Docker高阶实战：从Redis集群到C++微服务，全面解析镜像优化与生产环境部署+镜像制作常见问题详解

文章目录

• 本篇摘要
• 一.基于`dockerfile`构建`redis`集群
• 二.基于dockerfile编写C++微服务
• 三.docker镜像制作常见问题
• 四.本篇小结

本篇摘要

本文基于Dockerfile实战搭建Redis集群与C++微服务，详解多阶段构建、镜像优化及Nginx负载均衡配置，并解析虚悬镜像与中间层镜像的核心区别与处理方法。

一.基于dockerfile构建redis集群

在这里插入图片描述

• 下面通过容器化技术搭建6个redis的集群；然后第七个redis把他们够成集群后退出。

搭建过程结构：

#mermaid-svg-egQFtxlb4p43OyTD{font-family:\”trebuchet ms\”,verdana,arial,sans-serif;font-size:16px;fill:#333;}@keyframes edge-animation-frame{from{stroke-dashoffset:0;}}@keyframes dash{to{stroke-dashoffset:0;}}#mermaid-svg-egQFtxlb4p43OyTD .edge-animation-slow{stroke-dasharray:9,5!important;stroke-dashoffset:900;animation:dash 50s linear infinite;stroke-linecap:round;}#mermaid-svg-egQFtxlb4p43OyTD .edge-animation-fast{stroke-dasharray:9,5!important;stroke-dashoffset:900;animation:dash 20s linear infinite;stroke-linecap:round;}#mermaid-svg-egQFtxlb4p43OyTD .error-icon{fill:#552222;}#mermaid-svg-egQFtxlb4p43OyTD .error-text{fill:#552222;stroke:#552222;}#mermaid-svg-egQFtxlb4p43OyTD .edge-thickness-normal{stroke-width:1px;}#mermaid-svg-egQFtxlb4p43OyTD .edge-thickness-thick{stroke-width:3.5px;}#mermaid-svg-egQFtxlb4p43OyTD .edge-pattern-solid{stroke-dasharray:0;}#mermaid-svg-egQFtxlb4p43OyTD .edge-thickness-invisible{stroke-width:0;fill:none;}#mermaid-svg-egQFtxlb4p43OyTD .edge-pattern-dashed{stroke-dasharray:3;}#mermaid-svg-egQFtxlb4p43OyTD .edge-pattern-dotted{stroke-dasharray:2;}#mermaid-svg-egQFtxlb4p43OyTD .marker{fill:#333333;stroke:#333333;}#mermaid-svg-egQFtxlb4p43OyTD .marker.cross{stroke:#333333;}#mermaid-svg-egQFtxlb4p43OyTD svg{font-family:\”trebuchet ms\”,verdana,arial,sans-serif;font-size:16px;}#mermaid-svg-egQFtxlb4p43OyTD p{margin:0;}#mermaid-svg-egQFtxlb4p43OyTD .label{font-family:\”trebuchet ms\”,verdana,arial,sans-serif;color:#333;}#mermaid-svg-egQFtxlb4p43OyTD .cluster-label text{fill:#333;}#mermaid-svg-egQFtxlb4p43OyTD .cluster-label span{color:#333;}#mermaid-svg-egQFtxlb4p43OyTD .cluster-label span p{background-color:transparent;}#mermaid-svg-egQFtxlb4p43OyTD .label text,#mermaid-svg-egQFtxlb4p43OyTD span{fill:#333;color:#333;}#mermaid-svg-egQFtxlb4p43OyTD .node rect,#mermaid-svg-egQFtxlb4p43OyTD .node circle,#mermaid-svg-egQFtxlb4p43OyTD .node ellipse,#mermaid-svg-egQFtxlb4p43OyTD .node polygon,#mermaid-svg-egQFtxlb4p43OyTD .node path{fill:#ECECFF;stroke:#9370DB;stroke-width:1px;}#mermaid-svg-egQFtxlb4p43OyTD .rough-node .label text,#mermaid-svg-egQFtxlb4p43OyTD .node .label text,#mermaid-svg-egQFtxlb4p43OyTD .image-shape .label,#mermaid-svg-egQFtxlb4p43OyTD .icon-shape .label{text-anchor:middle;}#mermaid-svg-egQFtxlb4p43OyTD .node .katex path{fill:#000;stroke:#000;stroke-width:1px;}#mermaid-svg-egQFtxlb4p43OyTD .rough-node .label,#mermaid-svg-egQFtxlb4p43OyTD .node .label,#mermaid-svg-egQFtxlb4p43OyTD .image-shape .label,#mermaid-svg-egQFtxlb4p43OyTD .icon-shape .label{text-align:center;}#mermaid-svg-egQFtxlb4p43OyTD .node.clickable{cursor:pointer;}#mermaid-svg-egQFtxlb4p43OyTD .root .anchor path{fill:#333333!important;stroke-width:0;stroke:#333333;}#mermaid-svg-egQFtxlb4p43OyTD .arrowheadPath{fill:#333333;}#mermaid-svg-egQFtxlb4p43OyTD .edgePath .path{stroke:#333333;stroke-width:2.0px;}#mermaid-svg-egQFtxlb4p43OyTD .flowchart-link{stroke:#333333;fill:none;}#mermaid-svg-egQFtxlb4p43OyTD .edgeLabel{background-color:rgba(232,232,232, 0.8);text-align:center;}#mermaid-svg-egQFtxlb4p43OyTD .edgeLabel p{background-color:rgba(232,232,232, 0.8);}#mermaid-svg-egQFtxlb4p43OyTD .edgeLabel rect{opacity:0.5;background-color:rgba(232,232,232, 0.8);fill:rgba(232,232,232, 0.8);}#mermaid-svg-egQFtxlb4p43OyTD .labelBkg{background-color:rgba(232, 232, 232, 0.5);}#mermaid-svg-egQFtxlb4p43OyTD .cluster rect{fill:#ffffde;stroke:#aaaa33;stroke-width:1px;}#mermaid-svg-egQFtxlb4p43OyTD .cluster text{fill:#333;}#mermaid-svg-egQFtxlb4p43OyTD .cluster span{color:#333;}#mermaid-svg-egQFtxlb4p43OyTD div.mermaidTooltip{position:absolute;text-align:center;max-width:200px;padding:2px;font-family:\”trebuchet ms\”,verdana,arial,sans-serif;font-size:12px;background:hsl(80, 100%, 96.2745098039%);border:1px solid #aaaa33;border-radius:2px;pointer-events:none;z-index:100;}#mermaid-svg-egQFtxlb4p43OyTD .flowchartTitleText{text-anchor:middle;font-size:18px;fill:#333;}#mermaid-svg-egQFtxlb4p43OyTD rect.text{fill:none;stroke-width:0;}#mermaid-svg-egQFtxlb4p43OyTD .icon-shape,#mermaid-svg-egQFtxlb4p43OyTD .image-shape{background-color:rgba(232,232,232, 0.8);text-align:center;}#mermaid-svg-egQFtxlb4p43OyTD .icon-shape p,#mermaid-svg-egQFtxlb4p43OyTD .image-shape p{background-color:rgba(232,232,232, 0.8);padding:2px;}#mermaid-svg-egQFtxlb4p43OyTD .icon-shape .label rect,#mermaid-svg-egQFtxlb4p43OyTD .image-shape .label rect{opacity:0.5;background-color:rgba(232,232,232, 0.8);fill:rgba(232,232,232, 0.8);}#mermaid-svg-egQFtxlb4p43OyTD .label-icon{display:inline-block;height:1em;overflow:visible;vertical-align:-0.125em;}#mermaid-svg-egQFtxlb4p43OyTD .node .label-icon path{fill:currentColor;stroke:revert;stroke-width:revert;}#mermaid-svg-egQFtxlb4p43OyTD :root{–mermaid-font-family:\”trebuchet ms\”,verdana,arial,sans-serif;}

