在 ubuntu22.04 中 安装 EdgeX 并运行 device-sdk-c sample

EdgeX Foundry 作为一个开源的边缘计算框架，为开发者提供了一个灵活、可扩展的平台，用于构建和管理边缘设备及服务。然而，对于初次接触 EdgeX 的用户来说，从零开始安装并运行其相关组件（如 device-sdk-c 示例）可能会面临不少挑战。

无论是环境的配置、依赖的安装，还是编译与调试的过程，每一步都需要细致的操作和清晰的思路。本文将以 Ubuntu 22.04 为基础平台，详细记录如何从克隆 EdgeX 源码到成功运行 device-sdk-c 的 sample 示例的全过程。希望通过这篇博客，不仅能为新手提供一份实用的操作指南，也能让有经验的开发者在遇到问题时找到解决思路。

1.克隆源码-EdgeX

在开始安装 EdgeX 之前，我们需要先获取其源码，而 GitHub 是获取最新版本的最佳途径。为了让后续操作更加有条理，我们首先在用户主目录下创建一个专属的工作目录，用于存放 EdgeX 相关的文件。以下是具体操作：

• 打开终端，输入以下命令创建一个名为 edgex 的文件夹，并进入该目录：

  mkdir edgex && cd edgex
  

• 接下来，我们需要从 GitHub 上克隆 EdgeX 的 Compose 文件仓库，这是运行 EdgeX 服务的核心配置文件集合。执行以下命令：

  git clone https://github.com/edgexfoundry/edgex-compose.git
  cd edgex-compose
  

  结果显示如下：

  root@ubuntu:/home/ubuntu# mkdir edgex && cd edgex
  root@ubuntu:/home/ubuntu/edgex# git clone https://github.com/edgexfoundry/edgex-compose.git
  Cloning into 'edgex-compose'...
  remote: Enumerating objects: 5826, done.
  remote: Counting objects: 100% (3563/3563), done.
  remote: Compressing objects: 100% (319/319), done.
  remote: Total 5826 (delta 3504), reused 3250 (delta 3244), pack-reused 2263 (from 3)
  Receiving objects: 100% (5826/5826), 1.59 MiB | 1.76 MiB/s, done.
  Resolving deltas: 100% (4949/4949), done.
  root@ubuntu:/home/ubuntu/edgex# ls
  edgex-compose
  

  输出信息显示了克隆进度，最终 ls 命令确认目录中已存在 edgex-compose 文件夹，说明这一步成功完成。

2.安装工具

EdgeX 的部署离不开 Docker，因为它通过容器化技术简化了服务的管理和运行。因此，我们需要先在 Ubuntu 系统中安装 Docker。以下是详细的安装步骤：

• 使用官方脚本安装 Docker，并指定阿里云镜像源以加速下载：

  curl -fsSL https://get.docker.com -o get-docker.sh
  sh get-docker.sh --mirror Aliyun
  docker -v
  

  第一条命令通过 curl 下载安装脚本并保存为 get-docker.sh 文件；第二条命令执行脚本，--mirror Aliyun 参数指定使用阿里云镜像源，避免默认源下载缓慢；最后通过 docker -v 检查版本，确保安装成功。安装过程的输出类似：

  root@ubuntu:/home/ubuntu# curl -fsSL https://get.docker.com -o get-docker.sh
  root@ubuntu:/home/ubuntu# sh get-docker.sh --mirror Aliyun
  # Executing docker install script, commit: 4c94a56999e10efcf48c5b8e3f6afea464f9108e
  ...
  Client: Docker Engine - Community
   Version:           28.0.1
   API version:       1.48
   ...
  root@ubuntu:/home/ubuntu# docker -v
  Docker version 28.0.1, build 068a01e
  

• 如果下载速度过慢，可能是镜像源的问题。这时可以调整 Ubuntu 的软件源以提高效率。首先检查系统版本：

  root@ubuntu:/home/ubuntu# more /etc/os-release
  PRETTY_NAME="Ubuntu 22.04.5 LTS"
  NAME="Ubuntu"
  VERSION_ID="22.04"
  VERSION="22.04.5 LTS (Jammy Jellyfish)"
  VERSION_CODENAME=jammy
  ID=ubuntu
  ID_LIKE=debian
  HOME_URL="https://www.ubuntu.com/"
  SUPPORT_URL="https://help.ubuntu.com/"
  BUG_REPORT_URL="https://bugs.launchpad.net/ubuntu/"
  PRIVACY_POLICY_URL="https://www.ubuntu.com/legal/terms-and-policies/privacy-poli
  cy"
  UBUNTU_CODENAME=jammy
  

