OpenStorage is a clustered implementation of the Open Storage specification and relies on the Docker runtime. It allows you to run stateful services in Docker in a multi-host environment. It plugs into Docker volumes to provide storage to a container and plugs into Swarm to operate in a clustered environment.
When you install openstorage on a Linux host, you will automatically get a stateful storage layer that integrates with the Docker runtime and operates in a multi host environment. It starts an Open Storage Daemon - OSD
that currently supports Docker and will support any Linux container runtime that conforms to the OCI.
OSD will work with any distributed scheduler that is compatible with the Docker remote API.
OSD integrates with Docker Volumes and provisions storage to a container on behalf of any third party OSD driver and ensures the volumes are available in a multi host environment.
OpenStorage provides support for the Graph Driver in addition to Docker Volumes
. When used as a graph driver, the container's layers will be stored on a volume provided by the OSD.
The diagram below shows OSD integrated with Docker and Swarm to allow for provisioning of storage to containers in a multi node environment.
There are default drivers built-in for NFS, AWS and BTRFS. By using openstorage, you can get container granular, stateful storage provisioning to Linux containers with the backends supported by openstorage. We are working with the storage ecosystem to add more drivers for various storage providers.
Providers that support a multi-node environment, such as AWS or NFS to name a few, can provide highly available storage to linux containers across multiple hosts.
libopenstorage is written in the Go programming language. If you haven't set up a Go development environment, please follow these instructions to install golang
and set up GOPATH. Your version of Go must be at least 1.5 - we use the golang 1.5 vendor experiment https://golang.org/s/go15vendor. Note that the version of Go in package repositories of some operating systems is outdated, so please download the latest version.
After setting up Go, you should be able to go get
libopenstorage as expected (we use -d
to only download):
$ GO15VENDOREXPERIMENT=1 go get -d github.com/libopenstorage/openstorage/...
At this point you can build openstorage from the source folder:
$ cd $GOPATH/src/github.com/libopenstorage/openstorage
$ make install
or run only unit tests:
$ cd $GOPATH/src/github.com/libopenstorage/openstorage
$ make test
OSD is both the openstorage daemon and the CLI. When run as a daemon, the OSD is ready to receive RESTful commands to operate on volumes and attach them to a Docker container. It works with the Docker volumes plugin interface will communicate with Docker version 1.7 and later. When this daemon is running, Docker will automatically communicate with the daemon to manage a container's volumes.
Note: OSD needs to be run as root.
To start the OSD in daemon mode:
$GOPATH/bin/osd -d -f etc/config/config.yaml
where, config.yaml is the daemon's configuiration file and its format is explained below.
To have OSD persist the volume mapping across restarts, you must use an external key value database such as etcd or consul. The URL of your key value database must be passed into the OSD using the --kvdb
option. For example:
$GOPATH/bin/osd -d -f etc/config/config.yaml -k etcd-kv://localhost:4001
To use the OSD cli, see the CLI help menu:
NAME:
osd - Open Storage CLI
USAGE:
osd [global options] command [command options] [arguments...]
VERSION:
v1
COMMANDS:
driver Manage drivers
cluster Manage cluster
version Display version
aws Manage aws storage
btrfs Manage btrfs volumes
buse Manage buse storage
coprhd Manage coprhd storage
nfs Manage nfs volumes
pwx Manage pwx storage
vfs Manage vfs volumes
chainfs Manage chainfs graph storage
layer0 Manage layer0 graph storage
proxy Manage proxy graph storage
GLOBAL OPTIONS:
--json, -j output in json
--daemon, -d Start OSD in daemon mode
--driver [--driver option --driver option] driver name and options: name=btrfs,home=/var/openstorage/btrfs
--kvdb, -k "kv-mem://localhost" uri to kvdb e.g. kv-mem://localhost, etcd-kv://localhost:4001, consul-kv://localhost:8500
--file, -f file to read the OSD configuration from.
--help, -h show help
--version, -v print the version
The OSD daemon loads a YAML configuration file that tells the daemon what drivers to load and the driver specific attributes. Here is an example of config.yaml:
osd:
drivers:
nfs:
server: "171.30.0.20"
path: "/nfs"
btrfs:
aws:
aws_access_key_id: your_aws_access_key_id
aws_secret_access_key: your_aws_secret_access_key
coprhd:
restUrl: coprhd_rest_url
user: rest_user_name
password: rest_user_password
consistency_group: consistency_group_id
project: project_id
varray: varray_id
vpool: vpool_id
graphdrivers:
proxy:
layer0:
The above example initializes the OSD
with three drivers: NFS, BTRFS and AWS. Each have their own configuration sections.
