Keep containers alive during daemon downtime

By default, when the Docker daemon terminates, it shuts down running containers. Starting with Docker Engine 1.12, you can configure the daemon so that containers remain running if the daemon becomes unavailable. The live restore option helps reduce container downtime due to daemon crashes, planned outages, or upgrades.

Note: Live restore is not supported on Windows containers, but it does work for Linux containers running on Windows.

Enable the live restore option

There are two ways to enable the live restore setting to keep containers alive when the daemon becomes unavailable:

  • If the daemon is already running and you don’t want to stop it, you can add the configuration to the daemon configuration file. For example, on a linux system the default configuration file is /etc/docker/daemon.json.

Use your favorite editor to enable the live-restore option in the daemon.json.

{
"live-restore": true
}

You have to send a SIGHUP signal to the daemon process for it to reload the configuration. For more information on how to configure the Docker daemon using config.json, see daemon configuration file.

  • When you start the Docker daemon, pass the --live-restore flag:

      $ sudo dockerd --live-restore
    

Live restore during upgrades

The live restore feature supports restoring containers to the daemon for upgrades from one minor release to the next. For example from Docker Engine 1.12.1 to 1.13.2.

If you skip releases during an upgrade, the daemon may not restore connection the containers. If the daemon is unable restore connection, it ignores the running containers and you must manage them manually. The daemon won’t shut down the disconnected containers.

Live restore upon restart

The live restore option only works to restore the same set of daemon options as the daemon had before it stopped. For example, live restore may not work if the daemon restarts with a different bridge IP or a different graphdriver.

Impact of live restore on running containers

A lengthy absence of the daemon can impact running containers. The containers process writes to FIFO logs for daemon consumption. If the daemon is unavailable to consume the output, the buffer will fill up and block further writes to the log. A full log blocks the process until further space is available. The default buffer size is typically 64K.

You must restart Docker to flush the buffers.

You can modify the kernel’s buffer size by changing /proc/sys/fs/pipe-max-size.

Live restore and swarm mode

The live restore option is not compatible with Docker Engine swarm mode. When the Docker Engine runs in swarm mode, the orchestration feature manages tasks and keeps containers running according to a service specification.

© 2017 Docker, Inc.
Licensed under the Apache License, Version 2.0.
Docker and the Docker logo are trademarks or registered trademarks of Docker, Inc. in the United States and/or other countries.
Docker, Inc. and other parties may also have trademark rights in other terms used herein.
https://docs.docker.com/v1.13/engine/admin/live-restore/