13.1 性能优化概述
性能问题识别
常见性能瓶颈：
Jenkins性能问题分类：
1. 系统资源瓶颈
- CPU使用率过高
- 内存不足或泄漏
- 磁盘I/O瓶颈
- 网络带宽限制
2. 应用层面问题
- JVM配置不当
- 垃圾回收频繁
- 线程池配置问题
- 数据库连接池不足
3. 架构设计问题
- 单点瓶颈
- 负载分布不均
- 缓存策略不当
- 同步操作过多
4. 配置和使用问题
- 插件冲突或性能差
- 构建配置不合理
- 并发设置不当
- 日志级别过详细
性能监控指标：
关键性能指标(KPI)：
1. 响应时间指标
- 页面加载时间
- API响应时间
- 构建启动延迟
- 队列等待时间
2. 吞吐量指标
- 并发构建数量
- 每分钟构建数
- 用户并发数
- 请求处理速率
3. 资源利用率
- CPU使用率
- 内存使用率
- 磁盘使用率
- 网络使用率
4. 错误率指标
- 构建失败率
- 系统错误率
- 超时错误率
- 连接失败率
性能测试方法：
性能测试策略：
1. 基准测试
- 建立性能基线
- 定期性能回归测试
- 版本间性能对比
- 配置变更影响评估
2. 负载测试
- 模拟正常负载
- 测试系统稳定性
- 验证性能指标
- 识别性能拐点
3. 压力测试
- 测试系统极限
- 识别瓶颈点
- 验证故障恢复
- 评估扩展需求
4. 容量规划
- 预测增长需求
- 评估硬件需求
- 规划扩展策略
- 成本效益分析
性能优化策略
分层优化方法：
优化层次结构：
┌─────────────────────────────────────┐
│ 应用层优化 │
│ - 代码优化 │
│ - 算法优化 │
│ - 缓存策略 │
│ - 异步处理 │
└─────────────────────────────────────┘
┌─────────────────────────────────────┐
│ 中间件优化 │
│ - JVM调优 │
│ - 数据库优化 │
│ - 网络优化 │
│ - 负载均衡 │
└─────────────────────────────────────┘
┌─────────────────────────────────────┐
│ 系统层优化 │
│ - 操作系统调优 │
│ - 硬件配置 │
│ - 存储优化 │
│ - 网络配置 │
└─────────────────────────────────────┘
优化原则：
1. 先测量，后优化
2. 优化最大瓶颈
3. 平衡各项指标
4. 持续监控验证
13.2 JVM调优
内存配置优化
堆内存配置：
# Jenkins启动脚本优化
#!/bin/bash
# 基础内存配置（适用于中等规模Jenkins）
JAVA_OPTS="
-Xms4g # 初始堆大小
-Xmx8g # 最大堆大小
-XX:NewRatio=1 # 新生代与老年代比例
-XX:SurvivorRatio=8 # Eden与Survivor比例
-XX:MaxMetaspaceSize=512m # 元空间最大大小
-XX:CompressedClassSpaceSize=128m
"
# 垃圾回收器配置（推荐G1GC）
GC_OPTS="
-XX:+UseG1GC # 使用G1垃圾回收器
-XX:MaxGCPauseMillis=200 # 最大GC暂停时间
-XX:G1HeapRegionSize=16m # G1堆区域大小
-XX:G1NewSizePercent=30 # 新生代初始占比
-XX:G1MaxNewSizePercent=40 # 新生代最大占比
-XX:G1MixedGCCountTarget=8 # 混合GC目标次数
-XX:InitiatingHeapOccupancyPercent=45 # 并发标记触发阈值
"
# 大规模Jenkins配置（高并发场景）
LARGE_SCALE_OPTS="
-Xms16g
-Xmx32g
-XX:NewRatio=1
-XX:SurvivorRatio=6
-XX:MaxMetaspaceSize=1g
-XX:+UseG1GC
-XX:MaxGCPauseMillis=100
-XX:G1HeapRegionSize=32m
-XX:ParallelGCThreads=16
-XX:ConcGCThreads=4
"
# 性能监控和调试选项
MONITORING_OPTS="
-XX:+PrintGC # 打印GC信息
-XX:+PrintGCDetails # 详细GC信息
-XX:+PrintGCTimeStamps # GC时间戳
-XX:+PrintGCApplicationStoppedTime # 应用暂停时间
-Xloggc:/var/log/jenkins/gc.log # GC日志文件
-XX:+UseGCLogFileRotation # GC日志轮转
-XX:NumberOfGCLogFiles=10 # GC日志文件数量
-XX:GCLogFileSize=100M # GC日志文件大小
-XX:+HeapDumpOnOutOfMemoryError # OOM时生成堆转储
-XX:HeapDumpPath=/var/log/jenkins/heapdump.hprof
"
# JIT编译器优化
JIT_OPTS="
-XX:+TieredCompilation # 分层编译
-XX:TieredStopAtLevel=4 # 编译级别
-XX:CompileThreshold=10000 # 编译阈值
-XX:+UseCodeCacheFlushing # 代码缓存清理
-XX:ReservedCodeCacheSize=256m # 代码缓存大小
"
# 网络和I/O优化
NETWORK_OPTS="
-Djava.net.preferIPv4Stack=true
-Djava.awt.headless=true
-Dfile.encoding=UTF-8
-Dsun.jnu.encoding=UTF-8
-Dhudson.model.DirectoryBrowserSupport.CSP=
-Djenkins.install.runSetupWizard=false
"
# 组合所有选项
export JAVA_OPTS="$JAVA_OPTS $GC_OPTS $MONITORING_OPTS $JIT_OPTS $NETWORK_OPTS"
# 启动Jenkins
java $JAVA_OPTS -jar jenkins.war --httpPort=8080
内存分析脚本：
#!/bin/bash
# jenkins_memory_analysis.sh
JENKINS_PID=$(pgrep -f jenkins.war)
if [ -z "$JENKINS_PID" ]; then
echo "Jenkins进程未找到"
exit 1
fi
echo "=== Jenkins内存分析报告 ==="
echo "时间: $(date)"
echo "PID: $JENKINS_PID"
echo
# 基本内存信息
echo "=== 基本内存信息 ==="
jcmd $JENKINS_PID VM.info | grep -E "(heap|metaspace|code cache)"
echo
# 堆内存使用情况
echo "=== 堆内存使用情况 ==="
jcmd $JENKINS_PID GC.run_finalization
jcmd $JENKINS_PID VM.memory
echo
# GC统计信息
echo "=== GC统计信息 ==="
jstat -gc $JENKINS_PID
echo
# 类加载统计
echo "=== 类加载统计 ==="
jstat -class $JENKINS_PID
echo
# 编译统计
echo "=== JIT编译统计 ==="
jstat -compiler $JENKINS_PID
echo
# 生成堆转储（可选）
read -p "是否生成堆转储文件？(y/N): " -n 1 -r
echo
if [[ $REPLY =~ ^[Yy]$ ]]; then
DUMP_FILE="/tmp/jenkins_heapdump_$(date +%Y%m%d_%H%M%S).hprof"
echo "生成堆转储文件: $DUMP_FILE"
jcmd $JENKINS_PID GC.run_finalization
jcmd $JENKINS_PID VM.memory
jhsdb jmap --heap --pid $JENKINS_PID
fi
# 内存使用趋势分析
echo "=== 内存使用趋势（最近10次采样） ==="
for i in {1..10}; do
echo "采样 $i:"
jstat -gc $JENKINS_PID | tail -1
sleep 5
done
垃圾回收优化
G1GC调优配置：
# G1GC详细配置
G1_TUNING_OPTS="
# 基础G1配置
-XX:+UseG1GC
-XX:MaxGCPauseMillis=200
# 堆区域配置
-XX:G1HeapRegionSize=16m
-XX:G1NewSizePercent=20
-XX:G1MaxNewSizePercent=30
# 并发标记配置
-XX:InitiatingHeapOccupancyPercent=45
-XX:G1MixedGCLiveThresholdPercent=85
-XX:G1HeapWastePercent=5
# 混合GC配置
-XX:G1MixedGCCountTarget=8
-XX:G1OldCSetRegionThreshold=10
# 并发线程配置
-XX:ConcGCThreads=4
-XX:ParallelGCThreads=16
# 字符串去重（Java 8u20+）
-XX:+UseStringDeduplication
# 大对象处理
-XX:G1ReservePercent=10
"
# GC日志详细配置
GC_LOGGING_OPTS="
-Xloggc:/var/log/jenkins/gc-%t.log
-XX:+UseGCLogFileRotation
