DocsICP FilingConsole
官方文档
首页

Diagnose an NLB instance

更新时间：
复制 MD 格式
产品详情
我的收藏

If your Network Load Balancer (NLB) instance has a problem, use the instance diagnostics feature to troubleshoot it. This feature checks your NLB instance for issues related to health checks, idle instances, security policies, costs, and listeners. It also identifies the causes of detected issues and suggests solutions to help you resolve them quickly.

Prerequisites

The first time you run diagnostics, the system prompts you to enable Network Intelligence Service (NIS) and automatically creates a service-linked role named AliyunServiceRoleForNis. For more information about AliyunServiceRoleForNis, see service-linked role.

Run an instance diagnosis

  1. Log on to the Network Load Balancer console.

  2. In the top navigation bar, select the region where the instance is deployed.

  3. On the Instances page, find the instance and click Diagnose in the Instance Diagnostics column.

  4. In the Instance Diagnostics panel, view the diagnosis progress, a summary of the results, and the details of each diagnostic item.

    • Anomalous items are displayed directly in the panel. You can view the details of each anomalous item.

    • In the Diagnostic Details section, select Show All Diagnostic Items to view all diagnostic items that NLB supports, along with their details.

    • Alternatively, click View Historical Diagnostics in NIS Console at the top of the Instance Diagnostics panel to open the NIS console and view detailed diagnostic history.

  5. After you review the diagnostic information, click Close.

Diagnostic items

Category

Diagnostic item and description

Health Check Diagnostics

  • Health Check Configuration: Checks whether health checks are enabled for the listeners of the NLB instance.

  • Health Check Errors: Checks whether any health check errors occurred for the listeners of the NLB instance.

Idle Instance Diagnostics

  • NLB Idle Status: Checks whether a listener is configured for the NLB instance.

  • Idle Server: Checks whether a backend server is configured for the NLB instance.

Security Policy Diagnostics

  • Interception by Cloud Firewall: Checks whether network traffic related to the EIP associated with the NLB instance is blocked by Cloud Firewall.

  • Penalty for Security Control: Checks whether network traffic related to the EIP associated with the NLB instance is penalized by Alibaba Cloud Security.

  • Anti-DDoS Protection Status: Checks whether the EIP associated with the NLB instance is protected by Anti-DDoS Protection.

  • Cloud Security Interception: Checks whether network traffic related to the EIP associated with the NLB instance is blocked by Alibaba Cloud Security.

Cost Diagnostics

  • Overdue Payments: Checks whether the NLB instance has overdue payments.

  • Alerts for Expiration: Checks whether an alert for overdue payments has been triggered for the NLB instance.

Listener Diagnostics

  • Listener Status: Checks whether any listeners of the NLB instance are in the stopped state.

FAQ

  • Can instance diagnostics check the historical state of an instance?

    No.

    Instance diagnostics performs real-time checks and cannot diagnose the historical state of an instance. The diagnostic data covers only the 15-minute period before the diagnosis is run.

    For example, if an instance becomes unavailable at 09:00:00 due to an issue and recovers at 09:30:00, a diagnosis run at 10:00:00 checks only the instance state from 09:45:00 to 10:00:00. It will not detect the root cause of the issue that occurred between 09:00:00 and 09:30:00.

  • My listener has a health check configured, so why does the diagnosis report an abnormal status? Can instance diagnostics identify the specific cause of the failure?

    Instance diagnostics checks the health of all listeners for an instance. The diagnosis reports an issue if a health check is not configured or if the health check status is abnormal.

    Many factors can cause health check failures. Instance diagnostics provides a Further Diagnosis feature to help identify the root cause. For example, if a service is not running on the listening port of a backend server, or network filtering rules such as iptables are configured on the backend server's operating system, the Further Diagnosis feature can help pinpoint the specific issue.

  • Why is Further Diagnosis unavailable for some backend servers?

    Currently, the Further Diagnosis feature supports only backend servers that run CentOS, Ubuntu, or Alibaba Cloud Linux. It does not support backend servers that run other operating systems, such as Windows.

Related documents

ALB and CLB also support instance diagnostics. For more information, see Diagnose an ALB instance and Diagnose a CLB instance.

Previous：Modify an NLB instanceNext：NLB listeners