API Gateway lets you view monitoring charts for regions, dedicated instances, API groups, and individual APIs — covering traffic, latency, HTTP status codes, and request counts.
Overview
API Gateway provides four monitoring levels: region, instance, group, and API. Each level lets you filter by deployment environment (Production, Pre, or Test) and view the last seven days of call data.
1. Region monitoring
Log on to the API Gateway console.
In the left navigation pane, click Overview. Each region card shows the number of API groups, APIs, dedicated instances in use, and instances expiring within 10 days. Click the monitoring icon on a region card to open its monitoring charts.
-
Region monitoring covers the last seven days. Filter by environment (Production, Pre, or Test) to view the following metrics:
API request count: total number of API calls in the selected region.
API request count by instance: request volume broken down per dedicated instance.
Traffic: request and response data volume.
Average latency: end-to-end processing time, split between API Gateway and backend.
HTTP status code distribution: breakdown of response codes across all calls.
2. Instance monitoring
Log on to the API Gateway console.
In the left navigation pane, choose . On the Dedicated Instances page, click Monitoring to open the instance monitoring page.
-
Instance monitoring covers the last seven days. Filter by environment (Production, Pre, or Test) to view the following metrics:
Instance request count: total API calls handled by this instance.
Instance traffic: request and response data volume for this instance.
Average latency: end-to-end processing time, split between API Gateway and backend.
HTTP status code distribution: breakdown of response codes for calls through this instance.
-
Concurrent connections:
Active connections: the number of connections in the ESTABLISHED state per second. With persistent connections, multiple file requests can be transferred over a connection at the same time.
Inactive connections: the number of connections not in the ESTABLISHED state per second.
Maximum concurrent connections: the total number of established connections.
-
Data packets:
Inbound packets: the number of TCP packets received by the instance per second.
Outbound packets: the number of TCP packets sent by the instance per second.
-
Dropped data packets:
Inbound packets discarded per second: the number of inbound TCP packets dropped per second.
Outbound packets discarded per second: the number of outbound TCP packets dropped per second.
New connections per second (CPS): rate of new connections established on this instance.
Dropped connections per second: rate of connections dropped by this instance.
Concurrent connections, data packets, dropped data packets, and new connections metrics include only Internet-transferred data.
3. Group monitoring
Log on to the API Gateway console.
In the left navigation pane, choose . On the Group List page, click the Monitoring icon for the group you want to inspect.
-
Group monitoring covers the last seven days. Filter by environment (Production, Pre, or Test) to view the following metrics:
API request count: total number of API calls in this group.
Traffic: request and response data volume for this group.
Average latency: end-to-end processing time, split between API Gateway and backend.
HTTP status code distribution: breakdown of response codes for calls in this group.
4. API monitoring
Log on to the API Gateway console.
In the left navigation pane, choose . Click the API you want to monitor to open its details page.
-
In the left navigation pane, click Monitoring Information. API monitoring covers the last seven days. Filter by environment (Production, Test, or Pre) to view the following metrics:
Number of requests: successful and abnormal call counts for this API.
Traffic: uplink and downlink data volume for this API.
Backend service response time: time taken by the backend to process and respond to requests.
Error distribution: client-side and server-side error breakdown for this API.
API Gateway monitoring covers the last seven days. To view data beyond this range or analyze call logs in detail, use Simple Log Service. For more information, see View API call logs using Simple Log Service.