Lists ACK features released each month in 2021 with supported regions and documentation links.
Background
Supported Kubernetes versions: 1.24, 1.22, and 1.20.
Supported operating systems: CentOS 7.9, Alibaba Cloud Linux 3.2104, Alibaba Cloud Linux 2.1903, Windows Server 2019, and Windows Server Core 1909.
December 2021
Feature | Description | Region | References |
ACK One | ACK One is a distributed cloud container platform for hybrid cloud, multi-cluster, distributed computing, and disaster recovery scenarios. Register external Kubernetes clusters from any region or infrastructure, then centrally manage computing, networking, storage, security, monitoring, logs, jobs, applications, and traffic. Compatible with open-source Kubernetes APIs. | All regions | |
Kubernetes 1.22.3 | Kubernetes 1.22.3 is now available when creating clusters. | All regions | |
Deployment sets for node pools | Associate a node pool with a deployment set to distribute ECS nodes across multiple physical servers. Configure pod affinity to place application pods on different nodes for availability and disaster recovery. | All regions | Best practices for associating deployment sets with node pools |
Workbench | Log in to containers using Workbench, which offers higher stability and compatibility than the previous terminal tool. On the Pods page, click Terminal in the Actions column. | All regions | |
Custom configurations for the NGINX Ingress controller | Configure the NGINX Ingress controller directly from the Add-ons page in the ACK console. Set resource requests and limits, enable host network mode, enable admission webhooks, and specify node selectors. Custom configurations are retained after controller updates. | All regions | — |
Prometheus monitoring dashboards | CCM (cloud controller manager) and kube-controller-manager monitoring dashboards are now available for ACK Pro clusters. View dashboards on the Prometheus Monitoring page in the Operations module of the ACK console. | All regions | |
Log center | Collect CCM logs using the log center feature, now available in ACK Serverless Pro clusters. | All regions | Collect logs of control plane components in ACK managed clusters |
OPA-based policy governance | The Policy Governance feature is now available, built on Open Policy Agent (OPA) and the Gatekeeper admission controller. Predefined policies cover more Kubernetes scenarios than pod security policies (PSPs), with simpler configuration. | All regions | |
Node pool priorities | Assign priorities to node pools in the auto scaling policy. When multiple node pools qualify for a scale-out activity, ACK selects the node pool with the highest priority. | All regions | |
ALB Ingress controller open-sourced on GitHub | The ALB Ingress controller (v2.2.0) is now open-sourced. Compatible with the NGINX Ingress controller, it provides Layer 7 traffic routing through ALB instances, including complex routing, automatic certificate discovery, and HTTP, HTTPS, and QUIC protocol support. | All regions |
November 2021
Feature | Description | Region | References |
ALB Ingress-based routing | Create ALB Ingresses directly in the ACK console. Define Ingress rules, customize configurations, and use ALB instances for Layer 7 load balancing. | All regions | |
Multiple route tables via CCM | Configure multiple route tables for a Virtual Private Cloud (VPC) using the CCM. Set this up from the Add-ons page or by modifying the CCM ConfigMap. | All regions | |
Desired number of nodes | Scale a node pool by setting a desired node count. Values above the current count trigger a scale-out; values below trigger a scale-in based on the configured scaling policy. | All regions | |
ACK quotas | Quota display and increase requests are optimized in the ACK console for all cluster types, including ACK managed, ACK dedicated, ACK Serverless, ACK edge, and registered clusters. The console also links to Quota Center. | All regions | |
IPv4/IPv6 dual stack for ACK Serverless clusters | Enable IPv4/IPv6 dual stack when creating an ACK Serverless cluster to allow clients to access services using IPv6 addresses. Prerequisites: Kubernetes version 1.20.11-aliyun.1 or later, and a VPC with dual stack enabled. | All regions | |
ContainerOS | ContainerOS is an Alibaba Cloud operating system for containerized workloads, fully compatible with Kubernetes. Built on Alibaba Cloud Linux 3, it provides enhanced security, faster startup, simplified system services, and preinstalled cloud-native components. Available in managed node pools with free technical support. | All regions |
October 2021
Feature | Description | Region | References |
Kubernetes 1.20.11 | CVE-2021-25741 is a high-severity vulnerability that allows attackers to access host directories by exploiting symbolic links with subPath volume mounts. This vulnerability is fixed in Kubernetes 1.20.11. Upgrade from Kubernetes 1.20 to 1.20.11 is supported. | All regions | Vulnerability CVE-2021-25741 in Kubernetes and (Discontinued) Kubernetes 1.20 release notes |
ClusterRole management | Manage the full lifecycle of ClusterRoles directly from the ACK console. | All regions | Use custom RBAC roles to restrict resource operations in a cluster |
ARMS monitoring integration | The network topology feature of Application Real-Time Monitoring Service (ARMS) is now integrated with ACK. View network topologies for Services, workloads, and Alibaba Cloud resources directly in the ACK console. | All regions | — |