下载配置redis集群

编写dockerfile制作镜像

启动容器测试镜像能否正常运行

编写集群docker-compose.yml

启动集群

测试功能-清理资源

首先下载对应源码：

Redis源码包

然后本地解压拿到对应的conf文件：

• 修改成集群相关配置。

修改如下:

• 对应文件拖转进来。

对应目录结构：

root@iv-ye1i2elts0wh2yp1ahah:/home/youxing/docker-learn/master_slave/rediscluster# tree
.
├── docker-compose.yml
└── redis
├── dockerfile
├── redis-7.0.11.tar.gz
└── redis.conf

1 directory, 4 files

下面看下对应的dockerfile文件：

# 构建阶段：编译Redis
FROM ubuntu:22.04 as buildstage

# 替换阿里云源并更新
RUN sed –i 's@//.*archive.ubuntu.com@//mirrors.aliyun.com@g' /etc/apt/sources.list && \\
apt-get update

# 安装编译工具
RUN apt–get install –y build–essential

# 添加Redis源码和配置
ADD redis–7.0.11.tar.gz /
ADD redis.conf /redis/

# 编译Redis
WORKDIR /redis–7.0.11
RUN make

# 移动编译好的文件
RUN mv src/redis–server /redis/ && \\
mv src/redis–cli /redis/

# 运行阶段：生产镜像
FROM ubuntu:22.04

# 安装运行时依赖
RUN apt–get update && \\
apt–get install –y —no–install–recommends \\
libc6 && \\
rm –rf /var/lib/apt/lists/*

# 创建目录
RUN mkdir -p /data/redis && \\
mkdir -p /redis

# 从构建阶段复制文件
COPY –from=buildstage /redis /redis

# 暴露端口
EXPOSE 6379

# 设置启动命令
ENTRYPOINT ["/redis/redis-server", "/redis/redis.conf"]

• 这里在第一个ubuntu环境进行解压下载编译等；得到可执行程序然后搞到第二个ubuntu里面（由于原先包解压自带对应运行环境，因此第二个容器还需要下对应环境）；实现了多级构建效果；大大减少了对应容器占用空间情况。

docker-compose.yml:

services:
redis01:
image: myredis:v1.0
build: ./redis
container_name: redis01
ports:
– "6379:6379"
healthcheck:
test: ["CMD", "/redis/redis-cli", "ping"]
interval: 10s
timeout: 5s
retries: 10

redis02:
image: myredis:v1.0
container_name: redis02
healthcheck:
test: ["CMD", "/redis/redis-cli", "ping"]
interval: 10s
timeout: 5s
retries: 10

redis03:
image: myredis:v1.0
container_name: redis03
healthcheck:
test: ["CMD", "/redis/redis-cli", "ping"]
interval: 10s
timeout: 5s
retries: 10

redis04:
image: myredis:v1.0
container_name: redis04
healthcheck:
test: ["CMD", "/redis/redis-cli", "ping"]
interval: 10s
timeout: 5s
retries: 10

redis05:
image: myredis:v1.0
container_name: redis05
healthcheck:
test: ["CMD", "/redis/redis-cli", "ping"]
interval: 10s
timeout: 5s
retries: 10

redis06:
image: myredis:v1.0
container_name: redis06
healthcheck:
test: ["CMD", "/redis/redis-cli", "ping"]
interval: 10s
timeout: 5s
retries: 10

redis07:
image: myredis:v1.0
container_name: redis07
entrypoint: >
/redis/redis–cli —cluster create
redis01:6379 redis02:6379 redis03:6379
redis04:6379 redis05:6379 redis06:6379
—cluster–replicas 1 –a 123456 —cluster–yes
depends_on:
redis01:
condition: service_healthy
redis02:
condition: service_healthy
redis03:
condition: service_healthy
redis04:
condition: service_healthy
redis05:
condition: service_healthy
redis06:
condition: service_healthy

• 之前build的镜像搞6个redis容器（也可以使用docker compose build直接构建+启动）；然后再让第七个把他们搞成集群服务再退出。

演示下效果：

• 可以发现对应先启动123456全部健康后才启动的7。

• 镜像构建成功+容器成功启动。

• 可以成功插入数据；发现被重定向到对应槽位所在的主节点了。

• 主从关系也正常。

• 停止服务（先停止的7然后剩下的接着）。

二.基于dockerfile编写C++微服务

• 基于nginx容器反向代理对应C++微服务：通过主机port映射到nginx容器的80端口；然后根据nginx的conf反向把对应请求发给对应同一compose网络里的cpp程序容器的8081端口；然后它就接收到请求把对应html发给nginx80端口；最后通过映射发给对应浏览器。