  • 确认版本为 jammy 后，访问南京大学镜像站（https://mirror.nju.edu.cn/mirrorz-help/ubuntu/），获取对应源配置，并编辑 /etc/apt/sources.list：

  # vim /etc/apt/sources.list 
  
  deb https://mirror.nju.edu.cn/ubuntu/ jammy main restricted universe multiverse
  deb https://mirror.nju.edu.cn/ubuntu/ jammy-updates main restricted universe multiverse
  deb https://mirror.nju.edu.cn/ubuntu/ jammy-backports main restricted universe multiverse
  deb http://security.ubuntu.com/ubuntu/ jammy-security main restricted universe multiverse
  

• 启动 docker 并查看状态：

  root@ubuntu:/home/ubuntu# systemctl restart docker
  root@ubuntu:/home/ubuntu# systemctl status docker
  ● docker.service - Docker Application Container Engine
       Loaded: loaded (/lib/systemd/system/docker.service; enabled; vendor preset>
       Active: active (running) since Thu 2025-03-13 12:21:05 CST; 6s ago
  TriggeredBy: ● docker.socket
         Docs: https://docs.docker.com
     Main PID: 28525 (dockerd)
        Tasks: 10
       Memory: 22.1M
          CPU: 1.228s
       CGroup: /system.slice/docker.service
               └─28525 /usr/bin/dockerd -H fd:// --containerd=/run/containerd/con>
  

  Active: active (running) 表示 Docker 已正常运行，为后续 EdgeX 部署奠定了基础。

3.编译运行-EdgeX

有了 Docker 支持后，我们可以开始部署 EdgeX 服务。EdgeX 使用 Docker Compose 管理多个容器，因此需要进入 compose-builder 目录执行构建和运行命令：

• 进入目录并运行无安全模式的 EdgeX：

  root@ubuntu:/home/ubuntu# cd /home/ubuntu/edgex/edgex-compose/compose-builder/
  root@ubuntu:/home/ubuntu/edgex/edgex-compose/compose-builder# make run no-secty
  echo MQTT_VERBOSE=
  MQTT_VERBOSE=
  docker compose  -p edgex -f docker-compose-base.yml -f add-postgres.yml -f add-mqtt-messagebus.yml -f add-mqtt-broker-mosquitto.yml convert --output docker-compose.yml.gen
  if [ "0" = "0" ]; then rm -rf ./gen_ext_compose; fi
  echo "# Generated with: Docker Compose version v2.33.1" > docker-compose.yml
  cat docker-compose.yml.gen >> docker-compose.yml
  rm docker-compose.yml.gen
  docker compose  -p edgex up -d 
  [+] Running 12/12
   ✔ Container edgex-ui-go                            Running                0.0s 
   ✔ Container edgex-postgres                         Running                0.0s 
   ✔ Container edgex-mqtt-broker                      Running                0.0s 
   ✔ Container edgex-core-keeper                      Running                0.0s 
   ✔ Container edgex-core-metadata                    Running                0.0s 
   ✔ Container edgex-core-common-config-bootstrapper  Running                0.0s 
   ✔ Container edgex-app-rules-engine                 Running                0.0s 
   ✔ Container edgex-support-scheduler                Running                0.0s 
   ✔ Container edgex-core-data                        Running                0.0s 
   ✔ Container edgex-kuiper                           Running                0.0s 
   ✔ Container edgex-support-notifications            Running                0.0s 
   ✔ Container edgex-core-command                     Started                0.4s 
  

  这表明所有容器已启动，包括 UI、数据库和核心服务等。

• 查看容器运行状态，显示如下，每个容器显示 Up 状态，说明 EdgeX 已成功运行：

  root@ubuntu:/home/ubuntu/edgex/edgex-compose/compose-builder# docker compose ps
  NAME                                    IMAGE                                                                  COMMAND                  SERVICE                           CREATED         STATUS         PORTS
  edgex-app-rules-engine                  nexus3.edgexfoundry.org:10004/app-service-configurable:latest          "/app-service-config…"   app-rules-engine                  7 minutes ago   Up 7 minutes   48095/tcp, 127.0.0.1:59701->59701/tcp
  edgex-core-command                      nexus3.edgexfoundry.org:10004/core-command:latest                      "/core-command --reg…"   core-command                      7 minutes ago   Up 5 minutes   127.0.0.1:59882->59882/tcp
  edgex-core-common-config-bootstrapper   nexus3.edgexfoundry.org:10004/core-common-config-bootstrapper:latest   "entrypoint.sh /core…"   core-common-config-bootstrapper   7 minutes ago   Up 7 minutes   
  edgex-core-data                         nexus3.edgexfoundry.org:10004/core-data:latest                         "/core-data --regist…"   core-data                         7 minutes ago   Up 7 minutes   127.0.0.1:59880->59880/tcp
  edgex-core-keeper                       nexus3.edgexfoundry.org:10004/core-keeper:latest                       "/core-keeper"           core-keeper                       7 minutes ago   Up 7 minutes   127.0.0.1:59890->59890/tcp
  edgex-core-metadata                     nexus3.edgexfoundry.org:10004/core-metadata:latest                     "/core-metadata ...
  

