All Products
Search
Document Center

Container Service for Kubernetes:CoreDNS

Last Updated:Nov 21, 2025

CoreDNS is the default DNS service discovery plug-in for ACK and ACK Edge clusters. You can also use the CoreDNS component for service discovery in ACK Serverless clusters. This topic describes the CoreDNS component, explains how to use it, and provides its change log.

Component introduction

CoreDNS is the component in Kubernetes clusters that handles DNS resolution. It resolves both internal custom service domain names and external domain names. The CoreDNS project is hosted by the Cloud Native Computing Foundation (CNCF). For more information about CoreDNS, see CoreDNS: DNS and Service Discovery.

CoreDNS has two editions: unmanaged and managed. For more information, see DNS for service discovery.

For more information about the mappings between CoreDNS versions and cluster versions, see CoreDNS version in Kubernetes.

Notes

Change log

Unmanaged edition

Version number

Applicable clusters

Image URL

Change date

Changes

Impact

v1.12.1.3

For clusters of v1.27 and later.

registry.{{.Region}}.aliyuncs.com/acs/coredns:v1.12.1.3

November 5, 2025

This upgrade does not affect your services.

v1.12.1.2

For clusters of v1.21 and later.

registry.{{.Region}}.aliyuncs.com/acs/coredns:v1.12.1.2

October 10, 2025

  • Optimization:

    • Removed the aliyun suffix from the component version number.

    • The default configuration of cache for the serve_stale plug-in is optimized to serve_stale 30s verify.

This upgrade does not affect your services.

v1.12.1.1-4035d7a99-aliyun

For clusters of v1.21 and later.

registry.{{.Region}}.aliyuncs.com/acs/coredns:v1.12.1.1-4035d7a99-aliyun

July 22, 2025

Upgraded to the community edition v1.12.1.

  • Added the multisocket plug-in.

  • Increased the CNAME lookup limit from 7 to 10.

  • plugin/kubernetes: Fixed an issue where DNS queries were resolved to deleted pods due to DeletionTimestamp (#7119) (#7131).

  • plugin/kubernetes: Reverted the change "Only create PTR records for endpoints with a defined hostname (#6898)".

  • plugin/forward: Added the failfast_all_unhealthy_upstreams option. This option returns servfail when all upstream servers are down.

  • The cache plug-in enables the serve_stale configuration by default. When an upstream DNS server is unavailable, CoreDNS can use an expired cache to provide an acknowledgement (the TTL of this acknowledgement is 0) and asynchronously tries to obtain the latest address of the domain name.

This upgrade does not affect your services.

v1.11.3.5-5321daf49-aliyun

For clusters of v1.21 and later.

registry.{{.Region}}.aliyuncs.com/acs/coredns:v1.11.3.5-5321daf49-aliyun

March 19, 2025

  • Updated the base image to fix related security vulnerabilities.

  • Added support for Node Lingjun pools. CoreDNS pods are not scheduled to Node Lingjun.

This upgrade does not affect your services.

v1.11.3.2-f57ea7ed6-aliyun

For clusters of v1.21 and later.

registry.{{.Region}}.aliyuncs.com/acs/coredns:v1.11.3.2-f57ea7ed6-aliyun

October 21, 2024

  • CoreDNS now supports the Firewall plug-in. The Forward plug-in can execute the next plug-in based on the return code.

  • You can customize the topology constraint options for CoreDNS component deployments on the Operations Management > Component Management page of the console.

This upgrade does not affect your services.

v1.9.3.16-4341f22f-aliyun

For ACK Serverless clusters of v1.20.4 and later only.

registry.{{.Region}}.aliyuncs.com/acs/coredns:v1.9.3.16-4341f22f-aliyun

May 9, 2023

The default memory request for scheduling CoreDNS containers is increased to 4 GiB to prevent CoreDNS from being scheduled to shared instances. You can customize the memory size in the component configuration.

The upgrade may result in the use of higher ECI specifications when you create CoreDNS.

v1.9.3.10-7dfca203-aliyun

For clusters of v1.20.4 and later.

registry.{{.Region}}.aliyuncs.com/acs/coredns:v1.9.3.10-7dfca203-aliyun

April 3, 2023

  • Optimized zone-level anti-affinity scheduling.

  • Reduced the possibility of pod eviction in elastic node scenarios.

Due to scheduling policy adjustments, if all schedulable nodes in the cluster are in a single zone, CoreDNS replicas may fail to be scheduled and the component upgrade may fail. To ensure DNS availability, scale out your cluster and distribute the schedulable nodes across multiple zones. This ensures that CoreDNS replicas can be scheduled and run.

v1.9.3.6-32932850-aliyun

For clusters of v1.20.4 and later.

registry.{{.Region}}.aliyuncs.com/acs/coredns:v1.9.3.6-32932850-aliyun

August 25, 2022

  • Supports the delivery of Kubernetes events.

  • In ACK Serverless clusters, the default CPU request is changed to 2 cores.

This upgrade does not affect your services.

v1.9.3.2-8850b5e7-aliyun

For clusters of v1.20.4 and later.

registry.{{.Region}}.aliyuncs.com/acs/coredns:v1.9.3.2-8850b5e7-aliyun

August 3, 2022

You can enable the CoreDNS log collection feature in Log Service with a single click.

This upgrade does not affect your services.

v1.9.3.1-5e7ba42d-aliyun

For clusters of v1.20.4 and later.

registry.{{.Region}}.aliyuncs.com/acs/coredns:v1.9.3.1-5e7ba42d-aliyun

July 11, 2022

  • Fixed several issues and added some features. For more information, see CoreDNS-1.9.3 Release.

