Sysadmin_scripts

Chapter 36: System Administration Scripts

Overview

This chapter covers practical system administration scripts that DevOps engineers use daily. These scripts help manage users, services, system monitoring, and common administrative tasks.

---

User Management Scripts

Create User with Home Directory

#!/usr/bin/env bash
# create_user.sh - Create system user with home directory

set -euo pipefail

USAGE="Usage: $0 <username> [shell]"

username="${1:-}"
shell="${2:-/bin/bash}"

# Validate input
if [[ -z "$username" ]]; then
    echo "$USAGE"
    exit 1
fi

# Check if user exists
if id "$username" &>/dev/null; then
    echo "Error: User $username already exists"
    exit 1
fi

# Create user with home directory and shell
sudo useradd -m -s "$shell" "$username"

# Set password
echo "Enter password for $username:"
sudo passwd "$username"

echo "User $username created successfully"

List Users by Login Shell

#!/usr/bin/env bash
# list_users_by_shell.sh - List users by their login shell

set -euo pipefail

shell="${1:-/bin/bash}"

echo "Users using shell: $shell"
echo "======================"

# Get users with specific shell
awk -F: -v shell="$shell" '$7 == shell {print $1, $6}' /etc/passwd | \
    while read -r user home; do
        echo "User: $user, Home: $home"
    done

---

Service Management Scripts

Service Health Monitor

#!/usr/bin/env bash
# monitor_services.sh - Monitor critical services

set -euo pipefail

# Define services to monitor
SERVICES=(
    "nginx"
    "postgresql"
    "redis"
    "docker"
)

log() { echo "[$(date)] $*"; }

check_service() {
    local service="$1"

    if systemctl is-active --quiet "$service"; then
        log "✓ $service is running"
        return 0
    else
        log "✗ $service is NOT running"
        return 1
    fi
}

main() {
    local failed=0

    log "Checking services..."

    for service in "${SERVICES[@]}"; do
        check_service "$service" || ((failed++))
    done

    if [[ $failed -gt 0 ]]; then
        log "WARNING: $failed service(s) not running"
        exit 1
    fi

    log "All services running"
}

main

Restart Service with Health Check

#!/usr/bin/env bash
# restart_service.sh - Restart service with health verification

set -euo pipefail

SERVICE="${1:-}"
HEALTH_URL="${HEALTH_URL:-http://localhost:8080/health}"
MAX_WAIT=60

log() { echo "[$(date)] $*"; }

usage() {
    echo "Usage: $0 <service> [health-url]"
    echo "Example: $0 nginx"
    exit 1
}

# Validate arguments
[[ -z "$SERVICE" ]] && usage

log "Restarting service: $SERVICE"

# Stop service
sudo systemctl stop "$SERVICE"
log "Service stopped"

# Start service
sudo systemctl start "$SERVICE"
log "Service started"

# Wait for health check
log "Waiting for health check..."
elapsed=0

while [[ $elapsed -lt $MAX_WAIT ]]; do
    if curl -sf "$HEALTH_URL" &>/dev/null; then
        log "Service is healthy"
        exit 0
    fi
    sleep 2
    ((elapsed += 2))
done

log "Health check failed after ${MAX_WAIT}s"
sudo systemctl status "$SERVICE"
exit 1

---

System Monitoring Scripts

System Resource Report

#!/usr/bin/env bash
# system_report.sh - Generate comprehensive system report

set -euo pipefail

REPORT_FILE="${1:-/tmp/system_report.txt}"

log() { echo "$*" | tee -a "$REPORT_FILE"; }

generate_report() {
    : > "$REPORT_FILE"

    log "=== System Report ==="
    log "Generated: $(date)"
    log ""

    # System info
    log "--- System Information ---"
    log "Hostname: $(hostname)"
    log "Kernel: $(uname -r)"
    log "OS: $(lsb_release -d 2>/dev/null | cut -f2 || cat /etc/os-release | grep PRETTY_NAME | cut -d= -f2)"
    log "Uptime: $(uptime -p)"
    log ""

    # CPU info
    log "--- CPU Information ---"
    log "Model: $(grep 'model name' /proc/cpuinfo | head -1 | cut -d: -f2 | xargs)"
    log "Cores: $(nproc)"
    log "Load Average: $(uptime | awk -F'load average:' '{print $2}')"
    log ""

    # Memory info
    log "--- Memory Information ---"
    free -h | tee -a "$REPORT_FILE"
    log ""

    # Disk info
    log "--- Disk Usage ---"
    df -h | grep -E '^/dev' | tee -a "$REPORT_FILE"
    log ""

    # Top processes
    log "--- Top 5 Processes by Memory ---"
    ps aux --sort=-%mem | head -6 | tee -a "$REPORT_FILE"
    log ""

    # Network connections
    log "--- Network Connections ---"
    log "Established: $(ss -tun | grep ESTAB | wc -l)"
    log "Listening: $(ss -tun | grep LISTEN | wc -l)"

    log ""
    log "Report saved to: $REPORT_FILE"
}

generate_report

Disk Space Alert

#!/usr/bin/env bash
# disk_alert.sh - Alert when disk space is low

set -euo pipefail

THRESHOLD="${1:-90}"

log() { echo "[$(date)] $*"; }

check_disk() {
    local usage
    usage=$(df / | awk 'NR==2 {print $5}' | tr -d '%')

    if [[ $usage -ge $THRESHOLD ]]; then
        log "ALERT: Disk usage at ${usage}%"