4.克隆源码-SDK-C

为了运行 device-sdk-c 的示例，我们还需要获取其源码。device-sdk-c 包含了丰富的 C 语言示例，是开发自定义设备服务的好起点：

• 在 /edgex 目录下克隆仓库：

  root@ubuntu:/home/ubuntu/edgex/edgex-compose/compose-builder# cd ../..
  root@ubuntu:/home/ubuntu/edgex# ls
  edgex-compose
  root@ubuntu:/home/ubuntu/edgex# git clone https://github.com/edgexfoundry/device-sdk-c.git
  Cloning into 'device-sdk-c'...
  remote: Enumerating objects: 5641, done.
  remote: Counting objects: 100% (527/527), done.
  remote: Compressing objects: 100% (213/213), done.
  remote: Total 5641 (delta 419), reused 337 (delta 307), pack-reused 5114 (from 4)
  Receiving objects: 100% (5641/5641), 1.93 MiB | 1.70 MiB/s, done.
  Resolving deltas: 100% (3210/3210), done.
  

5.添加第三方依赖库

编译 device-sdk-c 需要一系列依赖库，参考官方 README（https://github.com/edgexfoundry/device-sdk-c/blob/main/README.md）

• A Linux build host
• A version of GCC supporting C11.
• CMake version 3 or greater and make.
• Development libraries and headers for:
  • curl (version 7.56 or later)
  • microhttpd (version 0.9)
  • libyaml (version 0.1.6 or later)
  • libcbor (version 0.5)
  • paho (version 1.3.x)
  • libuuid (from util-linux v2.x)
  • IOTech’s C utilities package (version 1.5)

apt-get install libcurl4-openssl-dev libmicrohttpd-dev libyaml-dev libcbor-dev libpaho-mqtt-dev
apt-get install lsb-release apt-transport-https curl gnupg
curl -fsSL https://iotech.jfrog.io/artifactory/api/gpg/key/public | gpg --dearmor -o /usr/share/keyrings/iotech.gpg
echo "deb [signed-by=/usr/share/keyrings/iotech.gpg] https://iotech.jfrog.io/iotech/debian-release $(lsb_release -cs) main" | tee -a /etc/apt/sources.list.d/iotech.list
apt-get update
apt-get install iotech-iot-1.5-dev
apt install uuid-dev

6.编译-C SDK

在C SDK 顶层目录下, 运行 make，最终结果显示如下：

root@ubuntu:/home/ubuntu/edgex/device-sdk-c# make

...
[100%] Built target device-file

7.执行 random 设备服务示例

• 将编译好的可执行文件 device-random 复制到以下源文件目录，因为源文件目录提供了/res 里面包含预设的配置文件：

  cp device-sdk-c/build/release/c/examples/random/device-random  device-sdk-c/src/c/examples/random/
  cd /home/ubuntu/edgex/device-sdk-c/src/c/examples/random/
  

• 执行以下命令：-o 表示覆盖默认配置，-cp 指定配置提供者

  ./device-random -o -cp=keeper.http://localhost:59890
  

  如果遇到报错，按以下方法解决：

  • 安全 Token 报错：设置环境变量禁用安全服务：

    • root@ubuntu:/home/ubuntu/edgex/device-sdk-c/src/c/examples/random# ./device-random -o -cp=keeper.http://localhost:59890
      level=INFO ts=2025-03-13T06:44:41Z app=device-random msg="iot_threadpool_alloc (threads: 8 max_jobs: 0 default_priority: -1 affinity: -1)"
      level=INFO ts=2025-03-13T06:44:41Z app=device-random msg="iot_scheduler_alloc (priority: -1 affinity: -1)"
      level=ERROR ts=2025-03-13T06:44:41Z app=device-random msg="vault: Token file /tmp/edgex/secrets/device-random/secrets-token.json doesn't parse as JSON"
      Error: 4: Configuration is invalid
      

    • 解决：需要关闭Edges安全服务：

      参考https://docs.edgexfoundry.org/4.0/microservices/configuration/CommonEnvironmentVariables/#edgex_security_secret_store

      export EDGEX_SECURITY_SECRET_STORE="false"
      