  • Supports ACK One multi-cluster services.

This upgrade does not affect your services.

v1.8.4.5-2ce07fd2-aliyun

For clusters of v1.20.4 and later.

registry.{{.Region}}.aliyuncs.com/acs/coredns:v1.8.4.5-2ce07fd2-aliyun

April 8, 2022

Optimized the CoreDNS scheduling affinity configuration to allow all cluster nodes to be elastic scaling nodes.

This upgrade does not affect your services.

v1.8.4.3-644f4735-aliyun

For clusters of v1.20.4 and later.

registry.{{.Region}}.aliyuncs.com/acs/coredns:v1.8.4.3-644f4735-aliyun

February 22, 2022

  • Disabled caching for resolution results of the ServError type.

  • Changed anti-affinity scheduling by hostname from `preferred` to `required`. This enforces anti-affinity scheduling by node.

Because replica anti-affinity is enforced by node, some CoreDNS replicas will be in the Pending state if the number of CoreDNS replicas is greater than the number of nodes. Before you upgrade to this version, scale out the cluster nodes or scale in the CoreDNS replicas.

v1.8.4.2-7d597cff-aliyun

For clusters of v1.20.4 and later.

registry.{{.Region}}.aliyuncs.com/acs/coredns:v1.8.4.2-3a376cc-aliyun

January 10, 2022

  • Added support for custom parameters.

  • Enabled resolution logs by default.

This upgrade does not affect your services.

v1.8.4.1-3a376cc-aliyun

For clusters of v1.20.4 and later.

registry.{{.Region}}.aliyuncs.com/acs/coredns:v1.8.4.1-3a376cc-aliyun

October 26, 2021

  • Supports listeners for EndpointSlices resources.

  • Supports DNS queries using IPv6 addresses.

This upgrade does not affect your services.

v1.7.0.0-f59c03d-aliyun

For clusters of v1.14.8 and later.

registry.{{.Region}}.aliyuncs.com/acs/coredns:v1.7.0.0-f59c03d-aliyun

July 8, 2021

  • Modified the default graceful shutdown time for CoreDNS and the container memory limit in the CoreDNS deployment.

  • Updated metric names. If your monitoring system depends on CoreDNS metrics, make the necessary changes. For more information, see Metric Changes.

  • Fixed an issue where the Forward plug-in used only the first upstream DNS server.

  • Removed compatibility for the deprecated Upstream plug-in. If your Corefile configuration contains the Upstream plug-in, it is safely and automatically deleted during the upgrade.

If you previously modified the /etc/resolv.conf DNS configuration file on an ECS instance, upgrading or recreating a CoreDNS pod causes it to use the modified /etc/resolv.conf file. Before you upgrade, make sure that all DNS servers in the configuration are working correctly.

1.6.7.edge (Discontinued)

-

registry.{{.Region}}.aliyuncs.com/acs/coredns:1.6.7.edge

April 23, 2021

Built based on the community edition 1.6.7. For more information, see CoreDNS-1.6.7 Release.

This upgrade does not affect your services.

v1.7.0

-

registry.{{.Region}}.aliyuncs.com/acs/coredns:1.7.0

March 18, 2021

  • Removed compatibility for the deprecated Upstream plug-in. If your Corefile configuration contains the Upstream plug-in, it is safely and automatically deleted during the upgrade.

  • Updated metric names. If your monitoring system depends on CoreDNS metrics, make the necessary changes. For more information, see Metric Changes.

  • Fixed an issue where the Forward plug-in used only the first upstream DNS server.

If you previously modified the /etc/resolv.conf DNS configuration file on an ECS instance, upgrading or recreating a CoreDNS pod causes it to use the modified /etc/resolv.conf file. Before you upgrade, make sure that all DNS servers in the configuration are working correctly.

v1.6.7 (Discontinued)

-

registry.{{.Region}}.aliyuncs.com/acs/coredns:1.6.7

November 28, 2018

Built based on the community edition 1.6.7. For more information, see CoreDNS-1.6.7 Release.

This upgrade does not affect your services.

Managed edition

Version number

Applicable clusters

Image URL

Change date

Changes

Impact

v1.12.1.2

For clusters of v1.21 and later.

registry.{{.Region}}.aliyuncs.com/acs/coredns:v1.12.1.2

October 10, 2025

This upgrade does not affect your services.

v1.11.3.2-f57ea7ed6-aliyun

For clusters of v1.21 and later.

registry.{{.Region}}.aliyuncs.com/acs/coredns:v1.11.3.2-f57ea7ed6-aliyun

October 21, 2024

  • CoreDNS now supports the Firewall plug-in. The Forward plug-in can execute the next plug-in based on the return code.

  • You can customize the topology constraint options for CoreDNS component deployments on the Operations Management > Component Management page of the console.

This upgrade does not affect your services.

v1.9.3.10-7dfca203-aliyun

For clusters of v1.20.4 and later.

registry.{{.Region}}.aliyuncs.com/acs/coredns:v1.9.3.10-7dfca203-aliyun

April 3, 2023

  • Optimized zone-level anti-affinity scheduling.

  • Reduced the possibility of pod eviction in elastic node scenarios.

Due to scheduling policy adjustments, if all schedulable nodes in the cluster are in a single zone, CoreDNS replicas may fail to be scheduled and the component upgrade may fail. To ensure DNS availability, scale out your cluster and distribute the schedulable nodes across multiple zones. This ensures that CoreDNS replicas can be scheduled and run.