步骤如下：

#mermaid-svg-yu3TZDzZ5b3qyLlG{font-family:\”trebuchet ms\”,verdana,arial,sans-serif;font-size:16px;fill:#333;}@keyframes edge-animation-frame{from{stroke-dashoffset:0;}}@keyframes dash{to{stroke-dashoffset:0;}}#mermaid-svg-yu3TZDzZ5b3qyLlG .edge-animation-slow{stroke-dasharray:9,5!important;stroke-dashoffset:900;animation:dash 50s linear infinite;stroke-linecap:round;}#mermaid-svg-yu3TZDzZ5b3qyLlG .edge-animation-fast{stroke-dasharray:9,5!important;stroke-dashoffset:900;animation:dash 20s linear infinite;stroke-linecap:round;}#mermaid-svg-yu3TZDzZ5b3qyLlG .error-icon{fill:#552222;}#mermaid-svg-yu3TZDzZ5b3qyLlG .error-text{fill:#552222;stroke:#552222;}#mermaid-svg-yu3TZDzZ5b3qyLlG .edge-thickness-normal{stroke-width:1px;}#mermaid-svg-yu3TZDzZ5b3qyLlG .edge-thickness-thick{stroke-width:3.5px;}#mermaid-svg-yu3TZDzZ5b3qyLlG .edge-pattern-solid{stroke-dasharray:0;}#mermaid-svg-yu3TZDzZ5b3qyLlG .edge-thickness-invisible{stroke-width:0;fill:none;}#mermaid-svg-yu3TZDzZ5b3qyLlG .edge-pattern-dashed{stroke-dasharray:3;}#mermaid-svg-yu3TZDzZ5b3qyLlG .edge-pattern-dotted{stroke-dasharray:2;}#mermaid-svg-yu3TZDzZ5b3qyLlG .marker{fill:#333333;stroke:#333333;}#mermaid-svg-yu3TZDzZ5b3qyLlG .marker.cross{stroke:#333333;}#mermaid-svg-yu3TZDzZ5b3qyLlG svg{font-family:\”trebuchet ms\”,verdana,arial,sans-serif;font-size:16px;}#mermaid-svg-yu3TZDzZ5b3qyLlG p{margin:0;}#mermaid-svg-yu3TZDzZ5b3qyLlG .label{font-family:\”trebuchet ms\”,verdana,arial,sans-serif;color:#333;}#mermaid-svg-yu3TZDzZ5b3qyLlG .cluster-label text{fill:#333;}#mermaid-svg-yu3TZDzZ5b3qyLlG .cluster-label span{color:#333;}#mermaid-svg-yu3TZDzZ5b3qyLlG .cluster-label span p{background-color:transparent;}#mermaid-svg-yu3TZDzZ5b3qyLlG .label text,#mermaid-svg-yu3TZDzZ5b3qyLlG span{fill:#333;color:#333;}#mermaid-svg-yu3TZDzZ5b3qyLlG .node rect,#mermaid-svg-yu3TZDzZ5b3qyLlG .node circle,#mermaid-svg-yu3TZDzZ5b3qyLlG .node ellipse,#mermaid-svg-yu3TZDzZ5b3qyLlG .node polygon,#mermaid-svg-yu3TZDzZ5b3qyLlG .node path{fill:#ECECFF;stroke:#9370DB;stroke-width:1px;}#mermaid-svg-yu3TZDzZ5b3qyLlG .rough-node .label text,#mermaid-svg-yu3TZDzZ5b3qyLlG .node .label text,#mermaid-svg-yu3TZDzZ5b3qyLlG .image-shape .label,#mermaid-svg-yu3TZDzZ5b3qyLlG .icon-shape .label{text-anchor:middle;}#mermaid-svg-yu3TZDzZ5b3qyLlG .node .katex path{fill:#000;stroke:#000;stroke-width:1px;}#mermaid-svg-yu3TZDzZ5b3qyLlG .rough-node .label,#mermaid-svg-yu3TZDzZ5b3qyLlG .node .label,#mermaid-svg-yu3TZDzZ5b3qyLlG .image-shape .label,#mermaid-svg-yu3TZDzZ5b3qyLlG .icon-shape .label{text-align:center;}#mermaid-svg-yu3TZDzZ5b3qyLlG .node.clickable{cursor:pointer;}#mermaid-svg-yu3TZDzZ5b3qyLlG .root .anchor path{fill:#333333!important;stroke-width:0;stroke:#333333;}#mermaid-svg-yu3TZDzZ5b3qyLlG .arrowheadPath{fill:#333333;}#mermaid-svg-yu3TZDzZ5b3qyLlG .edgePath .path{stroke:#333333;stroke-width:2.0px;}#mermaid-svg-yu3TZDzZ5b3qyLlG .flowchart-link{stroke:#333333;fill:none;}#mermaid-svg-yu3TZDzZ5b3qyLlG .edgeLabel{background-color:rgba(232,232,232, 0.8);text-align:center;}#mermaid-svg-yu3TZDzZ5b3qyLlG .edgeLabel p{background-color:rgba(232,232,232, 0.8);}#mermaid-svg-yu3TZDzZ5b3qyLlG .edgeLabel rect{opacity:0.5;background-color:rgba(232,232,232, 0.8);fill:rgba(232,232,232, 0.8);}#mermaid-svg-yu3TZDzZ5b3qyLlG .labelBkg{background-color:rgba(232, 232, 232, 0.5);}#mermaid-svg-yu3TZDzZ5b3qyLlG .cluster rect{fill:#ffffde;stroke:#aaaa33;stroke-width:1px;}#mermaid-svg-yu3TZDzZ5b3qyLlG .cluster text{fill:#333;}#mermaid-svg-yu3TZDzZ5b3qyLlG .cluster span{color:#333;}#mermaid-svg-yu3TZDzZ5b3qyLlG div.mermaidTooltip{position:absolute;text-align:center;max-width:200px;padding:2px;font-family:\”trebuchet ms\”,verdana,arial,sans-serif;font-size:12px;background:hsl(80, 100%, 96.2745098039%);border:1px solid #aaaa33;border-radius:2px;pointer-events:none;z-index:100;}#mermaid-svg-yu3TZDzZ5b3qyLlG .flowchartTitleText{text-anchor:middle;font-size:18px;fill:#333;}#mermaid-svg-yu3TZDzZ5b3qyLlG rect.text{fill:none;stroke-width:0;}#mermaid-svg-yu3TZDzZ5b3qyLlG .icon-shape,#mermaid-svg-yu3TZDzZ5b3qyLlG .image-shape{background-color:rgba(232,232,232, 0.8);text-align:center;}#mermaid-svg-yu3TZDzZ5b3qyLlG .icon-shape p,#mermaid-svg-yu3TZDzZ5b3qyLlG .image-shape p{background-color:rgba(232,232,232, 0.8);padding:2px;}#mermaid-svg-yu3TZDzZ5b3qyLlG .icon-shape .label rect,#mermaid-svg-yu3TZDzZ5b3qyLlG .image-shape .label rect{opacity:0.5;background-color:rgba(232,232,232, 0.8);fill:rgba(232,232,232, 0.8);}#mermaid-svg-yu3TZDzZ5b3qyLlG .label-icon{display:inline-block;height:1em;overflow:visible;vertical-align:-0.125em;}#mermaid-svg-yu3TZDzZ5b3qyLlG .node .label-icon path{fill:currentColor;stroke:revert;stroke-width:revert;}#mermaid-svg-yu3TZDzZ5b3qyLlG :root{–mermaid-font-family:\”trebuchet ms\”,verdana,arial,sans-serif;}

