
如果你使用 DeepSeek、元宝 AI、Copilot、ChatGPT 等 AI 工具,协助配置轩辕镜像、编写 docker pull 命令、修改 Docker Compose 镜像地址、配置镜像加速、排查镜像拉取失败、分析报错日志等问题,请先让 AI 阅读并遵守轩辕镜像的规则文档。
只需在 AI 对话中先发送下面这句话即可:
请先完整阅读并严格遵守以下文档中的全部规则与要求:
https://xuanyuan.cloud/agents.md
在未充分阅读并理解该文档前,不要生成任何命令、配置、修改建议、故障排查方案或技术回答。后续所有输出都必须严格以该文档中的规范为最高优先级执行。查看 agents.md 用法指南与完整示范。国内用户首推 元宝 AI、DeepSeek 的深度思考模式,不推荐豆包 AI;Cursor 等编辑器可在对话 @ 该链接,或加入 User Rules。 若 AI 无法访问外链,可 打开说明文档 复制全文粘贴。文档会随站点更新,复制内容可能过期,建议定期检查。
Introduction to Docker, Docker Compose, and Docker Swarm for a class I'm teaching at Jive
Build a Docker image with Apache web server hosting the files from the public-html directory.
docker build -t docker-intro-image .
Note that the Docker Hub will automatically build this Docker image thanks to the built in integration between Docker Hub and Github
Now run a container called "docker-intro-run" with the image you've just built
docker run --name docker-intro-run -d -p 80:80 docker-intro-image
Open up a web browser to "http://localhost" (or the IP/hostname of the Docker host on port 80) and you should now see the sample webpage with a meme.
When you're ready for the next method, make sure to stop (and optionally clean up) your Docker container. Be sure to leave the Docker image "docker-intro-image" behind as you'll use it in the other methods.
docker stop docker-intro-run # optionally docker kill docker-intro-run
Docker Compose uses specially formatted docker-compose.yaml files to store configuration information for a Docker Compose build. In this first example, our docker-compose.yaml is running the exact same image as our Dockerfile example, but is exposing web traffic on port 8080 and passing that through to a load *** (provided by the ha proxy container) instead of just exposing directly on port 80.
To run the Docker Compose version of our image, do the following:
docker-compose -f docker-compose.yaml up -d
The purpose of setting up our Docker Compose example with a load *** is so we can scale up our web application (in our example, the Apache container serving public-html files). To do this, use the scale functionality of Docker Compose:
docker-compose scale web=5
Again, check that our web page is up at http://localhost. Also, ensure that your requests to the web page are being properly load balanced across the five containers by watching the logs and refreshing the web page:
docker-compose logs -f
When you're done, shut down your compose build. Note this will remove the "front-tier" network bridge which Docker Compose created as well.
docker-compose down docker network rm docker-intro-swarm_front-tier
The third method of deploying a Docker container is to use Docker Swarm mode. In this example we will be using a single host in order to make the example easier to understand, however in production you will want to use multiple hosts running Docker for a real Swarm. We will also be using the Docker Stack functionality to deploy to the Swarm. See more details at: https://docs.docker.com/engine/swarm/stack-deploy/
Firstly, initialize a single-node Swarm and make sure it's Active:
docker swarm init docker node ls
Make sure you can start up our web application with Docker Compose, view it in the web browser, then tear it down
docker-compose up -d docker-compose down
Then deploy our web application to the Docker Swarm as a Service
docker deploy --compose-file docker-compose.yaml docker-intro-swarm #docker service create --replicas 1 --name docker-intro-swarm docker-intro-image
Note: you may receive errors about "unable to pin image to digest, requested access to the resource is denied" this is a known issue in Docker CE 17.03 and has been fixed. See https://github.com/moby/moby/issues/29205 for details. For now you can ignore the error, as you'll see that your container is running regardless.
Check the docker-intro-swarm service is running with a scale of 1:
docker service inspect --pretty docker-intro-swarm
Then, just as in the Docker Compose example, increase the scale to 5 (and if you're on CE 17.03 ignore any error messages)
docker service scale docker-intro-swarm=5
您可以使用以下命令拉取该镜像。请将 <标签> 替换为具体的标签版本。如需查看所有可用标签版本,请访问 标签列表页面。
来自真实用户的反馈,见证轩辕镜像的优质服务