「五」DOCKER源码分析5 daemon端对container start的处理

func (s *containerRouter) postContainersStart(ctx context.Context, w http.ResponseWriter, r *http.Request, vars map[string]string) error {
	//获取hostConfig配置信息
	var hostConfig *container.HostConfig
	checkpoint := r.Form.Get("checkpoint")
	checkpointDir := r.Form.Get("checkpoint-dir")
	//调用ContainerStart进一步启动容器，2.详细分析
	err := s.backend.ContainerStart(vars["name"], hostConfig, checkpoint, checkpointDir)
}

func (daemon *Daemon) ContainerStart(name string, hostConfig *containertypes.HostConfig, checkpoint string, checkpointDir string) error {
	//可以根据全容器ID、容器名、容器ID前缀获取容器对象
	container, err := daemon.GetContainer(name)
	//调用containerStart进一步启动容器，3.详细分析
	daemon.containerStart(container, checkpoint, checkpointDir, true)
}

func (daemon *Daemon) containerStart(container *container.Container, checkpoint string, checkpointDir string, resetRestartManager bool) (err error) {
	//在创建的时候调用了一次，这次start又mount一次？
	err := daemon.conditionalMountOnStart(container)
	//3.1对网络进行初始化，后面再详细分析
	err := daemon.initializeNetworking(container)
	//调用daemon.containerd.Create()进一步启动，后面4.再详细分析
	err := daemon.containerd.Create(container.ID, checkpoint, checkpointDir, *spec, container.InitializeStdio, createOptions...)
}

func (daemon *Daemon) initializeNetworking(container *container.Container) error {
	//如果是container模式
	if container.HostConfig.NetworkMode.IsContainer() {
		//获取所要共享网络的container的hosts文件
		nc, err := daemon.getNetworkedContainer(container.ID, container.HostConfig.NetworkMode.ConnectedContainer())
		//将所要共享网络的container的HostnamePath、HostsPath和ResolvConfPath赋值给新container
		initializeNetworkingPaths(container, nc)
		//将所要共享网络的container的Hostname、Domainname赋值给新container
		container.Config.Hostname = nc.Config.Hostname
		container.Config.Domainname = nc.Config.Domainname
		//container模式从这里就返回了，不执行后面的操作
		return nil
	}
	//如果是host模式
	if container.HostConfig.NetworkMode.IsHost() {
		if container.Config.Hostname == "" {
			//将宿主机的hostname赋值给新建container
			container.Config.Hostname, err = os.Hostname()
			//这里host没有返回，回往下继续执行
		}
	}
	//下面3.1.1再详细分析
	err := daemon.allocateNetwork(container)
	//生成container的hosts文件
	return container.BuildHostnameFile()
}

func (daemon *Daemon) allocateNetwork(container *container.Container) error {
	//如果没有使用网络，或者是container网络，则直接返回
	if container.Config.NetworkDisabled || container.HostConfig.NetworkMode.IsContainer() {
		return nil
	}
	
	networks := make(map[string]*network.EndpointSettings)
	for n, epConf := range container.NetworkSettings.Networks {
		if n == defaultNetName {
			continue
		}

		networks[n] = epConf
	}		
	for netName, epConf := range networks {
		cleanOperationalData(epConf)
		//下面再详细分析
		if err := daemon.connectToNetwork(container, netName, epConf.EndpointSettings, updateSettings); err != nil {
			return err
		}
	}
	//将hosts配置信息存到disk中
	err := container.WriteHostConfig()
}

func (daemon *Daemon) connectToNetwork(container *container.Container, idOrName string, endpointConfig *networktypes.EndpointSettings, updateSettings bool) (err error) {
	//查找需要连接的网络,n为libnetwork.Network类型，config为*networktypes.NetworkingConfig
	n, config, err := daemon.findAndAttachNetwork(container, idOrName, endpointConfig)
	//获取controller
	controller := daemon.netController
	//获得sandbox
	sb := daemon.getNetworkSandbox(container)
	//从给定的network，也就是n，创建一个endpoint options
	createOptions, err := container.BuildCreateEndpointOptions(n, endpointConfig, sb, daemon.configStore.DNS)
	//创建endpoin
	ep, err := n.CreateEndpoint(endpointName, createOptions...)
	//由network n 创建endpoint Join options
	joinOptions, err := container.BuildJoinOptions(n)
	//将endpoint接入sandbox
	err := ep.Join(sb, joinOptions...)
}

func (clnt *client) Create(containerID string, checkpoint string, checkpointDir string, spec specs.Spec, attachStdio StdioCallback, options ...CreateOption) (err error) {
	//生成一个libcontainerd.container对象
	container := clnt.newContainer(filepath.Join(dir, containerID), options...)
	//创建目录
	err := idtools.MkdirAllAs(container.dir, 0700, uid, gid)
	//生成文件config.json
	f, err := os.Create(filepath.Join(container.dir, configFilename))
	//调用container.start()进一步启动container，在5.中继续分析
	return container.start(checkpoint, checkpointDir, attachStdio)
}

func (ctr *container) start(checkpoint string, checkpointDir string, attachStdio StdioCallback) (err error) {
	
	//CreateContainerRequest对象，应该是保存一些容器信息给libcontainer start时使用
	r := &containerd.CreateContainerRequest{
		Id:            ctr.containerID,
		BundlePath:    ctr.dir,
		Stdin:         ctr.fifo(syscall.Stdin),
		Stdout:        ctr.fifo(syscall.Stdout),
		Stderr:        ctr.fifo(syscall.Stderr),
		Checkpoint:    checkpoint,
		CheckpointDir: checkpointDir,
		// check to see if we are running in ramdisk to disable pivot root
		NoPivotRoot: os.Getenv("DOCKER_RAMDISK") != "",
		Runtime:     ctr.runtime,
		RuntimeArgs: ctr.runtimeArgs,
	}	
	//发送CreateContainerRequest RPC请求给docker-containerd，这部分还不是很了解
	resp, err := ctr.client.remote.apiClient.CreateContainer(context.Background(), r)
}