This page outlines the steps for getting a Storm cluster up and running. If you're on AWS, you should check out the storm-deploy project. storm-deploy completely automates the provisioning, configuration, and installation of Storm clusters on EC2. It also sets up Ganglia for you so you can monitor CPU, disk, and network usage.
Here's a summary of the steps for setting up a Storm cluster:
Storm uses Zookeeper for coordinating the cluster. Zookeeper is not used for message passing, so the load Storm places on Zookeeper is quite low. Single node Zookeeper clusters should be sufficient for most cases, but if you want failover or are deploying large Storm clusters you may want larger Zookeeper clusters. Instructions for deploying Zookeeper are here.
A few notes about Zookeeper deployment:
Next you need to install Storm's dependencies on Nimbus and the worker machines. These are:
These are the versions of the dependencies that have been tested with Storm. Storm may or may not work with different versions of Java and/or Python.
Next, download a Storm release and extract the zip file somewhere on Nimbus and each of the worker machines. The Storm releases can be downloaded from here.
The Storm release contains a file at
conf/storm.yaml that configures the Storm daemons. You can see the default configuration values here. storm.yaml overrides anything in defaults.yaml. There's a few configurations that are mandatory to get a working cluster:
1) storm.zookeeper.servers: This is a list of the hosts in the Zookeeper cluster for your Storm cluster. It should look something like:
storm.zookeeper.servers: - "111.222.333.444" - "555.666.777.888"
If the port that your Zookeeper cluster uses is different than the default, you should set storm.zookeeper.port as well.
2) storm.local.dir: The Nimbus and Supervisor daemons require a directory on the local disk to store small amounts of state (like jars, confs, and things like that). You should create that directory on each machine, give it proper permissions, and then fill in the directory location using this config. For example:
3) java.library.path: This is the load path for the native libraries that Storm uses (ZeroMQ and JZMQ). The default of "/usr/local/lib:/opt/local/lib:/usr/lib" should be fine for most installations, so you probably don't need to set this config.
4) nimbus.host: The worker nodes need to know which machine is the master in order to download topology jars and confs. For example:
5) supervisor.slots.ports: For each worker machine, you configure how many workers run on that machine with this config. Each worker uses a single port for receiving messages, and this setting defines which ports are open for use. If you define five ports here, then Storm will allocate up to five workers to run on this machine. If you define three ports, Storm will only run up to three. By default, this setting is configured to run 4 workers on the ports 6700, 6701, 6702, and 6703. For example:
supervisor.slots.ports: - 6700 - 6701 - 6702 - 6703
The last step is to launch all the Storm daemons. It is critical that you run each of these daemons under supervision. Storm is a fail-fast system which means the processes will halt whenever an unexpected error is encountered. Storm is designed so that it can safely halt at any point and recover correctly when the process is restarted. This is why Storm keeps no state in-process -- if Nimbus or the Supervisors restart, the running topologies are unaffected. Here's how to run the Storm daemons:
As you can see, running the daemons is very straightforward. The daemons will log to the logs/ directory in wherever you extracted the Storm release.
Last edited by nathanmarz,