How to Install Storm: A Step-by-Step Tutorial

Apache Storm is a versatile and widely-used distributed real-time computation system, perfect for processing large streams of data with low latency. This tutorial will guide you through the process of installing Storm on a Linux system, either for development or production use. We will cover the prerequisites, installing dependencies, setting up Storm and Zookeeper, and verifying the installation.

Prerequisites

Linux-based operating system (Ubuntu, CentOS, Debian, etc.)
Java Development Kit (JDK) 8 or higher installed
Apache Zookeeper installed and running (Storm uses Zookeeper for coordination)
Basic command-line knowledge
Network connectivity to download required packages

Step 1: Install Java Development Kit (JDK)

Apache Storm requires Java because it runs on the JVM. To check if Java is installed, run:

java -version

If Java is not installed, follow these commands for Ubuntu/Debian based systems:

sudo apt update
sudo apt install openjdk-8-jdk -y

For CentOS/Fedora:

sudo yum install java-1.8.0-openjdk-devel -y

Verify the installation again with java -version.

Step 2: Install Apache Zookeeper

Zookeeper is necessary for managing the Storm cluster state.

To install Zookeeper on Ubuntu:

sudo apt install zookeeperd -y

Start and enable Zookeeper if not running:

sudo systemctl start zookeeper
sudo systemctl enable zookeeper

Verify Zookeeper is running:

sudo systemctl status zookeeper

Step 3: Download and Extract Apache Storm

Visit the Apache Storm official site to get the latest release. Use wget for downloading (e.g., version 2.4.0):

wget https://downloads.apache.org/storm/apache-storm-2.4.0/apache-storm-2.4.0.tar.gz

Extract the tarball to /opt:

sudo tar -xvzf apache-storm-2.4.0.tar.gz -C /opt/

Step 4: Configure Environment Variables

For convenience, create symbolic links and add Storm to your PATH:

sudo ln -s /opt/apache-storm-2.4.0 /opt/storm
export STORM_HOME=/opt/storm
export PATH=$PATH:$STORM_HOME/bin

Add the above environment variables to ~/.bashrc or /etc/profile to load them every session.

Step 5: Configure Storm

Edit the Storm configuration file /opt/storm/conf/storm.yaml with your preferred editor.

Minimal example configuration:

storm.zookeeper.servers:
    - "localhost"

nimbus.seeds: ["localhost"]

storm.local.dir: "/var/storm"

supervisor.slots.ports:
    - 6700
    - 6701
    - 6702
    - 6703

This example assumes a single-node Storm cluster. For a multi-node cluster, update hostnames accordingly.

Step 6: Start Storm Services

Storm has several daemon processes:

nimbus: master node
supervisor: manages workers
ui: web UI

Start each in a separate terminal or create system service scripts for production. To start manually:

storm nimbus
storm supervisor
storm ui

The Storm UI is accessible at http://localhost:8080 by default.

Troubleshooting Tips

If Storm cannot connect to Zookeeper, ensure Zookeeper is running and the host is reachable.
Verify Java version compatibility.
Check storm.yaml carefully for YAML syntax errors.
Firewall settings might need adjustment to allow Storm ports.
Logs located in $STORM_HOME/logs can help diagnose problems.

Summary Checklist

✅ Installed JDK 8 or newer
✅ Installed and running Zookeeper
✅ Downloaded and extracted Apache Storm
✅ Configured environment variables
✅ Edited storm.yaml with correct settings
✅ Started Storm daemons and verified UI access

For more detailed insights on cluster management and real-time data processing pipelines, you can also explore our tutorial on How to Install Apache Flink, another powerful stream processing tool.