Cost analysis: application dashboards | Application dashboards in cost analysis now show cost trends, correlation analysis results, and cost-saving suggestions and plans. | All regions | |
Cloud-native AI component set | The cloud-native AI component set now supports model management, model evaluation, and login with non-Alibaba Cloud accounts to AI Dashboard and AI Developer Console. Fluid applications using JindoRuntime can be monitored with Prometheus Monitoring. | All regions | Cloud-native AI suite O&M guide, Cloud-native AI component set user guide, Manage models in MLflow Model Registry, Evaluate a model, and Log on to AI Developer Console |
ALB Ingress controller | The ALB Ingress controller is now available. It supports complex routing, automatic certificate discovery, and HTTP, HTTPS, and QUIC protocols for Layer 7 load balancing. Compatible with NGINX Ingresses. | All regions | |
ACK Serverless Pro cluster (public preview) | ACK Serverless Pro clusters are now in public preview. They offer higher reliability and security than standard ACK Serverless clusters, with an SLA that includes compensation clauses. Designed for enterprise workloads requiring strict stability and security at scale. | All regions |
September 2021
Feature | Description | Region | References |
ARM node pools | Create ARM node pools in ACK clusters using ECS instances from the g6r and c6r instance families. ARM instances reduce costs in general-purpose computing scenarios (NGINX, Redis, SQL) and deliver high concurrency and throughput for big data workloads. | All regions | |
Auto scaling for Windows node pools | Enable auto scaling for Windows node pools. | All regions | |
Custom Windows images for node pools | Specify Windows images when creating node pools. Custom images based on Windows Server 2019 with kernel versions later than 1809 are supported. | All regions | |
Multiple security groups for a node pool | Assign more than one security group to a node pool for fine-grained access control. | All regions | |
IPv4/IPv6 dual stack for ACK managed clusters | Enable IPv4/IPv6 dual stack when creating an ACK managed cluster to allow clients to connect to cluster applications via IPv6 addresses. | All regions | |
CIS Kubernetes V1.20 Benchmark v1.0.0 | The cluster inspection feature now supports CIS Kubernetes V1.20 Benchmark v1.0.0 for clusters running Kubernetes 1.20 and later. | All regions | — |
Node pool scale-out policy and scale-in settings | Configure the node pool scale-out policy (least-waste or random) and enable or disable scale-in when setting up auto scaling. The policy determines which node pools are prioritized during scale-out. | All regions | |
Backup center (public preview) | Back up, restore, and migrate stateless and stateful applications across ACK and self-managed clusters. The backup center (formerly the application backup feature) supports disaster recovery and migration for stateful applications in hybrid cloud and multi-cluster environments. | All regions | |
ACK resource plans | ACK supports resource plans to offset pay-as-you-go fees for ACK managed cluster Pro Edition. | All | See ACK resource plans. |
Model evaluation for AI project acceleration | Manage and evaluate models trained with the AI component set. Control model versions and evaluate models using metrics such as accuracy and recall to identify the best model. | All regions |
August 2021
Feature | Description | Region | References |
ACK Scheduler V1.20-ack-4.0: load-aware and elastic container instance-based scheduling | Load-aware scheduling places pods on nodes with lower historical load to prevent overloaded nodes. Elastic container instance-based scheduling lets you annotate workloads to use only ECS instances, only elastic container instances, or a mix — falling back to elastic container instances when ECS resources are insufficient. | All regions | Use load-aware pod scheduling and Use elastic resources to implement Elastic Container Instance-based scheduling (discontinued) |
CCM 2.0.1: weighted routing and vServer group reuse | CCM 2.0.1 adds four new annotations: | All regions | |
Subscription clusters | Create clusters with the subscription billing method. ECS nodes and the SLB instance attached to the Kubernetes API server are billed under subscription. Purchase resource plans for elastic IP addresses (EIPs), NAT gateways, and Log Service projects to consolidate cluster payments at creation time. | All regions | — |
Reuse an existing SLB instance for the NGINX Ingress controller | When creating a subscription cluster, associate a previously created subscription SLB instance with the NGINX Ingress controller to consolidate all cluster payments at creation time. | All regions | — |
Intelligent O&M: cluster diagnostics and global check | Run a one-click global check on cluster resources, components, and configurations to get issue-fix suggestions. Use cluster diagnostics to troubleshoot nodes, pods, and networks. | All regions | |
ACK Serverless Ingress ALB controller | Install the ALB Ingress controller when creating an ACK cluster or from the Add-ons page after creation. It supports complex routing, automatic certificate discovery, and HTTP, HTTPS, and QUIC protocols, and is compatible with the NGINX Ingress controller. | All regions | |
