Operating system console releases
The latest updates to the operating system console enhance system management and support evolving applications, delivering improved performance, efficiency, and user experience.
Background
-
The operating system console supports specific versions of CentOS, Alibaba Cloud Linux, Anolis, and Ubuntu.
-
For a list of supported regions, see the documentation on region limitations.
July 2026
|
Feature |
Module |
Description |
Release date |
Related documentation |
|
OS Copilot |
Intelligent assistant |
Upgrades OS Copilot to OS Agent to provide a unified service entry. Adds intelligent parsing of command output, personalized feature recommendations based on historical memory, and one-click optimization by the agent. |
2026-07-30 |
|
|
System health and diagnostics |
One-click diagnostics |
Adds diagnostics for DNS anomalies, XFS and overlay mount and unmount anomalies, and backend storage anomalies on PVC nodes, together with kernel disassembly and source code analysis for TCP RST scenarios. Adds detection of kubelet and containerd startup anomalies for AI jobs and analysis of anomalies on the physical machines that run the jobs. |
2026-07-30 |
|
|
System health and diagnostics |
Java diagnostics |
Adds AI-assisted Java analysis and provides Java diagnostics as skills. |
2026-07-30 |
|
|
GPU performance and diagnostics |
AI Profiling |
The GPU topology page now supports a list of abnormal events. Flame graph analysis can now incorporate user prompts. Adds an offline AI Profiling skill. |
2026-07-30 |
|
|
Agent observability |
Anomaly diagnostics |
Adds push notifications for agent abnormal events. Agent anomaly diagnostics now cover scenarios such as tool call failures. |
2026-07-30 |
June 2026
|
Feature |
Module |
Description |
Release date |
Related documentation |
|
FastOOM |
FastOOM |
FastOOM now supports cgroup v2 and uses PSI to detect memory pressure. |
2026-06-30 |
|
|
System health and diagnostics |
One-click diagnostics |
Supports startup anomaly analysis for PAI AI jobs. Adds diagnostics for high pod CPU usage and high pod I/O. |
2026-06-30 |
|
|
GPU performance and diagnostics |
AI Profiling |
Optimizes the frontend display. |
2026-06-30 |
|
|
GPU diagnostics |
Detects and diagnoses GPU hardware anomalies. |
2026-06-30 |
||
|
OS Copilot |
Host inspection |
Adds abnormal kernel log events and metric inspection. |
2026-06-30 |
May 2026
|
Feature |
Module |
Description |
Release date |
Related documentation |
|
Component management |
SysOM |
Adds management and diagnostics support for Alibaba Cloud Linux 4. |
2026-05-30 |
|
|
System diagnostics |
One-click diagnostics |
1. Adds diagnostics for anomalies in critical system services. 2. Adds diagnostics for high pod load. 3. Adds one-click diagnostics for Lingjun. 4. Adds multi-dimensional GPU performance diagnostics for DLC training jobs. 5. Adds CPU load performance diagnostics for DLC training jobs. 6. Adds diagnostics for EAS job snapshot (checkpoint, restore) anomalies. 7. Adds pod memory state checks and attribute configuration diagnostics. |
2026-05-30 |
|
|
Memory diagnostics |
Adds large-page memory analysis and supports Java flame graph analysis. |
2026-05-30 |
||
|
Scenario diagnostics |
Adds node and container disk usage analysis, with file analysis for specific directories. |
2026-05-30 |
||
|
Agent observability |
Token analysis |
Adds agent token consumption statistics and analysis of trace and session interruptions. |
2026-05-30 |
April 2026
|
Feature |
Module |
Description |
Release date |
Related documentation |
|
GPU performance and diagnostics |
GPU resource topology |
Improves pod labeling and functionality. |
2026-04-30 |
|
|
Component management |
Configuration management |
Adds an "Automatic" memory pressure option to FastOOM and supports kernel-level QoS. |
2026-04-30 |
|
|
OS Copilot |
Adds Skills support for O&M capabilities. |
2026-04-30 |
March 2026
|
Feature |
Module |
Description |
Release date |
Documentation |
|
system health and diagnosis |
system diagnosis |
Memory diagnosis now supports analysis of Go program memory and system memory, and offers an improved display for Glibc memory. |
2026-03-30 |
|
|
Speeds up one-click diagnosis and adds support for CentOS versions earlier than 7.6. |
2026-03-30 |
|||
|
system health |
Adds dentry exception detection for memory resources. |
2026-03-30 |
||
|
component management |
component management |
Improves error code messages and reduces component installation time. |
2026-03-30 |
January 2026
|
Feature name |
Module |
Description |
Release date |
Documentation |
|
OS Copilot |
intelligent agent |
Adds diagnostics for high CPU load, automatically identifying and analyzing sudden spikes in CPU usage to quickly locate hot threads or methods. |
2026-01-30 |
|
|
system diagnosis |
panoramic memory analysis |
Adds output support for local memory allocation in Glibc, enabling a more comprehensive analysis of native memory usage. |
2026-01-30 |
|
|
system health and diagnosis |
system health |
Adds intelligent diagnostics for abnormal alerts, leveraging large model capabilities to provide context-aware root cause analysis and remediation suggestions. |
2026-01-30 |
December 2025
|
Feature name |
Feature module |
Feature overview |
Release date |
Related documentation |
|
GPU performance and diagnostics |
GPU resource topology |
Provides two unified views for monitoring a GPU cluster's resource usage and analyzing collective communication in distributed training.
