Disk Space Monitor

  1#!/bin/bash
  2
  3#############################################################################
  4# Disk Space Monitor Script
  5#
  6# Description: Monitors disk space usage and alerts when threshold exceeded
  7# Author: glyph.sh
  8# Reference: https://glyph.sh/scripts/disk-space-monitor/
  9#
 10# Features:
 11# - Check all mounted filesystems
 12# - Configurable alert threshold (default 90%)
 13# - Email alerts or syslog logging
 14# - Identify largest directories consuming space
 15# - Exclude virtual/temporary filesystems
 16#############################################################################
 17
 18# Configuration
 19THRESHOLD=${THRESHOLD:-90}                    # Alert threshold (percentage)
 20EMAIL_TO=${EMAIL_TO:-"admin@example.com"}     # Email recipient
 21EMAIL_SUBJECT="[ALERT] Disk Space Warning"
 22USE_SYSLOG=${USE_SYSLOG:-true}               # Log to syslog (true/false)
 23USE_EMAIL=${USE_EMAIL:-false}                # Send email alerts (true/false)
 24TOP_DIRS=${TOP_DIRS:-10}                     # Number of largest dirs to show
 25MIN_DIR_SIZE=${MIN_DIR_SIZE:-100}            # Minimum dir size in MB to report
 26
 27# Filesystems to exclude from monitoring
 28EXCLUDE_FS="tmpfs|devtmpfs|squashfs|overlay|none|udev|cgroup"
 29
 30# Colors for output
 31RED='\033[0;31m'
 32YELLOW='\033[1;33m'
 33GREEN='\033[0;32m'
 34BLUE='\033[0;34m'
 35NC='\033[0m' # No Color
 36
 37#############################################################################
 38# Functions
 39#############################################################################
 40
 41print_header() {
 42    echo -e "${BLUE}========================================${NC}"
 43    echo -e "${BLUE} Disk Space Monitor${NC}"
 44    echo -e "${BLUE}========================================${NC}"
 45    echo ""
 46    echo "Threshold: ${THRESHOLD}%"
 47    echo "Date: $(date '+%Y-%m-%d %H:%M:%S')"
 48    echo ""
 49}
 50
 51log_message() {
 52    local level="$1"
 53    local message="$2"
 54
 55    # Log to syslog if enabled
 56    if [ "$USE_SYSLOG" = true ]; then
 57        logger -t "disk-space-monitor" -p "user.${level}" "$message"
 58    fi
 59
 60    # Print to console
 61    case "$level" in
 62        "warning")
 63            echo -e "${YELLOW}[WARNING]${NC} $message"
 64            ;;
 65        "error")
 66            echo -e "${RED}[ERROR]${NC} $message"
 67            ;;
 68        "info")
 69            echo -e "${GREEN}[INFO]${NC} $message"
 70            ;;
 71        *)
 72            echo "$message"
 73            ;;
 74    esac
 75}
 76
 77send_email_alert() {
 78    local subject="$1"
 79    local body="$2"
 80
 81    if [ "$USE_EMAIL" = true ] && command -v mail >/dev/null 2>&1; then
 82        echo "$body" | mail -s "$subject" "$EMAIL_TO"
 83        log_message "info" "Email alert sent to $EMAIL_TO"
 84    elif [ "$USE_EMAIL" = true ]; then
 85        log_message "error" "mail command not found. Cannot send email alerts."
 86    fi
 87}
 88
 89check_disk_space() {
 90    local alert_triggered=false
 91    local alert_message=""
 92
 93    echo -e "${BLUE}Checking filesystem usage...${NC}"
 94    echo ""
 95
 96    # Header
 97    printf "%-20s %8s %8s %8s %5s %s\n" "Filesystem" "Size" "Used" "Avail" "Use%" "Mounted"
 98    echo "--------------------------------------------------------------------------------"
 99
100    # Get filesystem usage, exclude virtual/temporary filesystems
101    while IFS= read -r line; do
102        # Parse df output
103        filesystem=$(echo "$line" | awk '{print $1}')
104        size=$(echo "$line" | awk '{print $2}')
105        used=$(echo "$line" | awk '{print $3}')
106        avail=$(echo "$line" | awk '{print $4}')
107        percent=$(echo "$line" | awk '{print $5}' | tr -d '%')
108        mountpoint=$(echo "$line" | awk '{print $6}')
109
110        # Skip if percent is not a number
111        if ! [[ "$percent" =~ ^[0-9]+$ ]]; then
112            continue