Windows Server 2019 for ACK edge clusters | Select Windows Server 2019 when creating cloud node pools in ACK edge clusters, improving cloud-edge coordination for Windows applications. | All regions | — |
CNFS: shared NAS volumes and auto expansion | Mount a NAS volume in sharepath mode to share it across multiple applications or pods. Configure auto expansion policies so NAS volumes expand automatically when usage exceeds a threshold. | All regions |
July 2021
Feature | Description | Region | References |
Kubernetes version update | Upgrade from Kubernetes 1.18 to Kubernetes 1.20 is now supported. | All regions | Manually update ACK clusters and (Discontinued) Kubernetes 1.20 release notes |
CoreDNS | CoreDNS is now supported on the Add-ons page. It is the default DNS-based service discovery plugin in ACK and ACK edge clusters, providing domain name resolution for services within a cluster. | All regions | |
Cost analysis by namespace | The cost analysis feature now provides resource usage trends and per-CPU-core cost estimates for applications and pods, broken down by namespace. | All regions | |
Enhanced security for registered clusters | Install security-inspector (security scanning), aliyun-acr-credential-helper (password-free image pulls), and Gatekeeper (OPA policy management) in registered clusters. | All regions | |
CNFS 1.0 | CNFS is now generally available. It abstracts NAS file systems as Kubernetes resources via CRD objects. Supports creating, deleting, setting quotas for, mounting, monitoring, and expanding NAS file systems declaratively. | All regions |
June 2021
Feature | Description | Region | References |
Resource group selection | Select a resource group from a dropdown when creating a cluster or node pool. The cluster and its ECS instances are assigned to the selected group. | All regions | — |
Visual network policy configuration | Configure Kubernetes network policies through a visual interface in the ACK console to control traffic at the IP address or port level. | All regions | |
ACK Terway Hubble | Deploy ACK Terway Hubble from App Catalog to gain observability into network traffic, network policies, and workload topology in ACK managed clusters. | All regions | Implement network observability by using ACK Terway and Cilium Hubble |
Cost analysis: node pool-level reporting | View cost allocations and trends for resources, applications, and containers at the node pool level, with cost optimization suggestions based on current costs and node pool pricing. | All regions | |
Auto scaling scan interval | Set the interval at which the cluster evaluates scaling conditions: 15 seconds, 30 seconds, or 1 minute. | All regions | |
Modifiable SANs for ACK Serverless clusters | Update custom subject alternative names (SANs) in the API server certificate of an ACK Serverless cluster after creation. Supports domain names, IP addresses, and URLs. | All regions | Customize the SANs of the API server certificate when you create an ACK cluster |
Security inspection for registered clusters | Use the inspection feature to detect security risks in the workloads of a registered cluster. | All regions | |
Topology-aware CPU scheduling | Topology-aware CPU scheduling now supports dynamic resource usage thresholds for workloads with different priorities, and L3 cache and Memory Bandwidth Allocation (MBA) for resource isolation across priority tiers. | All regions | Enable CPU Suppress and Enable resource isolation based on the L3 cache and MBA |
May 2021
Feature | Description | Region | References |
CIS reinforcement for worker nodes | Enable CIS reinforcement to harden OS security for worker nodes. Supports only Alibaba Cloud Linux 2, the default OS image in ACK clusters. | All regions | |
New region: China (Nanjing - Local Region) | ACK Pro clusters are now available in the China (Nanjing - Local Region) region. | China (Nanjing - Local Region) | |
New region: China North 2 Ali Gov | ACK Pro clusters are now available in the China North 2 Ali Gov region on Alibaba Gov Cloud. | China North 2 Ali Gov | |
Cost analysis | Cost analysis helps IT administrators analyze resource usage and allocate costs. Capabilities include cloud resource cost analysis, cost trends, cost-saving suggestions, real-time forecasting, namespace-based cost allocation, and application cost optimization. | All regions | |
Custom SSL certificates for ACK Serverless clusters | Specify custom SSL certificates for SLB instances using annotations when creating Ingresses in ACK Serverless clusters. SSL certificates no longer need to be set through Kubernetes Secrets. | All regions | — |
Topology-aware scheduling for AMD CPUs | resource-controller V1.2.1-d1e280f-aliyun is released. Works with ack-scheduler for Kubernetes 1.20.4 to support topology-aware scheduling on AMD CPUs. | All regions |
April 2021
Feature | Description | Region | References |
Kubernetes 1.20 | Kubernetes 1.20 is now available when creating clusters. | All regions | |
Hot migration from ACK dedicated clusters to ACK Pro clusters | Migrate workloads from ACK dedicated clusters to ACK Pro clusters without service interruptions. | All regions | Hot migration from ACK dedicated clusters to ACK Pro clusters |