-XX:NumberOfGCLogFiles=10
-XX:GCLogFileSize=100M
-XX:+PrintGC
-XX:+PrintGCDetails
-XX:+PrintGCTimeStamps
-XX:+PrintGCDateStamps
-XX:+PrintGCApplicationStoppedTime
-XX:+PrintGCApplicationConcurrentTime
-XX:+PrintStringDeduplicationStatistics
"
GC分析脚本：
#!/usr/bin/env python3
# gc_analysis.py
import re
import sys
from datetime import datetime
from collections import defaultdict
class GCAnalyzer:
def __init__(self, log_file):
self.log_file = log_file
self.gc_events = []
self.pause_times = []
self.heap_usage = []
def parse_gc_log(self):
"""解析GC日志文件"""
with open(self.log_file, 'r') as f:
for line in f:
self._parse_line(line.strip())
def _parse_line(self, line):
"""解析单行GC日志"""
# 解析G1GC暂停时间
pause_pattern = r'\[GC pause.*?([0-9.]+) secs\]'
pause_match = re.search(pause_pattern, line)
if pause_match:
pause_time = float(pause_match.group(1)) * 1000 # 转换为毫秒
self.pause_times.append(pause_time)
# 解析堆使用情况
heap_pattern = r'(\d+)M->(\d+)M\((\d+)M\)'
heap_match = re.search(heap_pattern, line)
if heap_match:
before = int(heap_match.group(1))
after = int(heap_match.group(2))
total = int(heap_match.group(3))
self.heap_usage.append({
'before': before,
'after': after,
'total': total,
'utilization': (after / total) * 100
})
def analyze(self):
"""分析GC性能"""
if not self.pause_times:
print("未找到GC暂停时间数据")
return
# 暂停时间统计
avg_pause = sum(self.pause_times) / len(self.pause_times)
max_pause = max(self.pause_times)
min_pause = min(self.pause_times)
# 计算百分位数
sorted_pauses = sorted(self.pause_times)
p95_pause = sorted_pauses[int(len(sorted_pauses) * 0.95)]
p99_pause = sorted_pauses[int(len(sorted_pauses) * 0.99)]
print("=== GC性能分析报告 ===")
print(f"总GC次数: {len(self.pause_times)}")
print(f"平均暂停时间: {avg_pause:.2f}ms")
print(f"最大暂停时间: {max_pause:.2f}ms")
print(f"最小暂停时间: {min_pause:.2f}ms")
print(f"95%暂停时间: {p95_pause:.2f}ms")
print(f"99%暂停时间: {p99_pause:.2f}ms")
# 堆使用情况分析
if self.heap_usage:
avg_utilization = sum(h['utilization'] for h in self.heap_usage) / len(self.heap_usage)
max_utilization = max(h['utilization'] for h in self.heap_usage)
print(f"\n=== 堆使用情况 ===")
print(f"平均堆使用率: {avg_utilization:.2f}%")
print(f"最大堆使用率: {max_utilization:.2f}%")
# 性能建议
self._provide_recommendations(avg_pause, max_pause, p95_pause)
def _provide_recommendations(self, avg_pause, max_pause, p95_pause):
"""提供优化建议"""
print("\n=== 优化建议 ===")
if avg_pause > 200:
print("- 平均暂停时间过长，建议减小MaxGCPauseMillis目标")
if max_pause > 1000:
print("- 最大暂停时间过长，建议增加堆大小或调整G1参数")
if p95_pause > 500:
print("- 95%暂停时间过长，建议优化应用代码减少对象分配")
if len(self.pause_times) > 1000:
print("- GC频率过高，建议增加堆大小")
if __name__ == "__main__":
if len(sys.argv) != 2:
print("用法: python3 gc_analysis.py <gc_log_file>")
sys.exit(1)
analyzer = GCAnalyzer(sys.argv[1])
analyzer.parse_gc_log()
analyzer.analyze()
13.3 系统级优化
操作系统调优
Linux系统优化：
#!/bin/bash
# jenkins_system_tuning.sh
echo "=== Jenkins系统优化脚本 ==="
# 1. 内核参数优化
echo "配置内核参数..."
cat >> /etc/sysctl.conf << EOF
# Jenkins系统优化参数
# 网络优化
net.core.somaxconn = 65535
net.core.netdev_max_backlog = 5000
net.ipv4.tcp_max_syn_backlog = 65535
net.ipv4.tcp_fin_timeout = 30
net.ipv4.tcp_keepalive_time = 1200
net.ipv4.tcp_max_tw_buckets = 6000
net.ipv4.ip_local_port_range = 1024 65535
net.ipv4.tcp_rmem = 4096 65536 16777216
net.ipv4.tcp_wmem = 4096 65536 16777216
net.core.rmem_default = 262144
net.core.rmem_max = 16777216
net.core.wmem_default = 262144
net.core.wmem_max = 16777216
# 内存管理
vm.swappiness = 1
vm.dirty_ratio = 15
vm.dirty_background_ratio = 5
vm.vfs_cache_pressure = 50
vm.min_free_kbytes = 65536
# 文件系统
fs.file-max = 2097152
fs.nr_open = 2097152
# 进程限制
kernel.pid_max = 4194304
kernel.threads-max = 4194304
EOF
# 应用内核参数
sysctl -p
# 2. 文件描述符限制
echo "配置文件描述符限制..."
cat >> /etc/security/limits.conf << EOF
# Jenkins用户限制
jenkins soft nofile 65535
jenkins hard nofile 65535
jenkins soft nproc 32768
jenkins hard nproc 32768
jenkins soft memlock unlimited
jenkins hard memlock unlimited
# 所有用户默认限制
* soft nofile 65535
* hard nofile 65535
EOF
# 3. systemd服务限制
echo "配置systemd服务限制..."
mkdir -p /etc/systemd/system/jenkins.service.d
cat > /etc/systemd/system/jenkins.service.d/limits.conf << EOF
[Service]
LimitNOFILE=65535
LimitNPROC=32768
LimitMEMLOCK=infinity
EOF
# 4. 磁盘I/O优化
echo "优化磁盘I/O..."
# 设置I/O调度器为deadline（适合SSD）
echo deadline > /sys/block/sda/queue/scheduler
# 禁用透明大页
echo never > /sys/kernel/mm/transparent_hugepage/enabled
echo never > /sys/kernel/mm/transparent_hugepage/defrag
# 5. CPU优化
echo "优化CPU设置..."
