Back to Details

acc-optimize-docker-php-fpm

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

PHP-FPM Configuration Optimization

PHP-FPM配置优化

Provides production-ready PHP-FPM tuning for Docker containers based on workload type and available resources.
本文提供适用于Docker容器的生产级PHP-FPM调优方案，基于工作负载类型和可用资源进行配置。

Process Manager Modes

进程管理器模式

┌─────────────────────────────────────────────────────────────────┐
│                  PHP-FPM PROCESS MANAGERS                        │
├─────────────────────────────────────────────────────────────────┤
│  STATIC:    [W][W][W][W][W][W][W][W]   Fixed pool              │
│             Best for: dedicated containers, predictable load    │
│  DYNAMIC:   [W][W][W][W][ ][ ][ ][ ]   Scales up/down         │
│             Best for: variable load, shared resources           │
│  ONDEMAND:  [ ][ ][ ][ ][ ][ ][ ][ ]   Starts on request      │
│             Best for: low-traffic, dev environments             │
└─────────────────────────────────────────────────────────────────┘
┌─────────────────────────────────────────────────────────────────┐
│                  PHP-FPM PROCESS MANAGERS                        │
├─────────────────────────────────────────────────────────────────┤
│  STATIC:    [W][W][W][W][W][W][W][W]   Fixed pool              │
│             Best for: dedicated containers, predictable load    │
│  DYNAMIC:   [W][W][W][W][ ][ ][ ][ ]   Scales up/down         │
│             Best for: variable load, shared resources           │
│  ONDEMAND:  [ ][ ][ ][ ][ ][ ][ ][ ]   Starts on request      │
│             Best for: low-traffic, dev environments             │
└─────────────────────────────────────────────────────────────────┘
静态模式：固定进程池，最适合专用容器、可预测负载的场景 动态模式：进程数可上下扩展，最适合负载多变、资源共享的场景 按需模式：有请求时才启动进程，最适合低流量、开发环境

Memory Calculation Formula

内存计算公式

max_children = (available_memory - system_overhead) / avg_worker_memory

Where:
  available_memory = container_memory_limit
  system_overhead  = ~50MB (OS + PHP-FPM master + nginx)
  avg_worker_memory = 30-60MB (depends on application)
Container MemorySystem OverheadWorker Memorymax_children
256MB50MB40MB5
512MB50MB40MB11
1024MB50MB40MB24
2048MB50MB50MB39
Measure actual worker memory:
bash
ps aux | grep "php-fpm: pool" | awk '{sum+=$6; n++} END {print sum/n/1024 " MB"}'
max_children = (available_memory - system_overhead) / avg_worker_memory

Where:
  available_memory = container_memory_limit
  system_overhead  = ~50MB (OS + PHP-FPM master + nginx)
  avg_worker_memory = 30-60MB (depends on application)
其中： available_memory = 容器内存限制 system_overhead = 约50MB（操作系统 + PHP-FPM主进程 + nginx） avg_worker_memory = 30-60MB（取决于应用程序）
容器内存系统开销工作进程内存max_children
256MB50MB40MB5
512MB50MB40MB11
1024MB50MB40MB24
2048MB50MB50MB39

Configuration by Workload

测量实际工作进程内存

API Workload (Many Fast Requests)

—
ini
[www]
pm = static
pm.max_children = 20
request_terminate_timeout = 30s
request_slowlog_timeout = 5s
pm.max_requests = 1000
slowlog = /proc/self/fd/2
bash
ps aux | grep "php-fpm: pool" | awk '{sum+=$6; n++} END {print sum/n/1024 " MB"}'

Worker Workload (Few Long Requests)

按工作负载类型配置

—

API工作负载（大量快速请求）

ini
[www]
pm = dynamic
pm.max_children = 8
pm.start_servers = 2
pm.min_spare_servers = 1
pm.max_spare_servers = 4
request_terminate_timeout = 300s
request_slowlog_timeout = 30s
pm.max_requests = 200
ini
[www]
pm = static
pm.max_children = 20
request_terminate_timeout = 30s
request_slowlog_timeout = 5s
pm.max_requests = 1000
slowlog = /proc/self/fd/2

Mixed Workload

工作进程负载（少量长耗时请求）

ini
[www]
pm = dynamic
pm.max_children = 15
pm.start_servers = 5
pm.min_spare_servers = 3
pm.max_spare_servers = 8
request_terminate_timeout = 60s
request_slowlog_timeout = 10s
pm.max_requests = 500
ini
[www]
pm = dynamic
pm.max_children = 8
pm.start_servers = 2
pm.min_spare_servers = 1
pm.max_spare_servers = 4
request_terminate_timeout = 300s
request_slowlog_timeout = 30s
pm.max_requests = 200

