Events tab on the pipeline run record page.
Enhancements & Updates
- Add support for editing the kubeconfig binding mode on the pipeline UI.
- Add a message indicating that the S2I and B2I features currently do not support the containerd runtime.
Bug Fixes
- Fix an issue where Jenkins updates are not synchronized in real time.
- Fix the cron expression check failure of Jenkins.
- Fix an issue in the Import Code Repository dialog box, where repositories and organizations are not completely displayed.
- Fix an issue where users fail to check the CI/CD template.
- Remove the
Deprecated
tag from the CI/CD template and replacekubernetesDeploy
withkubeconfig binding
at the deployment phase. - Fix an issue where pipeline parameters are not updated in time.
- Fix the automatic cleaning behavior of NoSCMPipeline, and keep it consistent with the cleaning configuration of Jenkins.
Multi-Tenancy & Multi-Cluster
Features
- Add support for grouping cluster by labels.
- Add a limit on the number of labels that can be added to a cluster.
- Add support for automatically renewing a cluster's kubeconfig.
- Add metrics for cluster certificate expiration. Users will be notified when the kubeconfig is about to expire in 7 days or 24 hours.
- Add support for removing the cluster warning message.
- Add the name of current cluster to the
kubesphere-config
configmap. - Support cluster member management and cluster role management.
Bug Fixes
- Set the ready status to false when a cluster fails to join the federation control plane.
Microservice
Features
- Support Spring Cloud and add Spring Cloud Operator.
- Support Spring Cloud configurations and Spring Cloud Microservice in Spring Cloud Operator.
- Support Spring Cloud Gateway in Spring Cloud Operator.
- Support Microservice Aggregated API in Spring Cloud Operator.
- Add GatewayRouteConfig in Spring Cloud Operator.
- Add the springcloud-controller helm chart.
- Add kubesphere-spring-cloud-starter.
Storage
Features
- Support PVC resizing, which can automatically expand PVCs when the remaining capacity is insufficient.
- Support volume snapshot content management and volume snapshot class management.
- Allow users to set authorization rules for storage classes so that storage classes can be used only in specific projects.
- Provide usage data of each disk.
Bug Fixes
- Set
hostpath
as a required option when users are mounting volumes. - Update storageClass-accessor, so that deleting storage resources no longer requires accessor validation.
Authentication & Authorization
Features
- Add support for manually disabling and enabling users.
Enhancements & Updates
- Check user permissions to clusters when users are creating or updating workspace templates.
Bug Fixes
- Add support for filtering workspace roles using the LabelSelector.
- Add support for customizing or randomly setting an initial admin password.
- Delete annotations in
role-template-manage-users
,role-template-view-members
androle-template-manage-roles
. - Fix an issue where
cluster-admin
cannot view and manage the configmap, secret, and service account. - Delete role
workspace-manager
. - Add role
platform-self-provisioner
. - Delete role
users-manager
. - Block
role-template-manage-groups
.
DMP
Features
DMP Framework
- Refactor monitoring by adding application monitoring information.
ClickHouse
- Support ClickHouse kernel version 21.1.3.32.
- Support more ClickHouse metrics.
- Support horizontal and vertical expansion.
- Support SVC-based cluster load balancing.
- Support ClickHouse HA.
- Support more cluster states.
- Support disk resizing with the QingCloud CSI.
- Add support for automatically provisioning and managing ZooKeeper clusters.
MySQL
- Add support for setting the S3 and NFS backup schedule in a unified manner.
- Add version validation for MySQL sidecar image.
- Add support for backing up to NFS address in the ip:/directory format.
- Add support for validating MySQL images against MySQL version upon cluster creation.
- Add support for switching to the single-node mode when necessary.
- Support Chaos Mesh in E2E testing.
- Simplify cluster NFS backup with nfsbcp.
- Support MySQL configuration templates.
- Add support for customizing the MySQL version.
- Add capacity check when backing up to NFS.
- Add more status fields to the backup CRD.
- Update the version of Go to 1.17.
- Support helm installation of mysql-operator.
ECK
- Support the Hot-Warm-Cold architecture.
- Add support for deploying Kibana in the console.
- Add Elasticsearch Metrics Exporter.
Kafka
- Support Kafka v3.1 and v3.2.
- Add support for installing and managing Kafka and ZooKeeper nodes in the console.
- Support automatic cluster failover and recovery.
- Support cluster capacity and storage expansion.
- Add Kafka Metrics Exporter.
- Support secure access to the internal with TLS.
- Add support for managing users, topics, and ACLs in the console.
- Add support forrebalancing clusters in the console.
MongoDB
- Add service monitoring for MongoDB clusters.
- Enhance database account management.
- Support dynamic disk expansion.
- Optimize the elegant start and stop logic.
- Add support for displaying more cluster states.
OpenSearch
- Support installation and management of OpenSearch, Dashboard, and Logstash.
- Support OpenSearch v2.3.0.
- Support vertical expansion of the cluster computing and storage resources
- Support adjusting the number of cluster nodes.
- Support rolling updates upon configurations change.
- Support the keystore feature.
- Add OpenSearch Metrics Exporter.
- Support secure access with TLS.
PostgreSQL
- Support PostgreSQL kernel versions v12.12, v13.8, v14.5, and v15.0.
- Support the PostGIS and Citus plugins.
- Support horizontal and vertical scaling.
- Support semi-sync streaming replication.
- Support more metrics in the console.
Redis
- Add Proxy for Redis clusters.
- Support Redis Cluster failover and auto-recovery.
- Support resource monitoring.
- Add requests and limits.
- Support password and password-free settings for Redis clusters.
- Add support for dynamically reloading configurations upon Redis configuration changes.