# 设置CPU调度器
echo performance > /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor
# 6. 创建优化的挂载选项
echo "优化文件系统挂载选项..."
cat >> /etc/fstab << EOF
# Jenkins工作目录优化挂载
/dev/sdb1 /var/lib/jenkins ext4 defaults,noatime,nodiratime,barrier=0 0 2
EOF
echo "系统优化完成，建议重启系统使所有设置生效"
性能监控脚本：
#!/bin/bash
# jenkins_performance_monitor.sh
LOG_FILE="/var/log/jenkins/performance.log"
INTERVAL=60 # 监控间隔（秒）
# 创建日志目录
mkdir -p $(dirname $LOG_FILE)
echo "Jenkins性能监控启动，日志文件: $LOG_FILE"
echo "监控间隔: ${INTERVAL}秒"
while true; do
TIMESTAMP=$(date '+%Y-%m-%d %H:%M:%S')
# 获取Jenkins进程信息
JENKINS_PID=$(pgrep -f jenkins.war)
if [ -n "$JENKINS_PID" ]; then
# CPU使用率
CPU_USAGE=$(ps -p $JENKINS_PID -o %cpu --no-headers)
# 内存使用情况
MEMORY_INFO=$(ps -p $JENKINS_PID -o %mem,vsz,rss --no-headers)
MEM_PERCENT=$(echo $MEMORY_INFO | awk '{print $1}')
VSZ=$(echo $MEMORY_INFO | awk '{print $2}')
RSS=$(echo $MEMORY_INFO | awk '{print $3}')
# 文件描述符使用情况
FD_COUNT=$(lsof -p $JENKINS_PID 2>/dev/null | wc -l)
# 线程数量
THREAD_COUNT=$(ps -p $JENKINS_PID -o nlwp --no-headers)
# 系统负载
LOAD_AVG=$(uptime | awk -F'load average:' '{print $2}' | sed 's/^[ \t]*//')
# 磁盘使用情况
DISK_USAGE=$(df -h /var/lib/jenkins | tail -1 | awk '{print $5}' | sed 's/%//')
# JVM堆内存使用情况（如果jstat可用）
if command -v jstat >/dev/null 2>&1; then
HEAP_INFO=$(jstat -gc $JENKINS_PID | tail -1)
HEAP_USED=$(echo $HEAP_INFO | awk '{print ($3+$4+$6+$8)/1024}' | bc -l 2>/dev/null || echo "N/A")
HEAP_TOTAL=$(echo $HEAP_INFO | awk '{print ($1+$2+$5+$7)/1024}' | bc -l 2>/dev/null || echo "N/A")
else
HEAP_USED="N/A"
HEAP_TOTAL="N/A"
fi
# 记录性能数据
echo "$TIMESTAMP,CPU:${CPU_USAGE}%,MEM:${MEM_PERCENT}%,VSZ:${VSZ}KB,RSS:${RSS}KB,FD:${FD_COUNT},THREADS:${THREAD_COUNT},LOAD:${LOAD_AVG},DISK:${DISK_USAGE}%,HEAP_USED:${HEAP_USED}MB,HEAP_TOTAL:${HEAP_TOTAL}MB" >> $LOG_FILE
# 检查性能阈值并告警
if (( $(echo "$CPU_USAGE > 80" | bc -l) )); then
echo "[$TIMESTAMP] 警告: CPU使用率过高 ${CPU_USAGE}%" | tee -a $LOG_FILE
fi
if (( $(echo "$MEM_PERCENT > 85" | bc -l) )); then
echo "[$TIMESTAMP] 警告: 内存使用率过高 ${MEM_PERCENT}%" | tee -a $LOG_FILE
fi
if [ "$FD_COUNT" -gt 50000 ]; then
echo "[$TIMESTAMP] 警告: 文件描述符使用过多 $FD_COUNT" | tee -a $LOG_FILE
fi
if [ "$DISK_USAGE" -gt 85 ]; then
echo "[$TIMESTAMP] 警告: 磁盘使用率过高 ${DISK_USAGE}%" | tee -a $LOG_FILE
fi
else
echo "[$TIMESTAMP] Jenkins进程未运行" >> $LOG_FILE
fi
sleep $INTERVAL
done
存储优化
磁盘配置优化：
#!/bin/bash
# jenkins_storage_optimization.sh
echo "=== Jenkins存储优化 ==="
# 1. 创建优化的文件系统结构
echo "创建优化的目录结构..."
# Jenkins主目录
JENKINS_HOME="/var/lib/jenkins"
# 分离不同类型的数据
mkdir -p $JENKINS_HOME/{jobs,workspace,logs,plugins,tools,secrets}
mkdir -p /var/cache/jenkins/{builds,artifacts}
mkdir -p /tmp/jenkins/{workspace,builds}
# 2. 配置tmpfs用于临时文件
echo "配置tmpfs..."
cat >> /etc/fstab << EOF
# Jenkins临时文件系统
tmpfs /tmp/jenkins tmpfs defaults,size=4G,mode=1777 0 0
EOF
# 3. 设置合适的文件权限
echo "设置文件权限..."
chown -R jenkins:jenkins $JENKINS_HOME
chown -R jenkins:jenkins /var/cache/jenkins
chown -R jenkins:jenkins /tmp/jenkins
# 4. 配置日志轮转
echo "配置日志轮转..."
cat > /etc/logrotate.d/jenkins << EOF
/var/lib/jenkins/logs/*.log {
daily
missingok
rotate 30
compress
delaycompress
notifempty
copytruncate
su jenkins jenkins
}
/var/log/jenkins/*.log {
daily
missingok
rotate 30
compress
delaycompress
notifempty
copytruncate
su jenkins jenkins
}
EOF
# 5. 创建清理脚本
cat > /usr/local/bin/jenkins_cleanup.sh << 'EOF'
#!/bin/bash
# Jenkins存储清理脚本
JENKINS_HOME="/var/lib/jenkins"
RETENTION_DAYS=30
WORKSPACE_RETENTION_DAYS=7
echo "开始Jenkins存储清理..."
# 清理旧的构建日志
echo "清理构建日志..."
find $JENKINS_HOME/jobs/*/builds/*/log -type f -mtime +$RETENTION_DAYS -delete
# 清理旧的工作空间
echo "清理工作空间..."
find $JENKINS_HOME/workspace/* -type d -mtime +$WORKSPACE_RETENTION_DAYS -exec rm -rf {} + 2>/dev/null
# 清理临时文件
echo "清理临时文件..."
find /tmp/jenkins -type f -mtime +1 -delete
find /var/cache/jenkins -type f -mtime +$RETENTION_DAYS -delete
# 清理旧的插件缓存
echo "清理插件缓存..."
find $JENKINS_HOME/plugins -name "*.tmp" -delete
find $JENKINS_HOME/plugins -name "*.bak" -mtime +7 -delete
# 压缩旧的构建产物
echo "压缩构建产物..."
find $JENKINS_HOME/jobs/*/builds/*/archive -type f -name "*.jar" -mtime +7 ! -name "*.gz" -exec gzip {} \;
# 统计清理结果
echo "清理完成，当前磁盘使用情况:"
df -h $JENKINS_HOME
echo "Jenkins目录大小:"
du -sh $JENKINS_HOME
EOF
chmod +x /usr/local/bin/jenkins_cleanup.sh
# 6. 设置定时清理任务
echo "设置定时清理任务..."