113        fi
114
115        # Determine status color
116        if [ "$percent" -ge "$THRESHOLD" ]; then
117            color=$RED
118            status="ALERT"
119            alert_triggered=true
120            alert_message="${alert_message}\n${filesystem} mounted on ${mountpoint}: ${percent}% used (${used} / ${size})"
121        elif [ "$percent" -ge $((THRESHOLD - 10)) ]; then
122            color=$YELLOW
123            status="WARN"
124        else
125            color=$GREEN
126            status="OK"
127        fi
128
129        # Print filesystem info
130        printf "${color}%-20s %8s %8s %8s %4s%% %s${NC}\n" \
131            "$filesystem" "$size" "$used" "$avail" "$percent" "$mountpoint"
132
133        # Log if over threshold
134        if [ "$percent" -ge "$THRESHOLD" ]; then
135            log_message "warning" "Filesystem $filesystem on $mountpoint is ${percent}% full"
136        fi
137    done < <(df -h 2>/dev/null | tail -n +2 | grep -vE "$EXCLUDE_FS")
138
139    echo ""
140
141    # Send alert if threshold exceeded
142    if [ "$alert_triggered" = true ]; then
143        local email_body="WARNING: Disk space usage has exceeded ${THRESHOLD}%\n\nAffected filesystems:${alert_message}\n\nHostname: $(hostname)\nDate: $(date)\n\nPlease investigate and free up space."
144        send_email_alert "$EMAIL_SUBJECT" "$(echo -e "$email_body")"
145    fi
146}
147
148find_largest_directories() {
149    echo -e "${BLUE}Finding largest directories...${NC}"
150    echo ""
151
152    # Find directories larger than MIN_DIR_SIZE
153    # This can take time on large filesystems
154    local search_paths=("/var" "/home" "/usr" "/opt" "/tmp")
155
156    for path in "${search_paths[@]}"; do
157        if [ -d "$path" ]; then
158            echo -e "${YELLOW}Analyzing: ${path}${NC}"
159
160            # Use du to find largest directories
161            # -x: stay on one filesystem (don't cross mount points)
162            # -h: human readable
163            # --max-depth=2: limit depth to avoid long scans
164            du -x -h --max-depth=2 "$path" 2>/dev/null | \
165                sort -rh | \
166                head -n "$TOP_DIRS" | \
167                while read -r size dir; do
168                    # Convert size to MB for comparison
169                    size_mb=$(echo "$size" | sed 's/[^0-9.]//g')
170                    unit=$(echo "$size" | sed 's/[0-9.]//g')
171
172                    # Convert to MB
173                    case "$unit" in
174                        G|g) size_mb=$(echo "$size_mb * 1024" | bc 2>/dev/null || echo "$size_mb") ;;
175                        K|k) size_mb=$(echo "$size_mb / 1024" | bc 2>/dev/null || echo "$size_mb") ;;
176                    esac
177
178                    printf "  %8s  %s\n" "$size" "$dir"
179                done
180            echo ""
181        fi
182    done
183}
184
185show_inode_usage() {
186    echo -e "${BLUE}Checking inode usage...${NC}"
187    echo ""
188
189    printf "%-20s %12s %12s %12s %5s %s\n" "Filesystem" "Inodes" "IUsed" "IFree" "IUse%" "Mounted"
190    echo "--------------------------------------------------------------------------------"
191
192    # Get inode usage
193    while IFS= read -r line; do
194        filesystem=$(echo "$line" | awk '{print $1}')
195        inodes=$(echo "$line" | awk '{print $2}')
196        iused=$(echo "$line" | awk '{print $3}')
197        ifree=$(echo "$line" | awk '{print $4}')
198        ipercent=$(echo "$line" | awk '{print $5}' | tr -d '%')
199        mountpoint=$(echo "$line" | awk '{print $6}')
200
201        # Skip if percent is not a number
202        if ! [[ "$ipercent" =~ ^[0-9]+$ ]]; then
203            continue
204        fi
205
206        # Determine status color
207        if [ "$ipercent" -ge "$THRESHOLD" ]; then
208            color=$RED
209            log_message "warning" "Inode usage on $filesystem ($mountpoint) is ${ipercent}% full"
210        elif [ "$ipercent" -ge $((THRESHOLD - 10)) ]; then
