System load diagnosis helps you analyze the cause of an abnormal one-minute average load (the load1 metric), providing detailed information and suggesting solutions.
Limitations
Region restrictions
This feature is available only in mainland China and the China (Hong Kong) region.
Operating system restrictions
Architecture
Operating system
x86 architecture
Rocky Linux 9.5
Rocky Linux 9.1
Ubuntu 20.04
Alibaba Cloud Linux 3 Container Optimized
Rocky Linux 8.8
Ubuntu 22.04
Alibaba Cloud Linux 3 Pro
Alibaba Cloud Linux 2/3
CentOS 7.6 or later, or CentOS 8
Anolis OS 7/8
Ubuntu 24.04
ARM architecture
Alibaba Cloud Linux 3 Pro
Alibaba Cloud Linux 3
Use cases
This topic describes common scenarios where you can use this feature to run diagnostics and resolve issues based on the provided suggestions.
Your monitoring system indicates that the system load metric is high.
Your system experiences short-term anomalies related to network or I/O operations.
Prerequisites
If you use a RAM user, ensure the primary Alibaba Cloud account has granted the required system policies,
AliyunECSReadOnlyAccessandAliyunSysomFullAccess, to the RAM user.Activate console access.
When you sign in to the Operating System Console for the first time, click Activate Service.
Procedure
Go to the Operating System Console.
In the left-side navigation pane, click System Diagnosis.
In the upper-left corner, select your target instance's region.
In the Diagnosis Type list, select Scheduling Diagnosis. In the Diagnosis Item list, select System Load Diagnosis. Select an Instance ID and click Run Diagnosis.
NoteIf SysOM is not installed, clicking Run Diagnosis automatically installs SysOM on the target ECS instance and starts the diagnosis.
In the Diagnosis Records section, click View Report.
Diagnostic report
Basic information
The Basic information section contains details about the diagnosis, including the instance ID (resource ID), diagnosis item, report ID, and start time.
Diagnosis conclusion
If a load anomaly is found, this section summarizes the results and provides the detailed root cause.
Suggestions
This section provides recommended actions based on the diagnostic findings.
Diagnosis details
If a load anomaly is found, this section provides detailed information about the anomaly.
Example
This example demonstrates how to use the system load diagnosis feature to troubleshoot a load anomaly caused by an improper core binding configuration.
Symptoms
The container monitoring system detects a high system load.
Diagnostic analysis
To diagnose the target ECS instance in this scenario, use the system load diagnosis feature in the Operating System Console. The following figures show the results.
Conclusion and suggestions
Diagnosis conclusion: The primary cause of the high system load is pressure on CPU resources.
Suggestions: The analysis indicates that an improper core binding configuration is causing an uneven CPU load, leading to task accumulation on a few CPUs.
Diagnosis details
This section provides more information about the system load, as shown in the following figure.
Event overview: This section displays the overall system load value and provides an analysis of kernel-side pressure from system calls, hard interrupts, soft interrupts, and I/O.
Load details: This section shows the distribution of processes in runnable (R) and uninterruptible sleep (D) states. In this example, it reveals a large number of
stressprocesses in the R state.
Solution
Add idle CPUs to the core binding scope or remove the core binding configuration to distribute the load across more CPUs.