1. 编写代码

2. 编写微服务Dockerfile

3. 配置Nginx负载均衡编写Nginx Dockerfile

4. 编写docker-compose.yml构建镜像

5. 启动服务

6. 测试功能

对应工程目录：

root@iv-ye1i2elts0wh2yp1ahah:/home/youxing/docker-learn/C++_micro_server# tree
.
├── cppweb
│?? ├── dockerfile
│?? ├── main.cpp
│?? └── mycppweb
├── docker-compose.yml
└── nginx
├── default.conf
└── dockerfile

2 directories, 6 files

c++代码：

#include <iostream>
#include <netinet/in.h>
#include <string.h>
#include <assert.h>
#include <unistd.h>
#include <pthread.h>

using namespace std;

struct pthread_data {
struct sockaddr_in client_addr;
int sock_fd;
};

void *serverForClient(void *arg);

int main() {
int socket_fd;
int conn_fd;
int res;
socklen_t len = sizeof(sockaddr_in);

struct sockaddr_in sever_add;
memset(&sever_add, 0, sizeof(sever_add));
sever_add.sin_family = AF_INET;
sever_add.sin_port = htons(8081);
sever_add.sin_addr.s_addr = htonl(INADDR_ANY);

socket_fd = socket(AF_INET, SOCK_STREAM, 0);
assert(socket_fd > 0);

int option = 1;
setsockopt(socket_fd, SOL_SOCKET, SO_REUSEADDR, &option, sizeof(option));

res = bind(socket_fd, (struct sockaddr*)&sever_add, len);
assert(res != –1);

res = listen(socket_fd, 5);
assert(res != –1);

cout << "server init" << endl;

while(1) {
struct sockaddr_in client;
socklen_t client_len = sizeof(client);

conn_fd = accept(socket_fd, (struct sockaddr *)&client, &client_len);
if (conn_fd < 0) {
perror("accept");
continue;
}