|
2025-12-31 |
|
|
GPU diagnostics |
Adds a slow node diagnosis feature for GPUs. Tailored for large model training scenarios, this feature helps you quickly identify nodes that cause performance bottlenecks during training. |
2025-12-31 |
||
|
System diagnostics |
Memory analysis |
The memory analysis feature now supports Java memory diagnostics with large file uploads and unmanaged diagnostics. This update introduces on-heap/off-heap flame graphs and statistics on memory changes during profiling. |
2025-12-31 |
|
|
OS Copilot |
Intelligent diagnostics |
Introduces diagnostic support for the MCP service. Third-party agents can use this service to access the console's diagnostic capabilities and diagnose issues such as high memory usage, high CPU utilization, and performance jitter. |
2025-12-31 |
November 2025
|
Feature name |
Feature module |
Description |
Release date |
Related documentation |
|
component management |
configuration management |
Adds support for displaying regex match details in FastOOM. |
2025-11-30 |
|
|
Adds a feature to reclaim zombie Memcg. |
2025-11-30 |
|||
|
system diagnosis |
one-click diagnosis |
Scans for common OS issues with one click. |
2025-11-30 |
|
|
Memory Panorama |
Adds Java memory diagnosis to Memory Panorama. |
2025-11-30 |
||
|
OOM diagnosis |
SysOM enhances OOM data collection capabilities. |
2025-11-30 |
||
|
Historical Diagnosis |
Introduces Historical Diagnosis for analyzing past OOM events on ACS and ECI instances. |
2025-11-30 |
||
|
OS Copilot |
Intelligent Memory Diagnosis |
The intelligent assistant guides you to collect information for an issue, then automatically uses the Memory Panorama, Java memory diagnosis, and OOM diagnosis features to generate a diagnostic report with relevant links. |
2025-11-30 |
October 2025
|
Feature |
Module |
Description |
Release date |
Documentation |
|
component management |
component management |
Supports Rocky Linux 8.8 and later. |
2025-10-30 |
|
|
abnormal event push |
abnormal event alert |
Event content now includes k8s label information and GPU exception information. Downtime alerts now include diagnostic information. |
2025-10-30 |
|
|
AI performance diagnostics |
AI flame graph |
Introduces the AI flame graph for performance profiling of AI applications. |
2025-10-30 |
|
|
homepage overview |
system overview |
The console now includes a new overview page that displays the following:
|
2025-10-30 |
August 2025
|
Feature name |
Feature module |
Description |
Release date |
Related documentation |
|
Java memory diagnostics |
System diagnostics |
Adds memory analysis for Java applications. This feature scans the memory of a Java process and its pod, breaking down consumption from both the JVM's perspective and the process's actual memory footprint. |
2025-08-30 |
|
|
Pod diagnostics |
System diagnostics |
Pod diagnostics now supports a memory diagnostics mode for pods in ACK and ACS. |
2025-08-30 |
|
|
Process hotspot tracing |
System observability |
You can now query the history of process hotspots. |
2025-08-30 |
|
|
Abnormal event push |
Abnormal event alerting |
You can now receive push notifications for crash events. |
2025-08-30 |
July 2025
|
Feature |
Module |
Description |
Release date |
Related documentation |
|
FastOOM |
component management |
When enabled, this feature monitors node-level memory pressure and triggers FastOOM to terminate processes, preventing the system from becoming slow or unresponsive. |
2025-07-30 |
|
|
Pod diagnostics |
system diagnostics |
Diagnoses specific Pods in ACK clusters by namespace and name. |
2025-07-30 |
|
|