App Store
Features
- Add support for managing helm releases created by running the native helm command.
- Add new options in OpenPitrix.
Bug Fixes
- Fix a ks-controller-manager crash error caused by Helm controller NPE errors.
- Fix an issue that occurs while updating the traffic policy on the traffic monitoring page of an application.
Network
Enhancements & Updates
- Integrate OpenELB with KubeSphere for exposing LoadBalancer services.
Bug Fixes
- Fix an issue where the gateway of a project is not deleted after the project is deleted.
- Fix an issue during gateway log query, where the log query connection is not terminated after the query is completed.
- Fix an issue where users fail to create routing rules in IPv6 and IPv4 dual-stack environments.
- Update the OpenELB check URL.
Service Mesh
Bug Fixes
- Resolve port conflicts of virtual services that use multiple protocols.
KubeEdge Integration
Features
- Add support for logging in to common cluster nodes and edge nodes from the KubeSphere web console.
Enhancements & Upgrades
- Upgrade KubeEdge from v1.7.2 to v1.9.2.
- Remove EdgeWatcher as KubeEdge v1.9.2 provides similar functions.
Bug Fixes
- Change the cluster module key from
kubeedge
toedgeruntime
.
License
Features
- Add Cluster ID to the license control mechanism.
- Add license for the Observability Center.
- Add license for Spring Cloud.
- Add metrics and alert rules for the KubeSphere license, and notifications will be sent if the KubeSphere license is about to expire in less than 30 days.
- Add support for showing the license ID on the license page.
Security
Features
- Add Ingress to support HTTPS.
Enhancements & Updates
- Enforce the password length to be at least 8 characters.
- Add the logic to check whether the username and password are the same.
- Disable password auto-filling.
User Experience
Features
- Display events for abnormal pods.
- Add in-memory cache in addition to the original Redis cache.
- Update the product name and logo to KubeSphere Enterprise.
- Add the End User License Agreements.
- Add support for more languages in the KubeSphere web console.
- Add the lifecycle management feature for containers.
- Add support for creating container environment variables in batches from secrets and configmaps.
- Add a message in the Audit Log Search dialog box, which prompts users to enable the audit logs feature.
- Add data units in the Create Storage Class dialog box.
- Add the cluster viewing permission to a user when the user adds a cluster.
- Add a time range selector on the Traffic Monitoring tab page.
- Add a prompt to remind users to select a language or artifact type when users are building images.
- Add support for changing the number of items displayed on each page of a table.
- Add support for batch stopping workloads.
- Add the creator annotation to ensure information displayed on the pod details page is consistent with other details pages.
- Add support for displaying the revision record when the workload type is
statefulsets
ordaemonsets
. - Add a support page.
- Add support for duplicate name validation of containers and service names.
Enhancements & Updates
- Add support for setting the default container registry.
- Support the shortcut dropdown of the relevant events of the pods.
- Support quick redirection to the destination page and settings of the rows displayed on the page.
- Add support for hiding external links in the console.
- Remove the limit to 100 of the cpu-percent of HPA.
- Optimize the service details area on the Service Topology page.
- Prevent passwords without uppercase letters set through the backend CLI.
- Optimize the UI texts of clusters, workspaces, and project deletion.
- Set the Token parameter on the webhook settings page as mandatory.
- Improve user experience in the web console of member clusters.
- Optimize the Service Type and External Access columns of the service list.
- Improve the registry verification API to resolve registry verification failures.
Bug Fixes
- Fix an issue where statefulset creation fails when a volume is mounted to an init container.
- Fix an app installation failure, which occurs when users click buttons too fast.
- Fix an incorrect message displayed during workload creation, which indicates that resource requests have exceeded limits.
- Fix an issue where no data is displayed on the Traffic Management and Tracing tab pages in a multi-cluster project.
- Fix an issue where data in the service details area on the Service Topology page is not updated automatically.
- Fix an issue where existing settings are not displayed when a stateful service is being edited.
- Fix an issue where container probes are still displayed after they are deleted.
- Fix incorrect container names in the service details area on the Service Topology page.
- Fix the incorrect number of worker nodes on the Cluster Nodes page.
- Fix a workspace API error caused by an invalid cluster role.
- Fix an incorrect error message displayed when kubeconfig is updated for a second time.
- Fix an incorrect private key type in kubeconfig information.
- Fix an issue where the key is not display in LoadBalancer.
- Fix an issue where no prompt is displayed when users do not fill in key and value in the LoadBalancer access mode.
- Fix an issue where the update time of a service is incorrect on the details page.
- Fix the inaccurate prompt when users are adding an init container.
- Fix an issue where an error occurs while users enter Chinese characters in secret.
- Fix an issue where the total number of pages is incorrectly displayed.
- Fix an issue where the update time of an application is incorrect.
- Fix a 404 issue in Documentation.
- Fix an issue where the status of a cluster remains true when the cluster fails to join Federation.
- Fix an issue where traffic allocation fails in the canary release mode.
- Fix an issue where configurations do not take effect when users set the pod request to 0.
- Fix an issue where canary release goes wrong when multiple operating systems are selected.
- Fix an issue where configmap configurations cannot be saved while users are creating a workload.
API Changes
- Upgrade the alerting API to v2beta1, and the v2alpha1 alerting API is deprecated.
- Expose kube-apiserver of the host cluster as a LoadBalancer service for member clusters to access.
- Provide RESTful APIs for ClusterTemplate.
- Provide template-related APIs.
- Change the KubeEdge proxy service to
http://edgeservice.kubeedge.svc/api/
. - Change the patch type of
PatchWorkspaceTemplate
fromMergePatchType
toJSONPatchType
.