Adding a driver is fairly straightforward:
-
Add your driver decleration in
volumes/drivers.go
-
Add your driver
mydriver
implementation in thevolumes/drivers/mydriver
directory. The driver must implement theVolumeDriver
interface specified involumes/volume.go
. This interface is an implementation of the specification available [here] (http://api.openstorage.org/). -
You're driver must be a
File Volume
driver or aBlock Volume
driver. AFile Volume
driver will not implement a few low level primatives, such asFormat
,Attach
andDetach
.
Here is an example of drivers.go
:
// To add a provider to openstorage, declare the provider here.
package drivers
import (
"github.com/libopenstorage/openstorage/volume/drivers/aws"
"github.com/libopenstorage/openstorage/volume/drivers/btrfs"
"github.com/libopenstorage/openstorage/volume/drivers/nfs"
"github.com/libopenstorage/openstorage/volume"
)
type Driver struct {
providerType volume.ProviderType
name string
}
var (
providers = []Driver{
// AWS provider. This provisions storage from EBS.
{providerType: volume.Block,
name: aws.Name},
// NFS provider. This provisions storage from an NFS server.
{providerType: volume.File,
name: nfs.Name},
// BTRFS provider. This provisions storage from local btrfs fs.
{providerType: volume.File,
name: btrfs.Name},
}
)
That's pretty much it. At this point, when you start the OSD, your driver will be loaded.
make test # test on your local machine
make docker-test # test within a docker container
Assuming you are using the NFS driver, to create a volume with a default size of 1GB and attach it to a Docker container, you can do the following
$ docker volume create -d nfs
9ccb7280-918b-464f-8a34-34e73e9214d2
$ docker run -v 9ccb7280-918b-464f-8a34-34e73e9214d2:/root --volume-driver=nfs -ti busybox
To update the source folder and all dependencies:
$GOPATH/src/github.com/libopenstorage/openstorage $ make update-test-deps
However note that all dependencies are vendored in the vendor directory, so this is not necessary in general as long as you have GO15VENDOREXPERIMENT
set:
export GO15VENDOREXPERIMENT=1
OSD can run inside of Docker:
make docker-build-osd
This builds a Docker image called openstorage/osd
. You can then run the image:
make launch
Pre-built Docker images of the OSD are available at https://hub.docker.com/r/openstorage/osd/
[Unit]
Description=Open Storage
[Service]
CPUQuota=200%
MemoryLimit=1536M
ExecStart=/usr/local/bin/osd
Restart=on-failure
[Install]
WantedBy=multi-user.target
The specification and code is licensed under the Apache 2.0 license found in
the LICENSE
file of this repository.
See the Style Guide.
docker pull quay.io/pedge/protoeasy
Add to your ~/.bashrc (or equivalent):
# to use protoeasy for now, you must have docker installed locally or in a vm
# if running docker using docker-machine etc, replace 192.168.10.10 with the ip of the vm
# if running docker locally, replace 192.168.10.10 with 0.0.0.0
export PROTOEASY_ADDRESS=127.0.0.1:6789
launch-protoeasy() {
docker rm -f protoeasy || true
docker run -d -p 6789:6789 --name=protoeasy quay.io/pedge/protoeasy
}
Then just run launch-protoeasy
before compiling the protocol buffers files, and then to compile:
make proto
The sign-off is a simple line at the end of the explanation for the patch, which certifies that you wrote it or otherwise have the right to pass it on as an open-source patch. The rules are pretty simple: if you can certify the below (from developercertificate.org):
Developer Certificate of Origin
Version 1.1
Copyright (C) 2004, 2006 The Linux Foundation and its contributors.
660 York Street, Suite 102,
San Francisco, CA 94110 USA
Everyone is permitted to copy and distribute verbatim copies of this
license document, but changing it is not allowed.
Developer's Certificate of Origin 1.1
By making a contribution to this project, I certify that:
(a) The contribution was created in whole or in part by me and I
have the right to submit it under the open source license
indicated in the file; or
(b) The contribution is based upon previous work that, to the best
of my knowledge, is covered under an appropriate open source
license and I have the right under that license to submit that
work with modifications, whether created in whole or in part
by me, under the same open source license (unless I am
permitted to submit under a different license), as indicated
in the file; or
(c) The contribution was provided directly to me by some other
person who certified (a), (b) or (c) and I have not modified
it.
(d) I understand and agree that this project and the contribution
are public and that a record of the contribution (including all
personal information I submit with it, including my sign-off) is
maintained indefinitely and may be redistributed consistent with
this project or the open source license(s) involved.
then you just add a line to every git commit message:
Signed-off-by: Joe Smith <joe@gmail.com>
using your real name (sorry, no pseudonyms or anonymous contributions.)
You can add the sign off when creating the git commit via git commit -s
.