alert notifications |
system overview |
Notifies you through multiple channels about abnormal events, such as high CPU utilization, so you can respond and resolve issues quickly. |
2025-07-30 |
|
|
memory overview |
system diagnostics |
Adds detection for |
2025-07-30 |
|
|
crash diagnostics |
system diagnostics |
Adds support for analyzing hard lockups and hung tasks. |
2025-07-30 |
|
|
AI observability |
AI observability |
You can now export diagnostic reports for offline viewing. |
2025-07-30 |
|
|
Adds support for Ubuntu 22.04 and Ubuntu 24.04. |
June 2025
|
Feature name |
Feature module |
Overview |
Release date |
Related documentation |
|
health anomaly event |
system overview |
Added checks for |
2025-06-30 |
|
|
configuration management |
component management |
Enables the configuration of FastOOM. Using flexible configurations, FastOOM terminates relevant processes to prevent the system from becoming slow or unresponsive before an Out-of-Memory (OOM) event. |
2025-06-30 |
|
|
canary deployment |
component management |
Enables canary deployment of components to an ACK cluster based on node count, percentage, or ACK node labels. |
2025-06-30 |
|
|
unmanaged diagnosis |
system diagnosis |
Enables system diagnosis on unmanaged instances. |
2025-06-30 |
|
|
AI observation |
AI observation |
|
2025-06-30 |
March 2025
|
Feature name |
Feature module |
Description |
Release date |
Related documentation |
|
Node monitoring |
System management |
View network, I/O, scheduling, and memory monitoring information for nodes. |
2025-03-31 |
|
|
AI diagnosis |
AI diagnosis |
Diagnose system exceptions in AI workloads and generate diagnostic results, including GPU and AI job runtime status. |
2025-03-26 |
|
|
Configure Java hotspot tracing |
Process hotspot tracing |
Trace hotspots in Java processes and collect Java stack information. |
2025-03-12 |
December 2024
|
Parameter |
Feature module |
Description |
Release date |
Related documentation |
|
Cluster health |
System overview |
View the health status across multiple clusters and switch between them. |
2024-12-30 |
|
|
Abnormal event feedback |
Provide feedback on abnormal events. The system uses this feedback to adjust the reporting frequency and baseline. |
2024-12-30 |
||
|
One-click IO diagnosis |
System diagnosis |
Diagnoses common issues such as high IO latency, IO burst, and IO wait. |
2024-12-30 |
|
|
Network jitter diagnosis |
Analyzes network jitter caused by issues like slow packet reception, soft interrupts, and the scheduler, then provides root causes and solutions. |
2024-12-30 |
||
|
System load diagnosis |
Analyzes anomalies in the one-minute average system load (the |
2024-12-30 |
||
|
Downtime diagnosis |
Analyzes the causes of downtime and recommends solutions. |
2024-12-30 |
||
|
AI infra observability |
System observability |
Uses dynamic injection of GPU profiling to collect and analyze data from AI jobs—including operators, scheduling, and kernel functions. This provides non-intrusive, flexible, and low-overhead observability. |
2024-12-30 |
|
|
Cluster deployment |
Component management |
Component management now supports onboarding ACK clusters and enabling auto scaling for their instances. |
2024-12-30 |
|
|
Cancel subscription |
Subscription management |
You can now cancel subscriptions for CentOS security updates and Alinux 2 ELS through the console. |
2024-12-30 |
|
|
Guided diagnosis |
OS Copilot |
Provides guided diagnosis for anomalies such as OOM. |
2024-12-30 |
|
|
Document link citation |
OS Copilot answers can now include citations and links to documentation. |
2024-12-30 |