访问 Outline 服务器性能指标
使用集合让一切井井有条
根据您的偏好保存内容并对其进行分类。
Outline 通过 Prometheus 提供详细的性能指标，可助您深入了解有关服务器的使用情况和健康状况。本指南为您详细介绍了检索和查看这些指标的相关流程。
重要注意事项：本指南在编写时假定您对 Prometheus 和 PromQL 有基本的了解。如果您刚刚接触 Prometheus，不妨先查阅相关的文档和教程，然后再探索 Outline 的各项指标。
前提条件
操作说明
转发 Prometheus 端口
使用 SSH 连接到服务器，并转发 9090 端口：
ssh root@your_server_ip -L 9090:localhost:9090
访问 Prometheus 网页界面
打开网络浏览器，并访问 http://localhost:9090/graph
查询 Prometheus 指标
使用 PromQL 查询检索您感兴趣的指标。
PromQL 查询示例
使用情况
数据字节数（按访问密钥、协议和方向统计）：
increase(shadowsocks_data_bytes[1d])
数据字节数（按访问密钥汇总）：
sum(increase(shadowsocks_data_bytes[1d])) by (access_key)
数据字节数（用于数据限值计算）：
sum(increase(shadowsocks_data_bytes{dir=~"c<p|p>t"}[30d])) by (access_key)
数据字节数（按位置、协议和方向统计）：
increase(shadowsocks_data_bytes_per_location[1d])
有效的访问密钥
sum(max(max_over_time(shadowsocks_data_bytes{access_key!=""} [1h])) by (access_key) > bool 0)
TCP 连接
UDP
Prometheus：
prometheus_build_info
outline-ss-server：
shadowsocks_build_info
Node.js：
nodejs_version_info
如需查看可用指标的完整列表，请参阅 outline-ss-server 源代码。
如未另行说明，那么本页面中的内容已根据知识共享署名 4.0 许可获得了许可，并且代码示例已根据 Apache 2.0 许可获得了许可。有关详情，请参阅 Google 开发者网站政策。Java 是 Oracle 和/或其关联公司的注册商标。
最后更新时间 (UTC)：2025-07-25。
[null,null,["最后更新时间 (UTC)：2025-07-25。"],[[["\u003cp\u003eOutline provides detailed performance metrics via Prometheus, offering insights into server usage and health.\u003c/p\u003e\n"],["\u003cp\u003eAccessing these metrics requires an Outline server with Prometheus enabled and SSH access to forward port 9090.\u003c/p\u003e\n"],["\u003cp\u003eYou can use PromQL queries to retrieve specific metrics, such as data bytes, active access keys, TCP connections, and UDP packets.\u003c/p\u003e\n"],["\u003cp\u003ePerformance metrics are also available for CPU usage and memory by process.\u003c/p\u003e\n"],["\u003cp\u003eBuild information for Prometheus, outline-ss-server, and Node.js are available as well, providing software versioning.\u003c/p\u003e\n"]]],["Performance metrics are accessed via Prometheus, requiring an Outline server with Prometheus enabled and SSH access. Port 9090 is forwarded via SSH to view metrics in a web browser at `http://localhost:9090/graph`. PromQL queries retrieve metrics like data bytes, active access keys, TCP connections, UDP packets/associations, and CPU/memory usage. Build information for Prometheus, `outline-ss-server`, and Node.js are also available. The source code provides a complete metrics list.\n"],null,["# Accessing Outline Server Performance Metrics\n\nOutline provides detailed performance metrics through\n[Prometheus](https://prometheus.io/), allowing you to gain deeper insights into\nyour server's usage and health. This guide will walk you through the process of\nretrieving and viewing these metrics.\n\n**Important Note:** This guide assumes you have a basic understanding of\nPrometheus and PromQL. If you're new to Prometheus, consider exploring its\ndocumentation and tutorials before diving into Outline's metrics.\n\nPrerequisites\n-------------\n\n- **Outline server with Prometheus enabled**: Ensure that your Outline server\n has Prometheus metrics enabled. (This is usually the default configuration).\n\n- **SSH access to your server**: You'll need SSH access to forward the\n Prometheus port.\n\nInstructions\n------------\n\n1. **Forward Prometheus Port**\n\n Connect to your server using SSH and forward port 9090: \n\n ssh root@your_server_ip -L 9090:localhost:9090\n\n2. **Access Prometheus Web Interface**\n\n Open your web browser and navigate to: \u003chttp://localhost:9090/graph\u003e\n Query Prometheus Metrics\n3. **Use PromQL queries to retrieve the specific metrics you're interested in.**\n\n### Example PromQL Queries\n\n#### Usage\n\n- **Data Bytes (by access key, protocol, and direction):**\n\n `increase(shadowsocks_data_bytes[1d])`\n- **Data Bytes (aggregated by access key):**\n\n `sum(increase(shadowsocks_data_bytes[1d])) by (access_key)`\n- **Data Bytes (for data limits calculation):**\n\n `sum(increase(shadowsocks_data_bytes{dir=~\"c\u003cp|p\u003et\"}[30d])) by (access_key)`\n- **Data Bytes (by location, protocol, and direction):**\n\n `increase(shadowsocks_data_bytes_per_location[1d])`\n\n#### Active Access Keys\n\n`sum(max(max_over_time(shadowsocks_data_bytes{access_key!=\"\"} [1h])) by (access_key) \u003e bool 0)`\n\n#### TCP Connections\n\n- **TCP Connections (by access key, location, and status):**\n\n `increase(shadowsocks_tcp_connections_closed[1d])`\n- **TCP Connections (by location):**\n\n `increase(shadowsocks_tcp_connections_opened[1d])`\n\n#### UDP\n\n- **UDP Packets (by location and status):**\n\n `increase(shadowsocks_udp_packets_from_client_per_location[1d])`\n- **UDP Associations (no breakdown):**\n\n `increase(shadowsocks_udp_nat_entries_added[1d])`\n\n#### Performance\n\n- **CPU Usage (by process):**\n\n `rate(process_cpu_seconds_total[10m])`\n- **Memory (by process):**\n\n `process_virtual_memory_bytes`\n\n#### Build Information\n\n- **Prometheus:**\n\n `prometheus_build_info`\n- **outline-ss-server:**\n\n `shadowsocks_build_info`\n- **Node.js:**\n\n `nodejs_version_info`\n\nThe complete list of available metrics can be found in the `outline-ss-server`\n[source code](https://github.com/Jigsaw-Code/outline-ss-server/blob/master/cmd/outline-ss-server/metrics.go)."]]
