All Products
Search
Document Center

API Gateway:API observability

Last Updated:Jan 17, 2025

This topic describes the observability features of API Gateway. These features include API call logs, API monitoring and alerting, and Tracing Analysis. These features ensure that you can fully learn and monitor the running of APIs and improve your system stability and performance.

API call logs

API call logs of the API Gateway.

  • Use Simple Log Service to manage logs of API calls.

    API Gateway seamlessly integrates with Simple Log Service for various features. For example, you can query logs, download logs, and perform multi-dimensional statistical analysis of logs in real time. Specifically:

    • Online log query: You can use keywords to query logs. Both exact match and fuzzy match are supported. Log query can be used for troubleshooting or statistical query.

    • Detailed log query: You can use the following fields to query detailed call logs.

    • Call log viewing: You can view API call logs.

  • Configure the logging of HTTP requests and responses

    You can configure to record the HTTP requests received by API Gateway and the HTTP responses returned to clients.

    Note

    This feature is available for dedicated instances only.

For more information, see API call logs.

Monitoring and alerting

The monitoring and alerting features of the original API Gateway. Specifically:

  • View monitoring data

    You can view API Gateway monitoring data by region, API group, and API. Specifically:

    • Region monitoring.

    • Instance monitoring (dedicated instances only).

    • API group monitoring.

    • API monitoring.

  • Configure alerting for APIs.

    You can use CloudMonitor to configure alerting for APIs that are published in API Gateway. This way, you can track the running status of APIs at all times to ensure the stability of the APIs.

  • Configure an alert rule for an outbound connection pool.

    Simple Log Service allows you to specify a check frequency and trigger conditions to configure an alert rule, and triggers alerts when the conditions are met.

    If a connection pool is fully occupied, API Gateway returns the Backend service connect failed 'Connection lease request time out' error to the client for excess requests. This error is recorded by the errorMessage field in logs. You can use this error message to configure an alert rule. For more information, see Configure an alert rule in Simple Log Service.

For more information, see Monitoring and alerting.

Configure tracing analysis

Important

This feature is available for only dedicated instances.

You can configure Tracing Analysis in the API Gateway console to upload logs to the Alibaba Cloud Tracing Analysis platform. Tracing Analysis provides a set of tools to help you develop applications and troubleshoot issues. These tools include trace mapping, request statistics, trace topology, and application dependency analysis. Specifically:

  • How to authorize Tracing Analysis.

  • How to configure tracing analysis in the API Gateway console.

For more information, see Configure tracing analysis.