211            color=$YELLOW
212        else
213            color=$GREEN
214        fi
215
216        printf "${color}%-20s %12s %12s %12s %4s%% %s${NC}\n" \
217            "$filesystem" "$inodes" "$iused" "$ifree" "$ipercent" "$mountpoint"
218    done < <(df -i 2>/dev/null | tail -n +2 | grep -vE "$EXCLUDE_FS")
219
220    echo ""
221}
222
223print_summary() {
224    echo -e "${BLUE}========================================${NC}"
225    echo -e "${BLUE} Summary${NC}"
226    echo -e "${BLUE}========================================${NC}"
227    echo ""
228    echo "Monitoring completed at $(date '+%Y-%m-%d %H:%M:%S')"
229    echo "Threshold: ${THRESHOLD}%"
230    echo ""
231
232    # Count filesystems over threshold
233    local count=$(df -h 2>/dev/null | tail -n +2 | grep -vE "$EXCLUDE_FS" | awk '{print $5}' | tr -d '%' | awk -v t="$THRESHOLD" '$1 >= t {count++} END {print count+0}')
234
235    if [ "$count" -gt 0 ]; then
236        echo -e "${RED}WARNING: $count filesystem(s) over ${THRESHOLD}% capacity${NC}"
237    else
238        echo -e "${GREEN}All filesystems are below ${THRESHOLD}% capacity${NC}"
239    fi
240    echo ""
241}
242
243show_help() {
244    cat << EOF
245Usage: $0 [OPTIONS]
246
247Disk Space Monitoring Script
248
249OPTIONS:
250    -t, --threshold NUM      Alert threshold percentage (default: 90)
251    -e, --email ADDRESS      Email address for alerts
252    -s, --syslog             Enable syslog logging (default: true)
253    -m, --mail               Enable email alerts (default: false)
254    -d, --top-dirs NUM       Number of largest directories to show (default: 10)
255    -i, --inodes             Show inode usage
256    -h, --help               Display this help message
257
258EXAMPLES:
259    # Basic usage with defaults
260    $0
261
262    # Set custom threshold
263    $0 --threshold 85
264
265    # Enable email alerts
266    $0 --email admin@example.com --mail
267
268    # Show top 20 largest directories
269    $0 --top-dirs 20
270
271ENVIRONMENT VARIABLES:
272    THRESHOLD        Alert threshold percentage
273    EMAIL_TO         Email recipient address
274    USE_SYSLOG       Enable syslog (true/false)
275    USE_EMAIL        Enable email alerts (true/false)
276    TOP_DIRS         Number of top directories to show
277    MIN_DIR_SIZE     Minimum directory size in MB to report
278
279EOF
280    exit 0
281}
282
283#############################################################################
284# Main Script
285#############################################################################
286
287# Parse command line arguments
288SHOW_INODES=false
289
290while [[ $# -gt 0 ]]; do
291    case $1 in
292        -t|--threshold)
293            THRESHOLD="$2"
294            shift 2
295            ;;
296        -e|--email)
297            EMAIL_TO="$2"
298            shift 2
299            ;;
300        -s|--syslog)
301            USE_SYSLOG=true
302            shift
303            ;;
304        -m|--mail)
305            USE_EMAIL=true
306            shift
307            ;;
308        -d|--top-dirs)
309            TOP_DIRS="$2"
310            shift 2
311            ;;
312        -i|--inodes)
313            SHOW_INODES=true
314            shift
315            ;;
316        -h|--help)
317            show_help
318            ;;
319        *)
320            echo "Unknown option: $1"
321            echo "Use -h or --help for usage information"
322            exit 1
323            ;;
324    esac
325done
326
327# Validate threshold
328if ! [[ "$THRESHOLD" =~ ^[0-9]+$ ]] || [ "$THRESHOLD" -lt 1 ] || [ "$THRESHOLD" -gt 100 ]; then
329    echo "Error: Threshold must be between 1 and 100"
330    exit 1
331fi
332
333# Main execution
334print_header
335check_disk_space
336echo ""
337find_largest_directories
338echo ""
339
340if [ "$SHOW_INODES" = true ]; then
341    show_inode_usage
342fi
343
344print_summary
345
346log_message "info" "Disk space monitoring completed"
347
348exit 0