cat > /etc/cron.d/jenkins-cleanup << EOF
# Jenkins存储清理任务
0 2 * * * jenkins /usr/local/bin/jenkins_cleanup.sh >> /var/log/jenkins/cleanup.log 2>&1
EOF
echo "存储优化配置完成"
存储监控脚本：
#!/usr/bin/env python3
# jenkins_storage_monitor.py
import os
import sys
import json
import time
from datetime import datetime
from pathlib import Path
class StorageMonitor:
def __init__(self, jenkins_home='/var/lib/jenkins'):
self.jenkins_home = Path(jenkins_home)
self.report_file = '/var/log/jenkins/storage_report.json'
def get_directory_size(self, path):
"""获取目录大小"""
total_size = 0
try:
for dirpath, dirnames, filenames in os.walk(path):
for filename in filenames:
filepath = os.path.join(dirpath, filename)
try:
total_size += os.path.getsize(filepath)
except (OSError, IOError):
continue
except (OSError, IOError):
pass
return total_size
def get_disk_usage(self, path):
"""获取磁盘使用情况"""
try:
statvfs = os.statvfs(path)
total = statvfs.f_frsize * statvfs.f_blocks
free = statvfs.f_frsize * statvfs.f_available
used = total - free
return {
'total': total,
'used': used,
'free': free,
'usage_percent': (used / total) * 100 if total > 0 else 0
}
except OSError:
return None
def analyze_jenkins_storage(self):
"""分析Jenkins存储使用情况"""
report = {
'timestamp': datetime.now().isoformat(),
'jenkins_home': str(self.jenkins_home),
'directories': {},
'disk_usage': {},
'recommendations': []
}
# 分析各个目录的大小
directories_to_check = [
'jobs',
'workspace',
'plugins',
'logs',
'tools',
'secrets',
'userContent',
'war'
]
total_jenkins_size = 0
for dir_name in directories_to_check:
dir_path = self.jenkins_home / dir_name
if dir_path.exists():
size = self.get_directory_size(dir_path)
total_jenkins_size += size
report['directories'][dir_name] = {
'size_bytes': size,
'size_mb': size / (1024 * 1024),
'size_gb': size / (1024 * 1024 * 1024)
}
report['total_jenkins_size'] = {
'size_bytes': total_jenkins_size,
'size_mb': total_jenkins_size / (1024 * 1024),
'size_gb': total_jenkins_size / (1024 * 1024 * 1024)
}
# 获取磁盘使用情况
disk_usage = self.get_disk_usage(self.jenkins_home)
if disk_usage:
report['disk_usage'] = {
'total_gb': disk_usage['total'] / (1024 * 1024 * 1024),
'used_gb': disk_usage['used'] / (1024 * 1024 * 1024),
'free_gb': disk_usage['free'] / (1024 * 1024 * 1024),
'usage_percent': disk_usage['usage_percent']
}
# 生成建议
self._generate_recommendations(report)
return report
def _generate_recommendations(self, report):
"""生成优化建议"""
recommendations = []
# 检查磁盘使用率
if 'disk_usage' in report and report['disk_usage']['usage_percent'] > 85:
recommendations.append({
'type': 'critical',
'message': f"磁盘使用率过高 ({report['disk_usage']['usage_percent']:.1f}%)，需要立即清理"
})
# 检查各目录大小
if 'directories' in report:
# 检查workspace目录
if 'workspace' in report['directories']:
workspace_size_gb = report['directories']['workspace']['size_gb']
if workspace_size_gb > 10:
recommendations.append({
'type': 'warning',
'message': f"workspace目录过大 ({workspace_size_gb:.1f}GB)，建议清理旧的工作空间"
})
# 检查jobs目录
if 'jobs' in report['directories']:
jobs_size_gb = report['directories']['jobs']['size_gb']
if jobs_size_gb > 20:
recommendations.append({
'type': 'warning',
'message': f"jobs目录过大 ({jobs_size_gb:.1f}GB)，建议清理旧的构建记录"
})
# 检查logs目录
if 'logs' in report['directories']:
logs_size_gb = report['directories']['logs']['size_gb']
if logs_size_gb > 5:
recommendations.append({
'type': 'info',
'message': f"logs目录较大 ({logs_size_gb:.1f}GB)，建议配置日志轮转"
})
report['recommendations'] = recommendations
def save_report(self, report):
"""保存报告到文件"""
os.makedirs(os.path.dirname(self.report_file), exist_ok=True)
with open(self.report_file, 'w') as f:
json.dump(report, f, indent=2)
def print_report(self, report):
"""打印报告"""
print("=== Jenkins存储分析报告 ===")
print(f"时间: {report['timestamp']}")
print(f"Jenkins主目录: {report['jenkins_home']}")
print()
# 总体使用情况
if 'total_jenkins_size' in report:
total_size = report['total_jenkins_size']
print(f"Jenkins总大小: {total_size['size_gb']:.2f} GB")
if 'disk_usage' in report:
disk = report['disk_usage']
print(f"磁盘使用情况: {disk['used_gb']:.1f}GB / {disk['total_gb']:.1f}GB ({disk['usage_percent']:.1f}%)")
print()
# 目录详情
print("=== 目录大小详情 ===")
if 'directories' in report:
for dir_name, info in sorted(report['directories'].items(),
key=lambda x: x[1]['size_gb'], reverse=True):
print(f"{dir_name:15}: {info['size_gb']:8.2f} GB")
print()
# 建议
if 'recommendations' in report and report['recommendations']:
print("=== 优化建议 ===")
for rec in report['recommendations']:
icon = {'critical': '🚨', 'warning': '⚠️', 'info': 'ℹ️'}.get(rec['type'], '')
print(f"{icon} {rec['message']}")
else:
print("✅ 存储使用情况良好，无需特别优化")
def run(self):
"""运行存储监控"""
report = self.analyze_jenkins_storage()
self.save_report(report)
self.print_report(report)
return report
if __name__ == '__main__':
jenkins_home = sys.argv[1] if len(sys.argv) > 1 else '/var/lib/jenkins'
monitor = StorageMonitor(jenkins_home)
monitor.run()
13.4 构建优化
Pipeline性能优化
并行化优化策略：
// 高性能Pipeline示例
pipeline {
agent none
options {
// 构建保留策略
buildDiscarder(logRotator(
numToKeepStr: '10',
daysToKeepStr: '30',
artifactNumToKeepStr: '5'
))
// 超时设置
timeout(time: 30, unit: 'MINUTES')
// 禁用并发构建
disableConcurrentBuilds()
// 跳过默认检出
skipDefaultCheckout()
}
environment {
// 优化环境变量
MAVEN_OPTS = '-Xmx2g -XX:+UseG1GC -Dmaven.repo.local=/var/cache/maven'
GRADLE_OPTS = '-Xmx2g -XX:+UseG1GC -Dorg.gradle.daemon=false'
DOCKER_BUILDKIT = '1'
}
stages {
stage('Checkout') {
agent { label 'fast-ssd' }
steps {
sh 'mvn clean compile -T 4'
}
}
}
}
资源池管理：
// 资源池管理脚本
class ResourcePoolManager {
def jenkins = Jenkins.instance
def pools = [:]
def initializePools() {
pools['build'] = [
maxConcurrent: 10,
current: 0,
queue: [],
nodes: ['build-1', 'build-2', 'build-3']
]
pools['test'] = [
maxConcurrent: 5,
current: 0,
queue: [],
nodes: ['test-1', 'test-2']
]
pools['deploy'] = [
maxConcurrent: 2,
current: 0,
queue: [],
nodes: ['deploy-1']
]
}
def requestResource(String poolName, Closure task) {
def pool = pools[poolName]
if (pool.current < pool.maxConcurrent) {
pool.current++
try {
task()
} finally {
pool.current--
processQueue(poolName)
}
} else {
pool.queue.add(task)
echo "任务已加入${poolName}队列，当前队列长度: ${pool.queue.size()}"
}
}
def processQueue(String poolName) {
def pool = pools[poolName]
if (pool.queue.size() > 0 && pool.current < pool.maxConcurrent) {
def nextTask = pool.queue.remove(0)
pool.current++
// 异步执行下一个任务
Thread.start {
try {
nextTask()
} finally {
pool.current--
processQueue(poolName)
}
}
}
}
}
// 使用示例
def resourceManager = new ResourcePoolManager()
resourceManager.initializePools()
pipeline {
agent none
stages {
stage('Build') {
steps {
script {
resourceManager.requestResource('build') {
node('build') {
sh 'mvn clean package'
}
}
}
}
}
}
}
13.5 网络优化
带宽优化
网络配置优化：
#!/bin/bash
# jenkins_network_optimization.sh
echo "=== Jenkins网络优化配置 ==="
# 1. TCP优化
echo "配置TCP参数..."