Dynamic Mode Formulas

混合工作负载

pm.start_servers     = (min_spare + max_spare) / 2
pm.min_spare_servers = max_children * 0.25
pm.max_spare_servers = max_children * 0.75
ini
[www]
pm = dynamic
pm.max_children = 15
pm.start_servers = 5
pm.min_spare_servers = 3
pm.max_spare_servers = 8
request_terminate_timeout = 60s
request_slowlog_timeout = 10s
pm.max_requests = 500

pm.max_requests by Application Type

动态模式计算公式

Application TypeRecommended Value
Stateless API1000-5000
Framework with ORM500-1000
Legacy application100-500
Memory-intensive processing50-200
pm.start_servers     = (min_spare + max_spare) / 2
pm.min_spare_servers = max_children * 0.25
pm.max_spare_servers = max_children * 0.75

Health Check and Monitoring

按应用程序类型设置pm.max_requests

ini
pm.status_path = /status
ping.path = /ping
ping.response = pong
dockerfile
HEALTHCHECK --interval=30s --timeout=5s --start-period=10s --retries=3 \
    CMD php-fpm-healthcheck || exit 1
Key status metrics to monitor:
  • active processes: current load
  • listen queue: requests waiting (should be 0)
  • max children reached: need more workers
  • slow requests: performance issues
应用程序类型推荐值
无状态API1000-5000
带ORM的框架500-1000
遗留应用程序100-500
内存密集型处理50-200

Docker-Specific Settings

健康检查与监控

ini
; Logging to stdout/stderr
error_log = /proc/self/fd/2
access.log = /proc/self/fd/2
clear_env = yes
catch_workers_output = yes
decorate_workers_output = no
ini
pm.status_path = /status
ping.path = /ping
ping.response = pong
dockerfile
HEALTHCHECK --interval=30s --timeout=5s --start-period=10s --retries=3 \
    CMD php-fpm-healthcheck || exit 1
需要监控的关键状态指标：
  • active processes（活跃进程数）： 当前负载
  • listen queue（监听队列数）： 等待处理的请求（应保持为0）
  • max children reached（达到最大子进程数）： 需要增加工作进程数
  • slow requests（慢请求）： 性能问题预警

Before/After Comparison

Docker专属配置

MetricDefaultOptimized (static)Improvement
max_children520+300% capacity
Idle memory (512MB)200MB wasted~50MB overhead-75% waste
Request latency (p99)500ms150ms-70%
Throughput (req/s)50200+300%
Memory leak recoveryNeverEvery 1000 reqPredictable
ini
; Logging to stdout/stderr
error_log = /proc/self/fd/2
access.log = /proc/self/fd/2
clear_env = yes
catch_workers_output = yes
decorate_workers_output = no

Complete Production Configuration

优化前后对比

ini
[global]
error_log = /proc/self/fd/2
log_level = warning
daemonize = no

[www]
user = www-data
group = www-data
listen = 0.0.0.0:9000
pm = static
pm.max_children = 20
pm.max_requests = 1000
request_terminate_timeout = 30s
request_slowlog_timeout = 5s
slowlog = /proc/self/fd/2
pm.status_path = /status
ping.path = /ping
ping.response = pong
clear_env = yes
catch_workers_output = yes
decorate_workers_output = no
access.log = /proc/self/fd/2
access.format = "%R - %u %t \"%m %r\" %s %{mili}dms %{mega}MMB %C%%"
指标默认配置优化后（静态模式）提升效果
max_children520+300% 处理能力
空闲内存（512MB容器）浪费200MB约50MB开销减少75%内存浪费
请求延迟（p99）500ms150ms降低70%
吞吐量（请求/秒）50200提升300%
内存泄漏恢复从不每1000次请求可预测的内存回收

Generation Instructions

完整生产环境配置

  1. Determine workload type: API, Worker, or Mixed
  2. Calculate max_children: Based on container memory and worker memory
  3. Select PM mode: Static for production, dynamic for variable load
  4. Configure timeouts: Based on expected request duration
  5. Set max_requests: Based on application memory behavior
  6. Enable monitoring: Status page, health check, slow log
ini
[global]
error_log = /proc/self/fd/2
log_level = warning
daemonize = no

[www]
user = www-data
group = www-data
listen = 0.0.0.0:9000
pm = static
pm.max_children = 20
pm.max_requests = 1000
request_terminate_timeout = 30s
request_slowlog_timeout = 5s
slowlog = /proc/self/fd/2
pm.status_path = /status
ping.path = /ping
ping.response = pong
clear_env = yes
catch_workers_output = yes
decorate_workers_output = no
access.log = /proc/self/fd/2
access.format = "%R - %u %t \"%m %r\" %s %{mili}dms %{mega}MMB %C%%"
—

配置生成步骤

—
  1. 确定工作负载类型： API、工作进程或混合类型
  2. 计算max_children： 基于容器内存和工作进程内存
  3. 选择进程管理器模式： 生产环境推荐静态模式，负载多变场景用动态模式
  4. 配置超时时间： 基于预期的请求时长
  5. 设置max_requests： 基于应用程序的内存使用行为
  6. 启用监控： 状态页面、健康检查、慢日志