Service diagnostics checks your Kubernetes Service configuration against its associated Classic Load Balancer (CLB) instance and flags mismatches, quota violations, and error events — so you can pinpoint and fix the root cause without manually inspecting each resource.
When you run Service diagnostics, ACS deploys a data collection program on every node in the cluster. The program collects system version, load metrics, container component status, and key error messages from system logs. It does not collect business data or sensitive information.
Diagnostic items
Diagnostic items vary based on cluster configuration. The items shown on the diagnostics page are authoritative.
| Category | What it checks |
|---|---|
| Service | Consistency between Service configurations and CLB instance configurations |
| ResourceQuotas | CLB quota limits that could block pod or Service updates |
Service
| Diagnostic item | Description | How to fix |
|---|---|---|
| LBInstanceChargeType | Checks whether the CLB instance billing method matches the billing method specified in the Service. | Update the Service annotation to match the billing method of the CLB instance. See Use annotations to configure CLB instances. |
| LBCertId | Checks whether the certificate on the CLB instance matches the certificate specified in the Service. | Update the certificate ID annotation on the Service to match the certificate used by the HTTPS listeners on the CLB instance. See Use annotations to configure CLB instances. |
| LBReusable | Checks whether the CLB instance meets the requirements for reuse. | Select a CLB instance that satisfies the reuse requirements. See Use an existing SLB instance to expose an application. |
| ServiceWarningEvents | Checks whether the Service has generated warning events. | Service error events contain the [%s] content. For more information about how to troubleshoot Service errors, see Service errors and solutions. |
| LBExisted | Checks whether the CLB instance associated with the Service exists. | If the CLB instance no longer exists, re-run diagnostics on the Service and follow the resolution steps. See Service errors and solutions. |
| ReadyPodNum | Checks the number of backend pods in the Ready state for the Service. | This diagnostic item checks whether the backend pods of a Service are in the Ready state. |
ResourceQuotas
When a CLB quota limit is reached, pod updates or Service updates can no longer sync to the CLB instance. Check your current usage on the Quota Management page.
| Diagnostic item | What it checks | Quota limit |
|---|---|---|
| SlbQuotaBackendserverAttachedNum | Number of CLB instances associated with an Elastic Compute Service (ECS) instance. | 50 backend server groups per ECS instance (default). After this limit is reached, pod updates cannot sync to CLB backend servers. |
| SlbQuotaBackendserversNum | Number of backend servers in a CLB instance. | 200 backend servers per CLB instance (default). After this limit is reached, pod updates cannot sync to the CLB instance. |
| SlbQuotaInstancesNum | Number of CLB instances under an Alibaba Cloud account. | 60 CLB instances per account (default). No new CLB instances can be created after this limit is reached. |
| SlbQuotaListenersNum | Number of listeners configured on a CLB instance. | 50 listeners per CLB instance (default). After this limit is reached, Service updates cannot sync to the CLB instance. |