cat >> /etc/sysctl.conf << EOF
# Jenkins网络优化
net.ipv4.tcp_window_scaling = 1
net.ipv4.tcp_timestamps = 1
net.ipv4.tcp_sack = 1
net.ipv4.tcp_no_metrics_save = 1
net.ipv4.tcp_moderate_rcvbuf = 1
net.ipv4.tcp_congestion_control = bbr
net.core.default_qdisc = fq
# 连接跟踪优化
net.netfilter.nf_conntrack_max = 1048576
net.netfilter.nf_conntrack_tcp_timeout_established = 7200
EOF
sysctl -p
# 2. 配置Jenkins反向代理
echo "配置Nginx反向代理..."
cat > /etc/nginx/sites-available/jenkins << 'EOF'
upstream jenkins {
server 127.0.0.1:8080 fail_timeout=0;
}
server {
listen 80;
server_name jenkins.company.com;
return 301 https://$server_name$request_uri;
}
server {
listen 443 ssl http2;
server_name jenkins.company.com;
# SSL配置
ssl_certificate /etc/ssl/certs/jenkins.crt;
ssl_certificate_key /etc/ssl/private/jenkins.key;
ssl_protocols TLSv1.2 TLSv1.3;
ssl_ciphers ECDHE-RSA-AES256-GCM-SHA512:DHE-RSA-AES256-GCM-SHA512;
ssl_prefer_server_ciphers off;
ssl_session_cache shared:SSL:10m;
ssl_session_timeout 10m;
# 性能优化
client_max_body_size 100M;
client_body_timeout 60s;
client_header_timeout 60s;
keepalive_timeout 65s;
send_timeout 60s;
# 压缩配置
gzip on;
gzip_vary on;
gzip_min_length 1024;
gzip_proxied any;
gzip_comp_level 6;
gzip_types
text/plain
text/css
text/xml
text/javascript
application/json
application/javascript
application/xml+rss
application/atom+xml
image/svg+xml;
# 缓存配置
location ~* \.(css|js|png|jpg|jpeg|gif|ico|svg)$ {
expires 1y;
add_header Cache-Control "public, immutable";
access_log off;
}
# Jenkins代理配置
location / {
proxy_set_header Host $http_host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto $scheme;
proxy_set_header X-Forwarded-Port $server_port;
proxy_pass http://jenkins;
proxy_read_timeout 90s;
proxy_redirect http://jenkins https://jenkins.company.com;
# WebSocket支持
proxy_http_version 1.1;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection "upgrade";
# 缓冲优化
proxy_buffering on;
proxy_buffer_size 128k;
proxy_buffers 4 256k;
proxy_busy_buffers_size 256k;
}
# 健康检查
location /health {
access_log off;
return 200 "healthy\n";
add_header Content-Type text/plain;
}
}
EOF
# 启用站点
ln -sf /etc/nginx/sites-available/jenkins /etc/nginx/sites-enabled/
nginx -t && systemctl reload nginx
echo "网络优化配置完成"
CDN配置：
# cloudfront-jenkins.yml
AWSTemplateFormatVersion: '2010-09-09'
Description: 'Jenkins CDN配置'
Parameters:
JenkinsOrigin:
Type: String
Default: 'jenkins.company.com'
Description: 'Jenkins服务器域名'
Resources:
JenkinsCDN:
Type: AWS::CloudFront::Distribution
Properties:
DistributionConfig:
Enabled: true
Comment: 'Jenkins CDN Distribution'
Origins:
- Id: jenkins-origin
DomainName: !Ref JenkinsOrigin
CustomOriginConfig:
HTTPPort: 443
HTTPSPort: 443
OriginProtocolPolicy: https-only
OriginSSLProtocols:
- TLSv1.2
DefaultCacheBehavior:
TargetOriginId: jenkins-origin
ViewerProtocolPolicy: redirect-to-https
AllowedMethods:
- GET
- HEAD
- OPTIONS
- PUT
- POST
- PATCH
- DELETE
CachedMethods:
- GET
- HEAD
- OPTIONS
Compress: true
ForwardedValues:
QueryString: true
Headers:
- Authorization
- Host
- X-Forwarded-For
- X-Forwarded-Proto
Cookies:
Forward: all
DefaultTTL: 0
MaxTTL: 31536000
MinTTL: 0
CacheBehaviors:
# 静态资源缓存
- PathPattern: '*.css'
TargetOriginId: jenkins-origin
ViewerProtocolPolicy: redirect-to-https
AllowedMethods: [GET, HEAD]
CachedMethods: [GET, HEAD]
Compress: true
ForwardedValues:
QueryString: false
Headers: []
DefaultTTL: 86400
MaxTTL: 31536000
MinTTL: 0
- PathPattern: '*.js'
TargetOriginId: jenkins-origin
ViewerProtocolPolicy: redirect-to-https
AllowedMethods: [GET, HEAD]
CachedMethods: [GET, HEAD]
Compress: true
ForwardedValues:
QueryString: false
Headers: []
DefaultTTL: 86400
MaxTTL: 31536000
MinTTL: 0
- PathPattern: '*.png'
TargetOriginId: jenkins-origin
ViewerProtocolPolicy: redirect-to-https
AllowedMethods: [GET, HEAD]
CachedMethods: [GET, HEAD]
Compress: false
ForwardedValues:
QueryString: false
Headers: []
DefaultTTL: 2592000
MaxTTL: 31536000
MinTTL: 0
PriceClass: PriceClass_100
ViewerCertificate:
AcmCertificateArn: !Ref SSLCertificate
SslSupportMethod: sni-only
MinimumProtocolVersion: TLSv1.2_2021
SSLCertificate:
Type: AWS::CertificateManager::Certificate
Properties:
DomainName: !Sub 'cdn.${JenkinsOrigin}'
ValidationMethod: DNS
Outputs:
CDNDomainName:
Description: 'CloudFront域名'
Value: !GetAtt JenkinsCDN.DomainName
Export:
Name: !Sub '${AWS::StackName}-CDN-Domain'
连接优化
连接池配置：
// Jenkins系统配置脚本
import jenkins.model.Jenkins
import org.jenkinsci.plugins.workflow.libs.GlobalLibraries
import org.jenkinsci.plugins.workflow.libs.LibraryConfiguration
import org.jenkinsci.plugins.workflow.libs.SCMSourceRetriever
import jenkins.plugins.git.GitSCMSource
// HTTP连接池优化
System.setProperty('hudson.model.ParametersAction.keepUndefinedParameters', 'true')
System.setProperty('hudson.model.DirectoryBrowserSupport.CSP', '')
System.setProperty('jenkins.model.Jenkins.slaveAgentPort', '50000')
System.setProperty('jenkins.model.Jenkins.slaveAgentPortEnforce', 'true')
// 网络超时设置
System.setProperty('hudson.remoting.Launcher.pingIntervalSec', '300')
System.setProperty('hudson.remoting.Launcher.pingTimeoutSec', '60')
System.setProperty('hudson.slaves.ChannelPinger.pingInterval', '5')
System.setProperty('hudson.slaves.ChannelPinger.pingTimeout', '10')
// Git连接优化
System.setProperty('org.jenkinsci.plugins.gitclient.Git.timeOut', '30')
System.setProperty('hudson.plugins.git.GitSCM.ALLOW_LOCAL_CHECKOUT', 'true')
// HTTP客户端优化
System.setProperty('hudson.ProxyConfiguration.DEFAULT_CONNECT_TIMEOUT_MILLIS', '20000')
System.setProperty('hudson.ProxyConfiguration.DEFAULT_READ_TIMEOUT_MILLIS', '60000')
println "网络连接优化配置完成"
13.6 监控和调优工具
性能分析工具
JProfiler集成脚本：
#!/bin/bash
# jenkins_profiling.sh
JPROFILER_HOME="/opt/jprofiler"
JENKINS_PID=$(pgrep -f jenkins.war)
if [ -z "$JENKINS_PID" ]; then
echo "Jenkins进程未找到"
exit 1
fi
echo "=== Jenkins性能分析 ==="
echo "Jenkins PID: $JENKINS_PID"
# 1. 启动JProfiler代理
echo "启动JProfiler代理..."