  • Curl 解析失败：编辑 /etc/hosts：

    • level=DEBUG ts=2025-03-13T06:45:41Z app=device-random msg="mqtt: subscribing to edgex/configs/edgex/v4/core-common-config-bootstrapper/all-services/#"
      level=DEBUG ts=2025-03-13T06:45:41Z app=device-random msg="mqtt: subscribing to edgex/configs/edgex/v4/core-common-config-bootstrapper/device-services/#"
      level=ERROR ts=2025-03-13T06:45:42Z app=device-random msg="Curl failed with code 6 (Couldn't resolve host name)"
      level=ERROR ts=2025-03-13T06:45:43Z app=device-random msg="Curl failed with code 6 (Couldn't resolve host name)"
      level=ERROR ts=2025-03-13T06:45:44Z app=device-random msg="Curl failed with code 6 (Couldn't resolve host name)"
      level=ERROR ts=2025-03-13T06:45:45Z app=device-random msg="Curl failed with code 6 (Couldn't resolve host name)"
      level=ERROR ts=2025-03-13T06:45:46Z app=device-random msg="Curl failed with code 6 (Couldn't resolve host name)"
      level=ERROR ts=2025-03-13T06:45:47Z app=device-random msg="Curl failed with code 6 (Couldn't resolve host name)"
      

    • 解决：在/etc/hosts 中添加127.0.0.1 edgex-core-metadata：

      /etc/hosts
      
      127.0.0.1 localhost
      127.0.0.1 edgex-core-metadata
      # The following lines are desirable for IPv6 capable hosts
      ::1 ip6-localhost ip6-loopback
      fe00::0 ip6-localnet
      ff00::0 ip6-mcastprefix
      ff02::1 ip6-allnodes
      ff02::2 ip6-allrouters
      ff02::3 ip6-allhosts
      

  • 健康检查间隔缺失：修改 res/configuration.yml：

    • level=DEBUG ts=2025-03-13T06:50:39Z app=device-random msg="HTTP response: 404"
      level=DEBUG ts=2025-03-13T06:50:39Z app=device-random msg="HTTP response: 400"
      level=ERROR ts=2025-03-13T06:50:39Z app=device-random msg="Register service failed: {"apiVersion":"v3","message":"AddRegistrationRequest.Registration.HealthCheck.Interval field is required","statusCode":400}
      "
      level=ERROR ts=2025-03-13T06:50:39Z app=device-random msg="Unable to register service in registry"
      Error: 21: HTTP request failed
      level=DEBUG ts=2025-03-13T06:50:39Z app=device-random msg="Scheduler thread stopping"
      

    • 解决：在res/configration.yml 中添加 HealthCheckInterval: 10s：

      Writable:
        LogLevel: DEBUG
      
      Service:
        Host: localhost
        Port: 59999
        StartupMsg: Example random device service started
        HealthCheckInterval: 10s
      

• 如果程序运行成功，将显示如下：

  ...
  level=INFO ts=2025-03-13T07:36:45Z app=device-random correlation-id=c26ef1f7-755c-4952-8091-5c7c436e039b msg="AutoEvent: RandomDevice1/SensorOne"
  level=INFO ts=2025-03-13T07:36:50Z app=device-random correlation-id=dc222997-3113-465b-9c5c-26829ce24901 msg="AutoEvent: RandomDevice1/SensorTwo"
  level=INFO ts=2025-03-13T07:36:55Z app=device-random correlation-id=049d560d-896a-4e5a-8101-f3364c2767f1 msg="AutoEvent: RandomDevice1/SensorOne"
  

• 打开浏览器进入：localhost:4000 可以看见一个设备服务，并能成功读取随机数，如下所示：

  
  

总结

通过以上步骤，我们成功地在 Ubuntu 22.04 上完成了 EdgeX 的安装，并运行了 device-sdk-c 的 random 设备服务示例。从源码的克隆到 Docker 环境的搭建，再到依赖库的安装和配置文件的调整，每一个环节都环环相扣，考验着我们的耐心和解决问题的能力。

过程中可能遇到的报错，如安全性 token 问题、Curl 解析失败或健康检查间隔配置缺失，也提醒我们在操作复杂系统时，细节的重要性不容忽视。

最终，当我们在浏览器中看到 localhost:4000 界面上显示的设备服务和随机数据时，那份成就感无疑是对所有努力的最好回报。这不仅是一个技术实现的过程，更是一次对 EdgeX 架构和边缘计算理念的深入学习之旅。未来，我们还可以基于此基础，进一步扩展功能、优化性能，让边缘计算的潜力在更多场景中得以释放！