        # Show detailed info
        df -h

        # Find large directories
        log "Large directories:"
        du -ahx / 2>/dev/null | sort -rh | head -10

        return 1
    fi

    log "Disk usage OK: ${usage}%"
    return 0
}

check_disk

---

Network Administration Scripts

Network Connectivity Check

#!/usr/bin/env bash
# network_check.sh - Check network connectivity

set -euo pipefail

HOSTS=(
    "8.8.8.8"
    "google.com"
    "cloudflare.com"
)

log() { echo "[$(date)] $*"; }

check_ping() {
    local host="$1"

    if ping -c 1 -W 2 "$host" &>/dev/null; then
        log "✓ $host is reachable"
        return 0
    else
        log "✗ $host is NOT reachable"
        return 1
    fi
}

main() {
    local failed=0

    log "Checking network connectivity..."

    for host in "${HOSTS[@]}"; do
        check_ping "$host" || ((failed++))
    done

    if [[ $failed -gt 0 ]]; then
        log "WARNING: $failed host(s) unreachable"
        exit 1
    fi

    log "All hosts reachable"
}

main

Port Scanner

#!/usr/bin/env bash
# port_scan.sh - Simple port scanner

set -euo pipefail

HOST="${1:-localhost}"
START_PORT="${2:-1}"
END_PORT="${3:-1024}"

log() { echo "Scanning $HOST ports $START_PORT-$END_PORT..."; }

scan_ports() {
    for port in $(seq $START_PORT $END_PORT); do
        (echo >/dev/tcp/$HOST/$port) 2>/dev/null && \
            echo "Port $port: OPEN"
    done
}

log
scan_ports

---

Log Management Scripts

Log Rotation

#!/usr/bin/env bash
# rotate_logs.sh - Rotate application logs

set -euo pipefail

LOG_DIR="${LOG_DIR:-/var/log/myapp}"
MAX_DAYS="${MAX_DAYS:-30}"
MAX_SIZE="${MAX_SIZE:-100M}"

log() { echo "[$(date)] $*"; }

rotate_logs() {
    log "Starting log rotation"
    log "Directory: $LOG_DIR"
    log "Max days: $MAX_DAYS"

    # Find and compress old logs
    find "$LOG_DIR" -name "*.log" -type f -mtime +$MAX_DAYS -exec gzip {} \;

    # Remove very old compressed logs
    find "$LOG_DIR" -name "*.log.gz" -type f -mtime +$((MAX_DAYS * 2)) -delete

    # Compress large logs
    find "$LOG_DIR" -name "*.log" -type f -size +$MAX_SIZE -exec gzip {} \;

    log "Log rotation complete"
}

rotate_logs

---

Package Management Scripts

Update System Script

#!/usr/bin/env bash
# update_system.sh - Update Arch Linux system

set -euo pipefail

log() { echo "[$(date)] $*"; }

error() { echo "[$(date)] ERROR: $*" >&2; }

# Check for root
if [[ $EUID -ne 0 ]]; then
   error "This script must be run as root"
   exit 1
fi

log "Starting system update..."

# Sync databasescing databases..."
pac
log "Synman -Sy

# Upgrade system
log "Upgrading packages..."
pacman -Syu --noconfirm

# Clean up
log "Cleaning up..."
pacman -Scc --noconfirm

# Check for orphans
if pacman -Qdtq &>/dev/null; then
    log "Removing orphan packages..."
    pacman -Rns $(pacman -Qdtq) --noconfirm
fi

log "System update complete"

---

Backup and Sync Scripts

Directory Backup

#!/usr/bin/env bash
# backup_dir.sh - Backup directories

set -euo pipefail

SOURCE_DIR="${1:-}"
BACKUP_DIR="${2:-/backup}"

log() { echo "[$(date)] $*"; }

usage() {
    echo "Usage: $0 <source_directory> [backup_directory]"
    exit 1
}

[[ -z "$SOURCE_DIR" ]] && usage

# Create timestamped backup
timestamp=$(date +%Y%m%d_%H%M%S)
backup_name="$(basename "$SOURCE_DIR")_$timestamp.tar.gz"
backup_path="$BACKUP_DIR/$backup_name"

log "Creating backup: $backup_path"

mkdir -p "$BACKUP_DIR"

# Create compressed archive
tar -czf "$backup_path" -C "$(dirname "$SOURCE_DIR")" "$(basename "$SOURCE_DIR")"

# Show backup size
log "Backup size: $(du -h "$backup_path" | cut -f1)"

# Cleanup old backups (keep last 7)
cd "$BACKUP_DIR"
ls -t *.tar.gz | tail -n +8 | xargs -r rm

log "Backup complete: $backup_path"

---

Summary

In this chapter, you learned:

• ✅ User management scripts
• ✅ Service monitoring and management
• ✅ System resource monitoring
• ✅ Disk space alerts
• ✅ Network connectivity checks
• ✅ Port scanning
• ✅ Log rotation
• ✅ System updates
• ✅ Directory backups

---

Next Steps

Continue to the next chapter to learn about Backup and Restore Scripts.

---

Previous Chapter: Code Style and Organization Next Chapter: Backup and Restore Scripts