$JPROFILER_HOME/bin/jpenable --pid=$JENKINS_PID --port=8849
# 2. 生成堆转储
echo "生成堆转储..."
HEAP_DUMP_FILE="/tmp/jenkins_heap_$(date +%Y%m%d_%H%M%S).hprof"
jcmd $JENKINS_PID GC.run_finalization
jhsdb jmap --heap --pid $JENKINS_PID > "${HEAP_DUMP_FILE}.txt"
jcmd $JENKINS_PID VM.memory >> "${HEAP_DUMP_FILE}.txt"
# 3. 线程转储
echo "生成线程转储..."
THREAD_DUMP_FILE="/tmp/jenkins_threads_$(date +%Y%m%d_%H%M%S).txt"
jstack $JENKINS_PID > $THREAD_DUMP_FILE
# 4. GC分析
echo "分析GC日志..."
GC_LOG_FILE="/var/log/jenkins/gc.log"
if [ -f "$GC_LOG_FILE" ]; then
# 使用GCViewer分析GC日志
java -jar $JPROFILER_HOME/lib/gcviewer.jar $GC_LOG_FILE
fi
# 5. 性能报告
echo "生成性能报告..."
cat > "/tmp/jenkins_performance_report_$(date +%Y%m%d_%H%M%S).txt" << EOF
Jenkins性能分析报告
生成时间: $(date)
Jenkins PID: $JENKINS_PID
=== 系统信息 ===
$(uname -a)
=== CPU信息 ===
$(lscpu)
=== 内存信息 ===
$(free -h)
=== 磁盘信息 ===
$(df -h)
=== 网络连接 ===
$(netstat -an | grep :8080)
=== Java进程信息 ===
$(ps -p $JENKINS_PID -o pid,ppid,cmd,%mem,%cpu,etime)
=== JVM信息 ===
$(jcmd $JENKINS_PID VM.info)
=== 类加载统计 ===
$(jstat -class $JENKINS_PID)
=== 编译统计 ===
$(jstat -compiler $JENKINS_PID)
=== GC统计 ===
$(jstat -gc $JENKINS_PID)
EOF
echo "性能分析完成，文件保存在 /tmp/ 目录"
ls -la /tmp/jenkins_*
自动化调优脚本：
#!/usr/bin/env python3
# jenkins_auto_tuning.py
import os
import re
import json
import subprocess
from datetime import datetime, timedelta
class JenkinsAutoTuner:
def __init__(self):
self.jenkins_home = '/var/lib/jenkins'
self.config_file = '/etc/jenkins/tuning.json'
self.metrics = {}
def collect_metrics(self):
"""收集性能指标"""
# 获取Jenkins进程信息
jenkins_pid = self._get_jenkins_pid()
if not jenkins_pid:
return False
# CPU使用率
cpu_usage = self._get_cpu_usage(jenkins_pid)
# 内存使用情况
memory_info = self._get_memory_info(jenkins_pid)
# GC信息
gc_info = self._get_gc_info(jenkins_pid)
# 响应时间
response_time = self._get_response_time()
# 构建队列长度
queue_length = self._get_queue_length()
self.metrics = {
'timestamp': datetime.now().isoformat(),
'cpu_usage': cpu_usage,
'memory': memory_info,
'gc': gc_info,
'response_time': response_time,
'queue_length': queue_length
}
return True
def analyze_performance(self):
"""分析性能并生成调优建议"""
recommendations = []
# CPU分析
if self.metrics['cpu_usage'] > 80:
recommendations.append({
'type': 'cpu',
'severity': 'high',
'message': 'CPU使用率过高，建议增加执行器或优化构建脚本',
'action': 'increase_executors'
})
# 内存分析
heap_usage = self.metrics['memory']['heap_usage_percent']
if heap_usage > 85:
recommendations.append({
'type': 'memory',
'severity': 'high',
'message': '堆内存使用率过高，建议增加堆大小',
'action': 'increase_heap_size'
})
# GC分析
gc_time_percent = self.metrics['gc']['time_percent']
if gc_time_percent > 5:
recommendations.append({
'type': 'gc',
'severity': 'medium',
'message': 'GC时间占比过高，建议调整GC参数',
'action': 'tune_gc_parameters'
})
# 响应时间分析
if self.metrics['response_time'] > 5000: # 5秒
recommendations.append({
'type': 'response',
'severity': 'medium',
'message': '响应时间过长，建议优化插件或增加资源',
'action': 'optimize_plugins'
})
# 队列分析
if self.metrics['queue_length'] > 20:
recommendations.append({
'type': 'queue',
'severity': 'medium',
'message': '构建队列过长，建议增加构建节点',
'action': 'add_build_nodes'
})
return recommendations
def apply_tuning(self, recommendations):
"""应用调优建议"""
applied_changes = []
for rec in recommendations:
if rec['action'] == 'increase_heap_size':
if self._increase_heap_size():
applied_changes.append('增加堆内存大小')
elif rec['action'] == 'tune_gc_parameters':
if self._tune_gc_parameters():
applied_changes.append('优化GC参数')
elif rec['action'] == 'increase_executors':
if self._increase_executors():
applied_changes.append('增加执行器数量')
return applied_changes
def _get_jenkins_pid(self):
"""获取Jenkins进程ID"""
try:
result = subprocess.run(['pgrep', '-f', 'jenkins.war'],
capture_output=True, text=True)
return result.stdout.strip() if result.returncode == 0 else None
except:
return None
def _get_cpu_usage(self, pid):
"""获取CPU使用率"""
try:
result = subprocess.run(['ps', '-p', pid, '-o', '%cpu', '--no-headers'],
capture_output=True, text=True)
return float(result.stdout.strip()) if result.returncode == 0 else 0
except:
return 0
def _get_memory_info(self, pid):
"""获取内存信息"""
try:
# 获取进程内存使用
ps_result = subprocess.run(['ps', '-p', pid, '-o', '%mem,vsz,rss', '--no-headers'],
capture_output=True, text=True)
mem_percent, vsz, rss = ps_result.stdout.strip().split()
# 获取JVM堆信息
jstat_result = subprocess.run(['jstat', '-gc', pid],
capture_output=True, text=True)
gc_data = jstat_result.stdout.strip().split('\n')[-1].split()
heap_used = float(gc_data[2]) + float(gc_data[3]) + float(gc_data[5]) + float(gc_data[7])
heap_total = float(gc_data[0]) + float(gc_data[1]) + float(gc_data[4]) + float(gc_data[6])
return {
'mem_percent': float(mem_percent),
'vsz_kb': int(vsz),
'rss_kb': int(rss),
'heap_used_kb': heap_used,
'heap_total_kb': heap_total,
'heap_usage_percent': (heap_used / heap_total) * 100 if heap_total > 0 else 0
}
except:
return {}
def _get_gc_info(self, pid):
"""获取GC信息"""
try:
result = subprocess.run(['jstat', '-gc', pid],
capture_output=True, text=True)
lines = result.stdout.strip().split('\n')
if len(lines) >= 2:
headers = lines[0].split()
values = lines[1].split()
gc_data = dict(zip(headers, values))
# 计算GC时间占比
gc_time = float(gc_data.get('GCT', 0))
uptime_result = subprocess.run(['ps', '-p', pid, '-o', 'etime', '--no-headers'],
capture_output=True, text=True)
uptime_str = uptime_result.stdout.strip()
uptime_seconds = self._parse_uptime(uptime_str)
time_percent = (gc_time / uptime_seconds) * 100 if uptime_seconds > 0 else 0
return {
'total_time': gc_time,
'time_percent': time_percent,
'young_gc_count': int(gc_data.get('YGC', 0)),
'full_gc_count': int(gc_data.get('FGC', 0))
