Drop these templates into ~/.rebar/templates/:
cp riak* ~/.rebar/templates
Verify Erlang installation (>= R15B01):
erl -version
Start with your first app:
mkdir firstapp
cd firstapp
wget http://cloud.github.com/downloads/basho/rebar/rebar && chmod u+x rebar
./rebar create template=riak_core appid=firstapp
Here is an excerpt of what the output should look like:
==> firstapp (create)
Writing .gitignore
Writing Makefile
Writing README.md
Writing rebar.config
...
Congratulations, you have the start of a Riak Core application that can be deployed to multiple nodes and joined together to form a multinode cluster. Lets compile and run it in console mode:
make rel
./rel/firstapp/bin/firstapp console
At this point you have a single node of firstapp running. Lets test it via the erlang console:
1> firstapp:ping().
{pong,753586781748746817198774991869333432010090217472}
Now shut it down:
2> q().
.. and lets test it using the cmdline, and test it via http request:
./rel/firstapp/bin/firstapp start
./rel/firstapp/bin/firstapp ping
curl http://localhost:8098/firstapp/ping
./rel/firstapp/bin/firstapp stop
Above we showed how to start a single node, but this isn't typically how other Riak Core based applications like Riak are tested. Instead, there is something called a devrel that allows one to easily set up a local N-node cluster. By defauly N is 4. You can change this by either editing the Makefile and changing DEVNODES to your prefered N, or by setting the variable DEVNODES before running the commands below.
Build the dev-nodes:
make devrel
This did create 4 separate instances under the dev/ dir; check it out:
ls dev
Now, lets start all the nodes:
for d in dev/dev*; do $d/bin/firstapp start; done
Verify that the nodes are up and running:
for d in dev/dev*; do $d/bin/firstapp ping; done
You should see four pong replies. At this point it is worth saying that you have four INDIVIDUAL firstapp nodes running. They are NOT aware of each other yet. In order to form a cluster you have to join the nodes. That has to be done only once. If a node, or the entire cluster, goes down it will remember the nodes it was connected to.
for d in dev/dev{2,3,4}; do $d/bin/firstapp-admin join [email protected]; done
Finally, to make sure they really all agree on the shape of the cluster you can ask if the ring is "ready."
./dev/dev1/bin/firstapp-admin ringready
Which returns something like:
TRUE All nodes agree on the ring ['[email protected]','[email protected]','[email protected]'],'[email protected]']
To verify you have a 4 node cluster and to see the distribution of the ring, run the member_status command:
./dev/dev1/bin/firstapp-admin member_status
Which returns:
================================= Membership ==================================
Status Ring Pending Node
-------------------------------------------------------------------------------
valid 25.0% -- '[email protected]'
valid 25.0% -- '[email protected]'
valid 25.0% -- '[email protected]'
valid 25.0% -- '[email protected]'
-------------------------------------------------------------------------------
Valid:4 / Leaving:0 / Exiting:0 / Joining:0 / Down:0
Pretty cool!! Your riak-core cluster is up and running.
And in case you want to stop all the nodes:
for d in dev/dev*; do $d/bin/firstapp stop; done
For rapid, iterative development both make rel and make devrel
have a stage counterpart. Running either:
make stage
or
make stagedevrel
will cause the deps and apps directories to be symlinked into the
release(s) lib directory. This means you can edit your apps source
code, recompile it, and load it on to the running node(s). You can
either start mochiweb reloader
reloader:start().
when attached to the node(s), which will cause any changes to be automatically reloaded. Or attach to a running node and run
`l(mod_name)`.
to reload a specific module.