NodeLocal DNSCache | ACK NodeLocal DNSCache is a local DNS cache solution based on the open-source NodeLocal DNSCache project. It runs a DNS caching DaemonSet and an admission controller Deployment that injects DNSConfig data, letting pods use the local cache to accelerate DNS lookups. | All regions | |
Registered cluster: event center and image pull | Registered clusters now support the Kubernetes event center feature and the aliyun-acr-credential-helper component for password-free image pulls. | All regions | Create a registered cluster and Use the aliyun-acr-credential-helper component to pull images without using a password |
Custom configuration for control plane components | ACK Pro clusters now support custom parameters for kube-apiserver and kube-controller-manager to meet production environment requirements. | All regions | Customize the parameters of control plane components in ACK Pro clusters |
Alerting | Centralized alert management is now available. Configure alerts using CRD objects deployed in the cluster. Enabled by default when creating clusters, with predefined metrics covering multiple scenarios. | All regions |
March 2021
Feature | Description | Region | References |
Node pool data export | Export node information from a node pool to CSV files. | All regions | |
Updatable SANs for ACK managed clusters | Update subject alternative names (SANs) in the API server certificate for ACK standard and ACK Pro clusters. | All regions | Customize the SANs of the API server certificate when you create an ACK cluster |
Temporary kubeconfig files | Generate temporary kubeconfig files for ACK cluster access, valid for 30 minutes to 3 days. | All regions | Obtain the kubeconfig file of a cluster and use kubectl to connect to the cluster |
containerd runtime | Select containerd as the container runtime when creating a cluster or node pool. Deploy both containerd and Docker containers in the same cluster. Migrating from Docker to containerd requires recreating pods; hot migration is not supported. | All regions |
February 2021
Feature | Description | Region | References |
ACK Edge Pro cluster | Create ACK Edge Pro clusters, which provide the same reliability, stability, and billing methods as ACK Pro clusters. | All regions | |
Log center | The log center is now available in the ACK console. View cluster logs and control plane component logs in one place. | All regions | Query the logs of control plane components and View cluster logs |
Prometheus monitoring: CoreDNS dashboard | A CoreDNS dashboard is now available on the Prometheus Monitoring page in the ACK console. | All regions | |
EIPs for node pools | Associate elastic IP addresses (EIPs) with regular and managed node pools. Nodes automatically attach EIPs at node pool creation, or configure a NAT gateway for cluster-wide internet access. | All regions | |
New region: China South 1 Finance | ACK Pro clusters are now available in the China South 1 Finance region. | China South 1 Finance | |
AI engineering acceleration suite is released for public preview | The ACK AI engineering acceleration suite improves heterogeneous platform resource utilization and AI engineering delivery speed through job orchestration and resource virtualization. Key optimization areas:
| All regions |
January 2021
Feature | Description | Region | References |
Control plane observability for ACK Pro clusters | API server and etcd observability is now enabled in ACK Pro clusters. Monitor these components in dashboards and receive alerts on exceptions. | All regions | |
Custom configuration for control plane components | Customize parameters for kube-apiserver and kube-controller-manager in ACK Pro clusters to meet production environment requirements. | All regions | Customize the parameters of control plane components in ACK Pro clusters |
Log collection for control plane components | Collect logs from kube-apiserver, kube-controller-manager, and kube-scheduler. Enable log collection by selecting Enable for Log Collection for Control Plane Components when creating a cluster. | All regions | |
Preemptible instances for node pools | Use preemptible instances as the billing method for a node pool to reduce computing costs. Bid for idle Alibaba Cloud resources; instances run until reclaimed by higher bids from other customers. | All regions | Set the ratio of preemptible instances to pay-as-you-go instances in a node pool |
Edge node pools | Create edge node pools in ACK edge clusters to group nodes with common attributes and manage them uniformly across regions. Supports basic and enhanced cloud-edge coordination networks; the enhanced network uses SDN for higher network quality and security. | All regions | |
Elastic node pools for registered clusters | Use node pools in registered clusters to manage ECS instances with identical attributes and add them to self-managed Kubernetes clusters or third-party public cloud clusters. Enables unified resource scheduling across data centers and cloud environments. | All regions | |
Application backup | Back up applications, volumes, and persistent volumes (PVs) in ACK and registered clusters, and restore to other clusters. Supports stateless and stateful applications, and all resources in a namespace. | All regions | |
Cost reduction policy | Set the ratio of preemptible instances to pay-as-you-go instances in a node pool to reduce costs while maintaining performance stability. | All regions | Set the ratio of preemptible instances to pay-as-you-go instances in a node pool |