pthread_data *pdata = new pthread_data;
pdata->client_addr = client;
pdata->sock_fd = conn_fd;
std::cout<<"in "<<conn_fd<<endl;
pthread_t pt;
pthread_create(&pt, NULL, serverForClient, (void*)pdata);
pthread_detach(pt);
}

close(socket_fd);
return 0;
}

void *serverForClient(void *arg) {
pthread_data *pdata = (pthread_data*)arg;
int conn_fd = pdata->sock_fd;
std::cout<<"process "<<conn_fd<<endl;
char request[1024];
int len = recv(conn_fd, request, sizeof(request)–1, 0);
if (len <= 0) {
close(conn_fd);
delete pdata;
return nullptr;
}
request[len] = '\\0';

const char* http_header = "HTTP/1.1 200 OK\\r\\n"
"Content-Type: text/html\\r\\n"
"Connection: close\\r\\n\\r\\n";

send(conn_fd, http_header, strlen(http_header), 0);

const char* html_content = "<!DOCTYPE html>\\n"
"<html>\\n"
"<head>\\n"
"<title>Welcome to C++ web server!</title>\\n"
"<style>\\n"
" body { width: 35em; margin: 0 auto;\\n"
" font-family: Tahoma, Verdana, Arial, sans-serif; }\\n"
"</style>\\n"
"</head>\\n"
"<body>\\n"
"<h1>Welcome to C++ web server!</h1>\\n"
"<p>If you see this page, the web server is successfully working.</p>\\n"
"</body>\\n"
"</html>";

send(conn_fd, html_content, strlen(html_content), 0);
close(conn_fd);
delete pdata;
return nullptr;
}

• 这里绑定默认8081端口（如果容器里进行就绑定容器的8081端口）；然后接受到请求就发送对应html。

默认nginx的conf：

# Nginx 负载均衡配置
upstream backend {
server mycppweb:8081 weight=1;
server mycppweb2:8081 weight=2;
}

server {
listen 80;
access_log off;

location / {
proxy_pass http://backend;
proxy_set_header Host $host;
proxy_set_header X–Real–IP $remote_addr;
}
}

• 这里默认进行反向代理；当访问到它的80号端口就转向对应的名字为mycppweb的服务对应的容器的8081端口。

总的docker-compose.yml就是启动对应两服务三容器：

services:
web:
image: mynginx:v3.0
build:
context: ./nginx
ports:
– "8112:80"
depends_on:
mycppweb:
condition: service_started

mycppweb:
build:
context: ./cppweb
image: mycppweb:v2.0

mycppweb2:
image: mycppweb:v2.0

对于cpp服务的dockerfile的c++程序编译及运行还是采取多级构建进行只保留exe的容器j镜像构建：

# 第一阶段：构建阶段（编译环境）
FROM ubuntu:22.04 as buildstage

# 配置国内软件源并更新
RUN sed –i 's@//.*archive.ubuntu.com@//mirrors.ustc.edu.cn@g' /etc/apt/sources.list && apt update

# 安装编译工具链
RUN apt install –y build–essential

# 设置工作目录并复制源码
WORKDIR /src
COPY ./main.cpp .

# 编译程序（链接pthread库）
RUN g++ main.cpp –o mycppweb –lpthread

# 指定运行时命令s
CMD ["/src/mycppweb"]

# 第二阶段：运行阶段（生产环境）
FROM ubuntu:22.04

# 从构建阶段仅复制最终可执行文件
COPY —from=buildstage /src/mycppweb /

# 指定容器启动命令
CMD ["/mycppweb"]

下面启动下：

• 构建对应的镜像。

• 容器构建成功。

• 容器正常启动，下面进行访问（打开对应的cpp服务日志监控）。

• 不停的访问对应网站；因为设置的是nginx根据web1:web2发送率是1:2的权重来进行请求分配；通过观察日志可以发现。

三.docker镜像制作常见问题

• 此类镜像已无用，可安全删除。使用 docker image ls -f dangling=true 查看，并用 prune 命令清理。

区别：

总结：

“它们的名称和表情都是<none>，但虚悬镜像是没用的垃圾（可删），中间层镜像是有用的基石（不可删）。” 所以，不能仅凭 REPOSITORY 和 TAG 字段为 <none> 就删除镜像，必须先用 docker image ls -f dangling=true 确认它是真正的“虚悬镜像”后再进行清理。

四.本篇小结

本篇通过完整项目案例掌握Docker生产级应用部署，深入理解镜像分层机制与资源优化策略，提升容器化开发与运维效率。