}
except:
pass
return {}
def _parse_uptime(self, uptime_str):
"""解析进程运行时间"""
# 格式: [[DD-]HH:]MM:SS
parts = uptime_str.split(':')
seconds = 0
if len(parts) == 2: # MM:SS
seconds = int(parts[0]) * 60 + int(parts[1])
elif len(parts) == 3: # HH:MM:SS
seconds = int(parts[0]) * 3600 + int(parts[1]) * 60 + int(parts[2])
elif '-' in uptime_str: # DD-HH:MM:SS
day_part, time_part = uptime_str.split('-')
days = int(day_part)
time_parts = time_part.split(':')
seconds = days * 86400 + int(time_parts[0]) * 3600 + int(time_parts[1]) * 60 + int(time_parts[2])
return seconds
def _get_response_time(self):
"""获取响应时间"""
try:
import time
import urllib.request
start_time = time.time()
urllib.request.urlopen('http://localhost:8080/api/json', timeout=10)
end_time = time.time()
return (end_time - start_time) * 1000 # 转换为毫秒
except:
return 0
def _get_queue_length(self):
"""获取构建队列长度"""
try:
import urllib.request
import json
response = urllib.request.urlopen('http://localhost:8080/queue/api/json', timeout=5)
data = json.loads(response.read().decode())
return len(data.get('items', []))
except:
return 0
def _increase_heap_size(self):
"""增加堆内存大小"""
# 这里应该修改Jenkins启动脚本
# 实际实现需要根据具体的部署方式
print("建议增加堆内存大小到当前的1.5倍")
return True
def _tune_gc_parameters(self):
"""调整GC参数"""
print("建议调整GC参数以减少GC时间")
return True
def _increase_executors(self):
"""增加执行器数量"""
print("建议增加执行器数量以提高并发处理能力")
return True
def run(self):
"""运行自动调优"""
print("=== Jenkins自动调优开始 ===")
# 收集指标
if not self.collect_metrics():
print("无法收集性能指标")
return
print(f"当前性能指标:")
print(f" CPU使用率: {self.metrics['cpu_usage']:.1f}%")
print(f" 内存使用率: {self.metrics['memory'].get('heap_usage_percent', 0):.1f}%")
print(f" GC时间占比: {self.metrics['gc'].get('time_percent', 0):.1f}%")
print(f" 响应时间: {self.metrics['response_time']:.0f}ms")
print(f" 队列长度: {self.metrics['queue_length']}")
# 分析性能
recommendations = self.analyze_performance()
if not recommendations:
print("✅ 系统性能良好，无需调优")
return
print(f"\n发现 {len(recommendations)} 个优化建议:")
for i, rec in enumerate(recommendations, 1):
print(f" {i}. [{rec['severity'].upper()}] {rec['message']}")
# 应用调优
applied_changes = self.apply_tuning(recommendations)
if applied_changes:
print(f"\n已应用以下优化:")
for change in applied_changes:
print(f" ✓ {change}")
print("\n=== 自动调优完成 ===")
if __name__ == '__main__':
tuner = JenkinsAutoTuner()
tuner.run()
本章小结
本章详细介绍了Jenkins的性能优化：
- 性能优化概述：了解性能问题识别和优化策略
- JVM调优：掌握内存配置和垃圾回收优化
- 系统级优化：学习操作系统和存储优化
- 构建优化：实现Pipeline和资源管理优化
- 网络优化：配置带宽和连接优化
- 监控调优工具：使用性能分析和自动调优工具
通过系统性的性能优化，可以显著提升Jenkins的运行效率和用户体验。
下一章预告
下一章我们将学习Jenkins的故障排除，包括常见问题诊断、日志分析和恢复策略。
练习与思考
理论练习
性能分析：
- 分析不同类型的性能瓶颈
- 设计性能监控方案
- 制定性能优化计划
调优策略：
- 比较不同JVM垃圾回收器的特点
- 设计资源分配策略
- 规划网络优化方案
实践练习
JVM调优：
- 配置G1GC参数
- 分析GC日志
- 优化内存配置
系统优化：
- 实施操作系统调优
- 配置存储优化
- 部署监控工具
思考题
优化平衡：
- 如何在性能和稳定性之间找到平衡？
- 如何评估优化效果？
- 如何避免过度优化？
持续改进：
如何建立性能优化的持续改进机制？
如何处理性能回归问题？
如何在团队中推广性能优化最佳实践？ {
script {
// 优化的检出策略
checkout([
class:′GitSCM′,branches:[[name:env.BRANCHNAME]],doGenerateSubmoduleConfigurations:false,extensions:[[class: 'GitSCM', branches: [[name: env.BRANCH_NAME]], doGenerateSubmoduleConfigurations: false, extensions: [ [class:′GitSCM′,branches:[[name:env.BRANCHNAME]],doGenerateSubmoduleConfigurations:false,extensions:[[class: ‘CloneOption’,
depth: 1,
noTags: true,
shallow: true],
[$class: ‘CheckoutOption’, timeout: 10]
],
userRemoteConfigs: [[url: env.GIT_URL]]
])
}// 缓存依赖 stash includes: '**', name: 'source-code' }}
stage(‘Parallel Build & Test’) {
parallel {
stage(‘Unit Tests’) {
agent { label ‘test-runner’ }
steps {
unstash ‘source-code’// 使用缓存的依赖 script { if (fileExists('pom.xml')) { sh ''' # Maven并行构建 mvn clean test \ -T 4 \ -Dmaven.test.failure.ignore=true \ -Dmaven.repo.local=/var/cache/maven \ -Dparallel=methods \ -DthreadCount=4 ''' } else if (fileExists('build.gradle')) { sh ''' # Gradle并行构建 ./gradlew test \ --parallel \ --max-workers=4 \ --build-cache \ --gradle-user-home=/var/cache/gradle ''' } } } post { always { publishTestResults( testResultsPattern: '**/target/surefire-reports/*.xml,**/build/test-results/**/*.xml', allowEmptyResults: true ) } } } stage('Code Quality') { agent { label 'sonar-scanner' } steps { unstash 'source-code' script { // 并行代码质量检查 parallel([ 'SonarQube': { sh ''' sonar-scanner \ -Dsonar.projectKey=${JOB_NAME} \ -Dsonar.sources=src \ -Dsonar.host.url=${SONAR_URL} \ -Dsonar.login=${SONAR_TOKEN} ''' }, 'Security Scan': { sh ''' # OWASP依赖检查 dependency-check.sh \ --project ${JOB_NAME} \ --scan . \ --format XML \ --out dependency-check-report.xml ''' } ]) } } } stage('Build Artifacts') { agent { label 'build-server' } steps { unstash 'source-code' script { if (fileExists('pom.xml')) { sh ''' # Maven优化构建 mvn clean package \ -T 4 \ -DskipTests \ -Dmaven.repo.local=/var/cache/maven \ -Dmaven.compile.fork=true \ -Dmaven.compiler.maxmem=1024m ''' } else if (fileExists('Dockerfile')) { sh ''' # Docker多阶段构建 docker build \ --build-arg BUILDKIT_INLINE_CACHE=1 \ --cache-from ${IMAGE_NAME}:cache \ -t ${IMAGE_NAME}:${BUILD_NUMBER} \ -t ${IMAGE_NAME}:latest . ''' } } // 存储构建产物 stash includes: '**/target/*.jar,**/build/libs/*.jar', name: 'artifacts' } } }}
stage(‘Integration Tests’) {
agent { label ‘integration-test’ }
when {
anyOf {
branch ‘main’
branch ‘develop’
changeRequest()
}
}
steps {
unstash ‘source-code’
unstash ‘artifacts’script { // 并行集成测试 def testStages = [:] ['api-tests', 'ui-tests', 'performance-tests'].each { testType -> testStages[testType] = { sh "./run-${testType}.sh" } } parallel testStages } }}
stage(‘Deploy’) {
agent { label ‘deployment’ }
when {
branch ‘main’
}
steps {
unstash ‘artifacts’script { // 蓝绿部署 sh ''' # 部署到蓝绿环境 ./deploy.sh --strategy=blue-green --timeout=300 ''' } }}
