Simplify Your Server Deployments with a Smart Installation Script

Server deployments can be tedious, especially when you need to configure each server manually. Tasks like setting up directories, copying scripts, installing dependencies, and scheduling periodic jobs can be repetitive and error-prone.

Wouldn’t it be great if all of this could be handled with a single command? That’s where a robust installation script comes in. In this post, we’ll walk through how to create a flexible and secure installation script for your server-side applications. As a bonus, we’ll demonstrate how you can execute the installation with a single-line command. It’s a basic version of an install script that you can use as an template for your own installs.

The Power of an Installation Script

A well-designed installation script brings several advantages:

• Consistency: Ensures the same setup across multiple servers.
• Efficiency: Saves time by automating repetitive tasks.
• Flexibility: Allows customization for different environments or configurations.
• Ease of Use: Makes it easy for others to install and run your software.

In this guide, we’ll focus on a real-world example: deploying a monitoring agent to multiple servers. The same principles can be applied to virtually any server-side application.

Building the Script

1. Start with Dependencies

The script should first check if all required tools are installed, such as curl and jq. If they’re missing, the script installs them automatically.

Copied!if ! command -v curl >/dev/null 2>&1 || ! command -v jq >/dev/null 2>&1; then
    echo "Installing required tools: curl and jq"
    if [ -f /etc/debian_version ]; then
        sudo apt update && sudo apt install -y curl jq
    elif [ -f /etc/redhat-release ]; then
        sudo yum install -y curl jq
    else
        echo "Unsupported OS. Please install 'curl' and 'jq' manually."
        exit 1
    fi
fi
if ! command -v curl >/dev/null 2>&1 || ! command -v jq >/dev/null 2>&1; then
    echo "Installing required tools: curl and jq"
    if [ -f /etc/debian_version ]; then
        sudo apt update && sudo apt install -y curl jq
    elif [ -f /etc/redhat-release ]; then
        sudo yum install -y curl jq
    else
        echo "Unsupported OS. Please install 'curl' and 'jq' manually."
        exit 1
    fi
fi

2. Automate Configuration

A configuration file is crucial for storing environment-specific settings like API keys or endpoint URLs. Here’s how to create one dynamically:

Copied!cat > /opt/agent/agent.conf 
cat > /opt/agent/agent.conf <<EOL
API_KEY="your-api-key"
BASE_URL="https://your-api-url"
API_URL="${BASE_URL}/api/agent_checks"
SUBMIT_URL="${BASE_URL}/api/submit"
DEBUG=true
LOG_FILE="/var/log/agent.log"
EOL

3. Handle Files and Directories

The script creates the necessary directories, copies the main agent script, and ensures proper permissions:

Copied!mkdir -p /opt/agent
cp agent.sh /opt/agent/
chmod +x /opt/agent/agent.sh
touch /var/log/agent.log
chmod 644 /var/log/agent.log
mkdir -p /opt/agent
cp agent.sh /opt/agent/
chmod +x /opt/agent/agent.sh
touch /var/log/agent.log
chmod 644 /var/log/agent.log

4. Schedule the Agent

Adding a cronjob ensures the agent runs periodically:

Copied!CRON_CMD="* * * * * root /opt/agent/agent.sh > /dev/null 2>&1"
if ! grep -q "$CRON_CMD" /etc/crontab; then
    echo "$CRON_CMD" >> /etc/crontab
fi
CRON_CMD="* * * * * root /opt/agent/agent.sh > /dev/null 2>&1"
if ! grep -q "$CRON_CMD" /etc/crontab; then
    echo "$CRON_CMD" >> /etc/crontab
fi

5. Wrap It All Up

Finally, combine everything into a single script (install_agent.sh) that runs these steps sequentially:

Copied!#!/bin/bash

if [ $# -ne 2 ]; then
    echo "Usage: $0 <api_key> <base_url>"
    exit 1
fi

API_KEY=$1
BASE_URL=$2

# Install dependencies
if ! command -v curl >/dev/null 2>&1 || ! command -v jq >/dev/null 2>&1; then
    sudo apt update && sudo apt install -y curl jq
fi

# Create directories and files
mkdir -p /opt/agent
cp agent.sh /opt/agent/
chmod +x /opt/agent/agent.sh

# Create configuration file
cat > /opt/agent/agent.conf  /dev/null 2>&1"
if ! grep -q "$CRON_CMD" /etc/crontab; then
    echo "$CRON_CMD" >> /etc/crontab
fi

echo "Installation completed successfully!"</base_url></api_key>
#!/bin/bash

if [ $# -ne 2 ]; then
    echo "Usage: $0 <API_KEY> <BASE_URL>"
    exit 1
fi

API_KEY=$1
BASE_URL=$2

# Install dependencies
if ! command -v curl >/dev/null 2>&1 || ! command -v jq >/dev/null 2>&1; then
    sudo apt update && sudo apt install -y curl jq
fi

# Create directories and files
mkdir -p /opt/agent
cp agent.sh /opt/agent/
chmod +x /opt/agent/agent.sh

# Create configuration file
cat > /opt/agent/agent.conf <<EOL
API_KEY="$API_KEY"
BASE_URL="$BASE_URL"
API_URL="${BASE_URL}/api/agent_checks"
SUBMIT_URL="${BASE_URL}/api/submit"
DEBUG=true
LOG_FILE="/var/log/agent.log"
EOL

touch /var/log/agent.log
chmod 644 /var/log/agent.log

# Add cronjob
CRON_CMD="* * * * * root /opt/agent/agent.sh > /dev/null 2>&1"
if ! grep -q "$CRON_CMD" /etc/crontab; then
    echo "$CRON_CMD" >> /etc/crontab
fi

echo "Installation completed successfully!"

Deploying the Script with a One-Liner

To make installation even easier, host your install_agent.sh and agent.sh files on a GitHub repository or a secure server. Then use a single-line command to download and run the installation script:

Copied!curl -sSL https://raw.githubusercontent.com/your-repo/install_agent.sh | bash -s -- your-api-key base_url
curl -sSL https://raw.githubusercontent.com/your-repo/install_agent.sh | bash -s -- your-api-key base_url

Replace your-repo with your repository URL and your-api-key with the appropriate key for the server.

An installation script like this simplifies deployments, reduces errors, and ensures consistency across servers. Whether you’re deploying a monitoring agent, a web application, or any server-side tool, the principles remain the same.

By combining automation with flexibility, you make life easier not only for yourself but also for anyone else who needs to use your software.

Feel free to customize this script and adapt it to your needs.