docker live-restore

docker 1.12 开始支持live restore，就是关闭docker daemon ，而不关闭容器

有两种方式开启:

• 如果docker daemon正在运行并且你不想停止它，你可以添加配置到docker daemon的配置文件。例如：在linux系统上默认的配置文件是/etc/docker/daemon.json

{"live-restore": true}

你必须传递一个SIGHUP信号给daemon进程来重载配置。更多有关使用config.json来配置docker daemon的信息，可以参考 daemon configuration file

• 在使用dockerd启动时指定--live-restore选项

$ sudo dockerd --live-restore

root@btsci-cloud-server-1:~# ps -ef|grep docker

root 5773 1 1 14:09 ? 00:00:00 /usr/bin/dockerd -H tcp://127.0.0.1:4243 -H unix:///var/run/docker.sock --bip=192.168.2.1/24 --mtu=1450

root 5785 5773 0 14:09 ? 00:00:00 docker-containerd -l unix:///var/run/docker/libcontainerd/docker-containerd.sock --metrics-interval=0 --start-timeout 2m --state-dir /var/run/docker/libcontainerd/containerd --shim docker-containerd-shim --runtime docker-runc

root 5982 5785 0 14:09 ? 00:00:00 docker-containerd-shim e4b6cbd3ea87347a8f6e8d640ea0350b5a35424905633b08a374e0433251f02c /var/run/docker/libcontainerd/e4b6cbd3ea87347a8f6e8d640ea0350b5a35424905633b08a374e0433251f02c docker-runc

root 6087 5699 0 14:09 pts/2 00:00:00 grep --color=auto docker

root@btsci-cloud-server-1:~#

root@btsci-cloud-server-1:~# ps -ef|grep docker

root 5982 1 0 14:09 ? 00:00:00 docker-containerd-shim e4b6cbd3ea87347a8f6e8d640ea0350b5a35424905633b08a374e0433251f02c /var/run/docker/libcontainerd/e4b6cbd3ea87347a8f6e8d640ea0350b5a35424905633b08a374e0433251f02c docker-runc

root 6166 5699 0 14:10 pts/2 00:00:00 grep --color=auto docker

可见 docker daemon关闭后，容器进程变孤儿进程，父进程号变为1

实现机制如下（1.13版本）

func (daemon *Daemon) Shutdown() error {
	daemon.shutdown = true
	// Keep mounts and networking running on daemon shutdown if
	// we are to keep containers running and restore them.
        //即设置live-restore后，damon不关闭容器，自己直接退出
	if daemon.configStore.LiveRestoreEnabled && daemon.containers != nil {
		// check if there are any running containers, if none we should do some cleanup
		if ls, err := daemon.Containers(&types.ContainerListOptions{}); len(ls) != 0 || err != nil {
			return nil
		}
	}
	if daemon.containers != nil {
		logrus.Debugf("start clean shutdown of all containers with a %d seconds timeout...", daemon.configStore.ShutdownTimeout)
		daemon.containers.ApplyAll(func(c *container.Container) {
			if !c.IsRunning() {
				return
			}
			logrus.Debugf("stopping %s", c.ID)
			if err := daemon.shutdownContainer(c); err != nil {
				logrus.Errorf("Stop container error: %v", err)
				return
			}
			if mountid, err := daemon.layerStore.GetMountID(c.ID); err == nil {
				daemon.cleanupMountsByID(mountid)
			}
			logrus.Debugf("container stopped %s", c.ID)
		})
	}

而daemon恢复时候，还是从文件列表获取容器信息，把容器信息重新加入到Daemon对象的containers字典里

func (daemon *Daemon) restore() error {
	var (
		currentDriver = daemon.GraphDriverName()
		containers    = make(map[string]*container.Container)
	)

	logrus.Info("Loading containers: start.")

	dir, err := ioutil.ReadDir(daemon.repository)
	if err != nil {
		return err
	}

	for _, v := range dir {
		id := v.Name()
                                //获取文件列表
		container, err := daemon.load(id)
		if err != nil {
			logrus.Errorf("Failed to load container %v: %v", id, err)
			continue
		}

		// Ignore the container if it does not support the current driver being used by the graph
		if (container.Driver == "" && currentDriver == "aufs") || container.Driver == currentDriver {
			rwlayer, err := daemon.layerStore.GetRWLayer(container.ID)
			if err != nil {
				logrus.Errorf("Failed to load container mount %v: %v", id, err)
				continue
			}
			container.RWLayer = rwlayer
			logrus.Debugf("Loaded container %v", container.ID)

			containers[container.ID] = container
		} else {
			logrus.Debugf("Cannot load container %s because it was created with another graph driver.", container.ID)
		}
	}