}
post {
always {
script {
// 清理工作空间
cleanWs(
cleanWhenAborted: true,
cleanWhenFailure: true,
cleanWhenNotBuilt: true,
cleanWhenSuccess: true,
cleanWhenUnstable: true,
deleteDirs: true
)
}
}success { script { // 成功通知 if (env.BRANCH_NAME == 'main') { slackSend( channel: '#deployments', color: 'good', message: "✅ 部署成功: ${env.JOB_NAME} #${env.BUILD_NUMBER}" ) } } } failure { script { // 失败通知和分析 emailext( subject: "构建失败: ${env.JOB_NAME} #${env.BUILD_NUMBER}", body: ''' 构建失败详情: 项目: ${env.JOB_NAME} 构建号: ${env.BUILD_NUMBER} 分支: ${env.BRANCH_NAME} 提交: ${env.GIT_COMMIT} 查看详情: ${env.BUILD_URL} ''', to: '${DEFAULT_RECIPIENTS}' ) } }}
}
**构建缓存优化：**
```groovy
// 共享库中的缓存管理
@Library('jenkins-shared-library') _
def buildWithCache(Map config) {
def cacheKey = generateCacheKey(config)
def cacheHit = false
stage('Cache Check') {
script {
// 检查缓存是否存在
cacheHit = checkCache(cacheKey)
if (cacheHit) {
echo "缓存命中: ${cacheKey}"
restoreCache(cacheKey)
} else {
echo "缓存未命中，开始构建"
}
}
}
if (!cacheHit) {
stage('Build') {
script {
// 执行构建
config.buildSteps()
// 保存缓存
saveCache(cacheKey, config.cachePatterns)
}
}
}
}
def generateCacheKey(Map config) {
// 基于文件内容生成缓存键
def checksums = []
config.cacheFiles.each { file ->
if (fileExists(file)) {
def checksum = sh(
script: "sha256sum ${file} | cut -d' ' -f1",
returnStdout: true
).trim()
checksums.add(checksum)
}
}
def combinedChecksum = sh(
script: "echo '${checksums.join(',')}' | sha256sum | cut -d' ' -f1",
returnStdout: true
).trim()
return "${config.projectName}-${combinedChecksum}"
}
def checkCache(String cacheKey) {
// 检查S3或其他缓存存储
def exitCode = sh(
script: "aws s3 ls s3://jenkins-cache/${cacheKey}.tar.gz",
returnStatus: true
)
return exitCode == 0
}
def restoreCache(String cacheKey) {
sh """
aws s3 cp s3://jenkins-cache/${cacheKey}.tar.gz cache.tar.gz
tar -xzf cache.tar.gz
rm cache.tar.gz
"""
}
def saveCache(String cacheKey, List patterns) {
def files = patterns.join(' ')
sh """
tar -czf cache.tar.gz ${files}
aws s3 cp cache.tar.gz s3://jenkins-cache/${cacheKey}.tar.gz
rm cache.tar.gz
"""
}
// 使用示例
pipeline {
agent any
stages {
stage('Build with Cache') {
steps {
script {
buildWithCache([
projectName: 'my-app',
cacheFiles: ['pom.xml', 'package.json', 'requirements.txt'],
cachePatterns: ['~/.m2/repository', 'node_modules', '.venv'],
buildSteps: {
sh 'mvn clean package'
sh 'npm install'
sh 'pip install -r requirements.txt'
}
])
}
}
}
}
}
资源管理优化
动态节点管理：
// 智能节点分配脚本
@Library('jenkins-shared-library') _
def allocateOptimalNode(Map requirements) {
def availableNodes = getAvailableNodes()
def optimalNode = selectOptimalNode(availableNodes, requirements)
if (optimalNode) {
return optimalNode
} else {
// 动态创建节点
return createDynamicNode(requirements)
}
}
def getAvailableNodes() {
def nodes = []
Jenkins.instance.computers.each { computer ->
if (computer.isOnline() && !computer.isTemporarilyOffline()) {
def node = computer.getNode()
def executor = computer.getExecutors().find { !it.isBusy() }
if (executor) {
nodes.add([
name: node.getNodeName(),
labels: node.getLabelString().split(' '),
cpu: getNodeCpuUsage(computer),
memory: getNodeMemoryUsage(computer),
disk: getNodeDiskUsage(computer),
load: getNodeLoad(computer)
])
}
}
}
return nodes
}
def selectOptimalNode(List nodes, Map requirements) {
// 过滤满足标签要求的节点
def candidateNodes = nodes.findAll { node ->
requirements.labels.every { label ->
node.labels.contains(label)
}
}
if (candidateNodes.isEmpty()) {
return null
}
// 计算节点得分
def scoredNodes = candidateNodes.collect { node ->
def score = calculateNodeScore(node, requirements)
[node: node, score: score]
}
// 选择得分最高的节点
def bestNode = scoredNodes.max { it.score }
return bestNode.node
}
def calculateNodeScore(Map node, Map requirements) {
def score = 0
// CPU得分（使用率越低得分越高）
score += (100 - node.cpu) * 0.3
// 内存得分
score += (100 - node.memory) * 0.3
// 磁盘得分
score += (100 - node.disk) * 0.2
// 负载得分
score += Math.max(0, 100 - node.load * 20) * 0.2
// 特殊要求加分
if (requirements.preferSSD && node.labels.contains('ssd')) {
score += 10
}
if (requirements.preferHighCpu && node.labels.contains('high-cpu')) {
score += 10
}
return score
}
def createDynamicNode(Map requirements) {
// 基于需求创建云节点
def nodeTemplate = selectNodeTemplate(requirements)
def cloudName = nodeTemplate.cloud
// 触发节点创建
def cloud = Jenkins.instance.getCloud(cloudName)
def provisionedNode = cloud.provision(nodeTemplate, 1)
// 等待节点上线
waitForNodeOnline(provisionedNode.name, 300) // 5分钟超时
return provisionedNode
}
def selectNodeTemplate(Map requirements) {
def templates = [
[
name: 'small-node',
cloud: 'aws-ec2',
instanceType: 't3.medium',
labels: ['linux', 'docker'],
cpu: 2,
memory: 4
],
[
name: 'medium-node',
cloud: 'aws-ec2',
instanceType: 't3.large',
labels: ['linux', 'docker', 'maven'],
cpu: 2,
memory: 8
],
[
name: 'large-node',
cloud: 'aws-ec2',
instanceType: 't3.xlarge',
labels: ['linux', 'docker', 'high-cpu'],
cpu: 4,
memory: 16
]
]
// 选择满足需求的最小模板
def suitableTemplates = templates.findAll { template ->
template.cpu >= requirements.minCpu &&
template.memory >= requirements.minMemory &&
requirements.labels.every { label ->
template.labels.contains(label)
}
}
return suitableTemplates.min { it.cpu + it.memory }
}
// 使用示例
pipeline {
agent none
stages {
stage('Lightweight Tasks') {
agent {
label allocateOptimalNode([
labels: ['linux', 'docker'],
minCpu: 1,
minMemory: 2,
preferSSD: false
]).name
}
steps {
sh 'echo "Running on optimized node"'
}
}
stage('Heavy Compilation') {
agent {
label allocateOptimalNode([
labels: ['linux', 'maven', 'high-cpu'],
minCpu: 4,
minMemory: 8,
preferSSD: true,
preferHighCpu: true
]).name
}
steps