You can monitor metrics for your elastic IP addresses (EIPs), such as bandwidth and packet rate. You can also configure threshold-based alerts or subscribe to anomaly events. This helps you understand EIP usage, detect issues promptly, and take quick action.
CloudMonitor alerts
Cloud Monitor (CMS) collects real-time monitoring metrics for elastic IP addresses (EIPs) and supports the following features:
View monitoring charts to understand the operational status of your EIPs.
Create threshold-based alert rules to receive notifications about abnormal monitoring data and take prompt action.
If a Resource Access Management (RAM) user needs to view EIP monitoring data, an Alibaba Cloud account must first grant the required permissions.
Option 1: Grant the RAM user a system policy that is associated with
cms:QueryMetricList.Available system policies: AliyunCloudMonitorFullAccess, AliyunCloudMonitorMetricDataReadOnlyAccess, AliyunCloudMonitorReadOnlyAccess, and AliyunCloudMonitorReadOnlyBasicAccess.
Option 2: Create or modify a custom policy. Add the
cms:QueryMetricListoperation to the `Allow` statement. Then, grant the custom policy to the RAM user.
EIP monitoring metrics
Monitoring metrics | Description |
Outbound bandwidth | The bandwidth that is consumed when the EIP instance accesses external services. Unit: bps. |
Inbound bandwidth | The bandwidth that is consumed when external services access the EIP instance. Unit: bps. |
Inbound bandwidth utilization | The percentage of an EIP instance's total bandwidth used for external access. |
Outbound bandwidth utilization | The percentage of total bandwidth that is consumed by outbound traffic from the EIP instance. |
Outbound packet rate | The number of data packets that are sent from the EIP instance per second. Unit: pps. |
Inbound packet rate | The number of data packets that are received by the EIP instance per second. Unit: pps. |
Outbound packet drop rate due to throttling | The rate at which outbound data packets from the EIP instance are dropped due to bandwidth throttling. Unit: pps. |
Inbound packet drop rate due to throttling | The rate at which inbound data packets to the EIP instance are dropped due to bandwidth throttling. Unit: pps. |
View EIP monitoring data
Console
Go to the Elastic IP Address console, click the ID of the target EIP, and then view the monitoring charts on the tab.
By default, the system displays monitoring data from the last hour. You can view data from the last 3 hours, 6 hours, or 12 hours, or from a custom period. You can query data from the last 31 days.
You can enable Auto-refresh to update the monitoring chart with the latest data every minute.
Go to the Cloud Service Monitoring - EIP page. For the target EIP, click Monitoring Charts in the Actions column to view the monitoring data.
By default, the system displays monitoring data from the last hour. You can view data from the last 3 hours, 6 hours, or 12 hours, or from 1 day, 3 days, 7 days, or 14 days, or from a custom period. You can query data from the last 31 days.
API
You can call the DescribeMetricList operation to query monitoring data for a specific metric of a cloud service.
Create a threshold-based alert rule
If an EIP is released, its associated threshold-based alert rules in CloudMonitor are also deleted.
Console
Go to the Cloud Monitor - Alert Rules page and click Create Alert Rule.
Product: Select Elastic IP Address.
Resource Scope: The resource scope cannot be changed after the alert rule is created. You can select All Resources to apply the rule to all instances of the selected Product in your account. You can also select Instance and configure the Associated Instance to apply the rule to specific instances.
Set rule conditions: Define the specific conditions under which monitoring data triggers an alert.
Click Add Rule and select the appropriate metric type from the drop-down list.
In the Set Rule Description panel, configure the Rule Name, Metric Type, and Metric.
Mute For: If an alert is not resolved, this is the interval at which alert notifications are resent.
Effective Period: The alert rule checks monitoring data only during its effective period.
Alert Contact Group: The contact group to which alerts are sent.
API
You can call the PutResourceMetricRules operation to set multiple threshold-based alert rules for a specific metric of a resource.
Fine-grained monitoring
If public bandwidth utilization reaches 100% and the EIP is not scaled out in time, service traffic may be rate-limited, which can cause packet loss and degrade the user experience. The fine-grained monitoring feature of EIP provides the following benefits:
It increases monitoring granularity to the second level. This lets you monitor traffic fluctuations in real time and helps you adjust the EIP bandwidth peak promptly to avoid access delays that are caused by bandwidth throttling.
It uses Alibaba Cloud Simple Log Service to provide comprehensive features such as data storage, reports, and alerting.
After you enable fine-grained monitoring, the system automatically adjusts the data storage policy of the associated Logstore. The default retention period is set to 7 days.
Each account receives a free quota of 500 MB per month. Usage beyond this quota is charged.
When an EIP is attached to an ECS instance or an elastic network interface (ENI), bandwidth rate limiting is applied at the ECS network interface controller (NIC) level. The corresponding rate-limiting drop metrics are not integrated with EIP fine-grained monitoring. Therefore, you can view rate-limiting drop data only for EIPs that are attached to NAT Gateways or Server Load Balancers, not for EIPs attached to ECS instances or ENIs.
Monitoring data
Peak outbound and inbound bandwidth per second: The peak outbound and inbound bandwidth of an EIP per second, measured in bps.
Outbound and inbound packet rate: The number of data packets per second in the outbound and inbound directions of an EIP. The unit is pps.
Outbound and inbound packet loss rate per second: The number of packets lost per second in the outbound and inbound directions for an EIP. The unit is pps.
Outbound and inbound new TCP connection rate: The rate of new TCP connections in the outbound and inbound directions for the EIP, measured in pps.
Enable fine-grained monitoring
Console
Go to the Virtual Private Cloud console > Internet Toolbox page and click Fine-grained Monitoring.
If the required permissions for fine-grained monitoring are not granted, follow the on-screen instructions to complete the resource access authorization.
If you use a RAM user to enable fine-grained monitoring, an Alibaba Cloud account must first grant the required resource access authorization.
In the Actions column for the target EIP, click Enable Fine-grained Monitoring. Select an existing log project and Logstore or create new ones to complete the log settings.
After you enable this feature, click the
icon in the Monitor column of the target EIP to view and analyze monitoring data.
API
You can call the SetHighDefinitionMonitorLogStatus operation to enable or disable fine-grained monitoring for an EIP.
Anomalies
You can integrate EIP with the Event Hub of Network Intelligence Service (NIS) to view abnormal events for the EIP from the last 30 days. This lets you identify potentially affected resources and make timely business adjustments.
After an EIP is attached to a cloud resource, you can view the anomalous activities of the EIP on the tab of the EIP details page.
You can view all anomalies that occurred within the last 30 days.
After you view an event, you can resolve it based on the provided suggestions.
You can subscribe to events to receive alert notifications for important events and make timely business adjustments.