Convoy is a Docker volume plugin for a variety of storage back-ends. It's designed to be a simple Docker volume plug-ins that supports vendor-specific extensions such as snapshots, backups and restore. It's written in Go and can be deployed as a standalone binary.
Docker has various drivers(aufs, device mapper, etc) for container's root image, but not for volumes. User can create volume through docker run -v volname, but it's disposable, cannot be easily reused for new containers or containers on the other hosts. For example, if you start a wordpress container with database, add some posts, remove the container, then the modified database would lost.
Before volume plugin, the only way to reuse the volume is using host bind mount feature of Docker, as docker run -v /host_path:/container_path, then maintain the content of the volume at /hostpath. You can also use --volume-from but that would require original container still exists on the same host.
Convoy used Docker volume plugin mechanism to provide persistent volume for Docker containers, and supports various of backends(e.g. device mapper, NFS, EBS) and more features like snapshot/backup/restore. So user would able to migrate the volumes between the hosts, share the same volume across the hosts, make scheduled snapshots of as well as recover to previous version of volume. It's much easier for user to manage data with Docker volumes with Convoy.
- Device Mapper
- Virtual File System(VFS)/Network File System(NFS)
- Amazon Elastic Block Store(EBS)
First let's make sure we have Docker 1.8 or above running.
docker --version
If not, install the latest Docker daemon as follows:
curl -sSL https://get.docker.com/ | sh
Once we have made sure we have the right Docker daemon running, we can install and configure Convoy volume plugin as follows:
wget https://github.com/rancher/convoy/releases/download/v0.3/convoy.tar.gz
tar xvf convoy.tar.gz
sudo cp convoy/convoy convoy/convoy-pdata_tools /usr/local/bin/
sudo mkdir -p /etc/docker/plugins/
sudo bash -c 'echo "unix:///var/run/convoy/convoy.sock" > /etc/docker/plugins/convoy.spec'
We can use file-backed lookback device to test and demo Convoy Device Mapper driver. Loopback device, however, is known to be unstable and should not be used in production.
truncate -s 100G data.vol
truncate -s 1G metadata.vol
sudo losetup /dev/loop5 data.vol
sudo losetup /dev/loop6 metadata.vol
Once we have the data and metadata device setup, we can start the Convoy plugin daemon as follows:
sudo convoy daemon --drivers devicemapper --driver-opts dm.datadev=/dev/loop5 --driver-opts dm.metadatadev=/dev/loop6
We can create a Docker container with a convoy volume. As a test, we create a file called /vol1/foo in the convoy volume:
sudo docker run -v vol1:/vol1 --volume-driver=convoy ubuntu touch /vol1/foo
Next we take a snapshot of the convoy volume. We backup the snapshot to a local directory: (Backup to NFS share or S3 objectore is also supported.)
sudo convoy snapshot create vol1 --name snap1vol1
sudo mkdir -p /opt/convoy/
sudo convoy backup create snap1vol1 --dest vfs:///opt/convoy/
The convoy backup command returns a URL string representing backup dataset. You can use the same URL string to recover the volume to another host:
sudo convoy create res1 --backup <backup_url>
The following command creates a new container and mounts the recovered convoy volume into that container:
sudo docker run -v res1:/res1 --volume-driver=convoy ubuntu ls /res1/foo
You should see the recovered file /res1/foo.
Ensure you have Docker 1.8 or above installed.
Download latest version of convoy and unzip it. Put the binaries in a directory in the execution $PATH of sudo and root users (e.g. /usr/local/bin).
wget https://github.com/rancher/convoy/releases/download/v0.3/convoy.tar.gz
tar xvf convoy.tar.gz
sudo cp convoy/convoy convoy/convoy-pdata_tools /usr/local/bin/
Run the following commands to setup the Convoy volume plugin for Docker:
sudo mkdir -p /etc/docker/plugins/
sudo bash -c 'echo "unix:///var/run/convoy/convoy.sock" > /etc/docker/plugins/convoy.spec'
You need to pass different arguments to convoy daemon depending on the choice of backend implementation.
