Skip to content

Latest commit

 

History

History
188 lines (140 loc) · 6.61 KB

getting-started.md

File metadata and controls

188 lines (140 loc) · 6.61 KB
Raw

Running your first Torus cluster

0) Get Torus

Download a release

Releases are available at http://github.com/coreos/torus/releases

Build from master

For builds, Torus assumes a Go installation and a correctly configured GOPATH. Simply checkout the repo and use the Makefile to build.

git clone [email protected]:coreos/torus $GOPATH/src/github.com/coreos/torus
cd $GOPATH/src/github.com/coreos/torus
make

This will create the binaries torusd, torusctl and torusblk in the "bin" directory. You can think of torusd as the storage daemon, torusctl as the administrative tool, and torusblk as the client daemon that mounts volumes on a host or exposes them through NBD, AoE, or similar.

On first build Torus will install and use glide locally to download its dependenices.

1) Get etcd

You need a v3.0 or higher etcd instance, as torus uses the v3 API natively and uses the latest client. You might try etcd v3.0.0-beta.0.

Running etcd manually

To run a single node cluster locally:

etcd --data-dir /tmp/etcd

Running etcd using rkt

To run a single node cluster locally, execute:

$ rkt fetch coreos.com/etcd:v3.0.0-beta.0
$ mkdir /tmp/etcd
$ rkt run coreos.com/etcd:v3.0.0-beta.0 \
     --volume=data-dir,kind=host,source=/tmp/etcd,readOnly=false \
     -- \
    -advertise-client-urls http://0.0.0.0:2379,http://0.0.0.0:4001 \
    -listen-client-urls http://0.0.0.0:2379,http://0.0.0.0:4001 \
    -listen-peer-urls http://0.0.0.0:2380
$ export ETCD_IP=$(rkt l --full=true --no-legend=true | grep 'etcd.*running' | cut -f8 | cut -d'=' -f2)

Clustering etcd for high availability and setting up a production etcd are covered by the etcd team.

2) init

We need to initialize Torus in etcd. This sets the global settings for the storage cluster, analogous to "formatting the cluster". The default settings are useful for most deployments.

./bin/torusctl init

And you're ready!

If torusctl can't connect to etcd, it takes the -C flag, just like etcdctl

./bin/torusctl -C $ETCD_IP:2379 init

(This remains true for all uses of torus binaries)

If you're curious about the other settings,

./bin/torusctl init --help

will tell you more, check the docs, or feel free to ask in IRC.

3) Run some storage nodes

Running manually

./bin/torusd --etcd 127.0.0.1:2379 --peer-address http://127.0.0.1:40000 --data-dir /tmp/torus1 --size 20GiB
./bin/torusd --etcd 127.0.0.1:2379 --peer-address http://127.0.0.1:40001 --data-dir /tmp/torus2 --size 20GiB

This runs a storage node without HTTP. Add --host and --port to open the HTTP endpoint for monitoring.

Multiple instances can be run, so long as the ports don't conflict and you keep separate data dirs.

Running with rkt

The following will start a local three node torus cluster::

$ rkt fetch quay.io/coreos/torus
$ mkdir -p /tmp/torus/{1,2,3}
$ rkt run --volume=volume-data,kind=host,source=/tmp/torus/1" \
    --set-env LISTEN_HOST=0.0.0.0 \
    --set-env PEER_ADDRESS=http://0.0.0.0:40000 \
    --set-env ETCD_HOST="${ETCD_IP}" quay.io/coreos/torus

Start two additional instances of torus replacing --volume=...source=/tmp/torus/{1,2,3}.

Running with Docker

With Host Networking
docker run \
--net=host \
-v /path/to/data1:/data \
-e STORAGE_SIZE=20GiB \
-e LISTEN_HTTP_PORT=4321 \
-e PEER_ADDRESS=http://127.0.0.1:40000 \
-e ETCD_HOST=127.0.0.1 \
quay.io/coreos/torus

If you want to run more than one storage node on the host, you can do so by offsetting the ports.

Non-host networking

You'll need to figure out non-host networking where all storage nodes are on the same subnet. Flannel, et al, are recommended here. But if you're okay with your docker networking...

docker run \
-v /path/to/data1:/data \
-e STORAGE_SIZE=20GiB \
-e PEER_ADDRESS=http://$NODE_IP:40000 \
-e ETCD_HOST=127.0.0.1 \
quay.io/coreos/torus

4) Check that everything is reporting in

./bin/torusctl list-peers

Should show your data nodes and their reporting status. Eg:

ADDRESS                 UUID                                  SIZE     USED  MEMBER  UPDATED       REB/REP DATA
http://127.0.0.1:40000  b2a2cbe6-38b7-11e6-ab37-5ce0c5527cf4  5.0 GiB  0 B   OK      1 second ago  0 B/sec
http://127.0.0.1:40002  b2a2cbf8-38b7-11e6-9404-5ce0c5527cf4  5.0 GiB  0 B   OK      1 second ago  0 B/sec
http://127.0.0.1:40001  b2a2cc9e-38b7-11e6-b607-5ce0c5527cf4  5.0 GiB  0 B   OK      1 second ago  0 B/sec
Balanced: true Usage:  0.00%

5) Activate storage on the peers

./bin/torusctl peer add --all-peers

You'll notice if you run torusctl list-peers again, the MEMBER column will have changed from Avail to OK. These nodes are now storing data. Peers can be added one (or a couple) at a time via:

./bin/torusctl peer add $PEER_ADDRESS [$PEER_UUID...]

To see which peers are in service (and other sharding details):

./bin/torusctl ring get

To remove a node from service:

./bin/torusctl peer remove $PEER_ADDRESS

Draining of peers will happen automatically. If this is a hard removal (ie, the node is gone forever) just remove it, and data will re-replicate automatically. Doing multiple hard removals above the replication threshold may result in data loss. However, this is common practice if you're familiar with RAID levels..

Even better fault tolerance with erasure codes and parity is on the roadmap.

6) Create a volume

./bin/torusctl volume create-block myVolume 10GiB

This creates a 10GiB virtual blockfile for use. It will be safely replicated and CRC checked, by default.

7) Mount that volume via NBD

sudo modprobe nbd
sudo ./bin/torusblk --etcd 127.0.0.1:2379 nbd myVolume /dev/nbd0

Specifying /dev/nbd0 is optional -- it will pick the first available device if unspecified.

The mount process is similar to FUSE for a block device; it will disconnect when killed, so make sure it's synced and unmounted.

At this point, you have a replicated, highly-available block device connected to your machine. You can format it and mount it using the standard tools you expect:

sudo mkfs.ext4 /dev/nbd0
sudo mount /dev/nbd0 -o discard,noatime /mnt/torus

torusblk nbd supports the TRIM SSD command to accelerate garbage collecting; the discard option enables this.