metrics-server

更新时间:
复制 MD 格式

Built on the open-source Metrics Server, this offline monitoring component collects pod CPU and memory metrics and exposes them through the Metrics API for HPA scaling and basic resource monitoring. Metrics Server , collects pod CPU and memory metrics and exposes them via the Metrics API for HPA scaling.

Usage notes

Pod resource metrics are available in Monitor basic resources.

See FAQ about metrics-server.

Release notes

June 2026

Version Image address Release date Description Impact
v0.3.14-079483b-aliyun registry.{REGION}.aliyuncs.com/acs/metrics-server:v0.3.14-079483b-aliyun June 2026 Aligned the calculation of allocated container resources (requests) with the bin-packing logic of the Kubernetes scheduler. Fixed an issue where the allocated node resources displayed for init containers and sidecar containers did not match the scheduler's actual allocation. No impact on workloads.

February 2026

Version Image address Release date Description Impact
v0.3.12-08c15e8-aliyun registry.{REGION}.aliyuncs.com/acs/metrics-server:v0.3.12-08c15e8-aliyun February 2026
  • Added basic resource monitoring for Hybrid nodes in hybrid cloud node pools.
  • Added support for displaying the resource usage of each workload in the workload list of the ACK console through an aggregation API.
No impact on workloads.

January 2024

Version Image address Release date Description Impact
v0.3.9.7-85b3699-aliyun registry.{REGION}.aliyuncs.com/acs/metrics-server:v0.3.9.7-85b3699-aliyun January 2024
  • Optimized monitoring performance for large-scale clusters.
  • Stabilized HPA scaling during rolling updates. To enable, set --enable-hpa-rolling-update-skipped=true. See FAQ about auto scaling.
No impact on workloads.

June 2023

Version Image address Release date Description Impact
v0.3.9.5-1e8461a-aliyun registry.{REGION}.aliyuncs.com/acs/metrics-server:v0.3.9.5-1e8461a-aliyun June 2023
  • Improved component security.
  • Reduced unnecessary log output.
No impact on workloads.

May 2023

Version Image address Release date Description Impact
v0.3.9.4-ff225cd-aliyun registry.{REGION}.aliyuncs.com/acs/metrics-server:v0.3.9.4-ff225cd-aliyun May 2023 Compatible with ACK clusters running Kubernetes 1.26. No impact on workloads.

March 2023

Version Image address Release date Description Impact
v0.3.9.3-ff225cd-aliyun registry.{REGION}.aliyuncs.com/acs/metrics-server:v0.3.9.3-ff225cd-aliyun March 2023 Fixed: metrics not collected from virtual nodes in ACK Edge clusters. No impact on workloads.
v0.3.9.3-628a248-aliyun registry.{REGION}.aliyuncs.com/acs/metrics-server:v0.3.9.3-628a248-aliyun February 2023 Fixed: CPU metrics spike when monitoring data collection is interrupted. No impact on workloads.

July 2022

Version Image address Release date Description Impact
v0.3.9-d554015-aliyun registry.{REGION}.aliyuncs.com/acs/metrics-server:v0.3.9-d554015-aliyun July 2022
  • Patched container image vulnerabilities.
  • Compatible with ACK clusters running Kubernetes 1.22.
No impact on workloads.

May 2022

Version Image address Release date Description Impact
v0.3.8.7-307cf45-aliyun registry.{REGION}.aliyuncs.com/acs/metrics-server:v0.3.8.5-307cf45-aliyun May 2022 Compatible with ACK clusters running Kubernetes 1.22. No impact on workloads.

February 2022

Version Image address Release date Description Impact
v0.3.8.6-307cf45-aliyun registry.{REGION}.aliyuncs.com/acs/metrics-server:v0.3.8.5-307cf45-aliyun February 2022 Compatible with ACK clusters running Kubernetes 1.22 and earlier. No impact on workloads.

May 2021

Version Image address Release date Description Impact
v0.3.8.5-307cf45-aliyun registry.{REGION}.aliyuncs.com/acs/metrics-server:v0.3.8.5-307cf45-aliyun May 2021
  • Compatible with ACK clusters running Kubernetes 1.20 and earlier.
  • Pod basic resource monitoring now supported by the latest CloudMonitor.
No impact on workloads.

April 2021

Version Image address Release date Description Impact
v0.2.2-b4bf266-aliyun registry.{REGION}.aliyuncs.com/acs/metrics-server:v0.2.2-b4bf266-aliyun April 2021
  • Added metric collection for Windows and Linux nodes.
  • Added scaling periods: 15, 20, 30, and 60 seconds.
  • Fixed: HPA adds excess pods during rolling updates.
Update to the latest version.