Assuming you have two devices created, one data device called /dev/convoy-vg/data and the other metadata device called /dev/convoy-vg/metadata. You run the following command to start the Convoy daemon:
sudo convoy daemon --drivers devicemapper --driver-opts dm.datadev=/dev/convoy-vg/data --driver-opts dm.metadatadev=/dev/convoy-vg/metadata
- A default Device Mapper volume size is 100G. You can override it with the
---driver-opts dm.defaultvolumesizeoption. - You can take a look at here if you want to know how much storage need to be allocated for metadata device.
First, mount the NFS share to the root directory used to store volumes. Substitute <vfs_path> to the appropriate directory of your choice:
sudo mkdir <vfs_path>
sudo mount -t nfs <nfs_server>:/path <vfs_path>
The NFS-based Convoy daemon can be started as follows:
sudo convoy daemon --drivers vfs --driver-opts vfs.path=<vfs_path>
Make sure you're running on EC2 instance, and has already configured AWS credentials correctly.
sudo convoy daemon --drivers ebs
Volumes can be created using the convoy create command:
sudo convoy create volume_name
- Device Mapper: Default volume size is 100G.
--sizeoption is supported. - EBS: Default volume size is 4G.
--sizeand some other options are supported.
We can also create a volume using the docker run command. If the volume does not yet exist, a new volume will be created. Otherwise the existing volume will be used.
sudo docker -it test_volume:/test --volume-driver=convoy ubuntu
sudo convoy delete <volume_name>
or
sudo docker rm -v <container_name>
- NFS or EBS: The
-r/--referenceoption instructs theconvoy deletecommand to only delete the reference to the volume from the current host and leave the underlying files on NFS server or EBS volume unchanged. This is useful when the volume need to be reused later. docker rm -vwould be treated asconvoy deletewith-r/--reference.
sudo convoy list
sudo convoy inspect vol1
sudo convoy snapshot create vol1 --name snap1vol1
sudo convoy snapshot delete snap1vol1
- Device Mapper: please make sure you keep the latest backed-up snapshot for the same volume available to enable incremental backup mechanism, since Convoy need it to calculate the differences between snapshots.
- Device Mapper or VFS: We can backup a snapshot to S3 object store or an NFS mount/local directory:
sudo convoy backup create snap1vol1 --dest s3://backup-bucket@us-west-2/
or
sudo convoy backup create snap1vol1 --dest vfs:///opt/backup/
The backup operation returns a URL string that uniquely idenfied the backup dataset.
s3://backup-bucket@us-west-2/?backup=f98f9ea1-dd6e-4490-8212-6d50df1982ea\u0026volume=e0d386c5-6a24-446c-8111-1077d10356b0
If you're using S3, please make sure you have AWS credential ready either at ~/.aws/credentials or as environment variables, as described here. You may need to put credentials to /root/.aws/credentials or setup sudo environment variables in order to get S3 credential works.
- EBS:
--destis not needed. Just doconvoy backup create snap1vol1.
sudo convoy create res1 --backup <url>
- EBS: Current host must be in the same region of backup to be restored.
We can use the standard docker run command to mount the restored volume into a Docker container:
sudo docker run -it -v res1:/res1 --volume-driver convoy ubuntu
You can mount an NFS-backed volume on multiple servers. You can use the standard docker run command to mount an existing NFS-backed mount into a Docker container. For example, if you have already created an NFS-based volume vol1 on one host, you can run the following command to mount the existing vol1 volume into a new container:
sudo docker run -it -v vol1:/vol1 --volume-driver=convoy ubuntu
If you need any help with Convoy, please join us at either our forum or #rancher IRC channel.
Feel free to submit any bugs, issues, and feature requests to Convoy Issues.
Contribution are welcome! Please take a look at Development Guide if you want to how to build Convoy from source or running test cases.
We love to hear new Convoy Driver ideas from you. Implemenations are most welcome! Please consider take a look at enhancement ideas if you want contribute.
And of course, bug fixes are always welcome!