1chmod +x disk-space-monitor.sh
2./disk-space-monitor.sh

1# Alert when usage exceeds 85%
2./disk-space-monitor.sh --threshold 85

1# Send alerts to specific email
2./disk-space-monitor.sh --email admin@example.com --mail

1# Show top 20 largest directories
2./disk-space-monitor.sh --top-dirs 20

1# Include inode usage in report
2./disk-space-monitor.sh --inodes

1# Configure via environment variables
2export THRESHOLD=85
3export EMAIL_TO="admin@example.com"
4export USE_EMAIL=true
5export TOP_DIRS=15
6./disk-space-monitor.sh

 1# Edit crontab
 2crontab -e
 3
 4# Check disk space every hour
 50 * * * * /usr/local/bin/disk-space-monitor.sh --threshold 90 --syslog
 6
 7# Check every 6 hours with email alerts
 80 */6 * * * /usr/local/bin/disk-space-monitor.sh --threshold 90 --email admin@example.com --mail
 9
10# Daily detailed report with largest directories
110 8 * * * /usr/local/bin/disk-space-monitor.sh --top-dirs 20 --inodes --email admin@example.com --mail

 1# /etc/systemd/system/disk-space-monitor.service
 2[Unit]
 3Description=Disk Space Monitor
 4After=network.target
 5
 6[Service]
 7Type=oneshot
 8ExecStart=/usr/local/bin/disk-space-monitor.sh --threshold 90 --syslog
 9User=root
10
11[Install]
12WantedBy=multi-user.target

 1# /etc/systemd/system/disk-space-monitor.timer
 2[Unit]
 3Description=Run Disk Space Monitor hourly
 4
 5[Timer]
 6OnCalendar=hourly
 7Persistent=true
 8
 9[Install]
10WantedBy=timers.target

1systemctl daemon-reload
2systemctl enable disk-space-monitor.timer
3systemctl start disk-space-monitor.timer

1# Debian/Ubuntu
2sudo apt-get install mailutils
3
4# RHEL/CentOS
5sudo yum install mailx
6
7# Configure with local MTA or relay host

1echo "Test message" | mail -s "Test Alert" admin@example.com

 1# Find large files
 2find /var -type f -size +100M -exec ls -lh {} \;
 3
 4# Find old log files
 5find /var/log -type f -mtime +30
 6
 7# Check for core dumps
 8find / -name "core.*" -type f 2>/dev/null
 9
10# Analyze package cache (Debian/Ubuntu)
11du -sh /var/cache/apt/archives/
12
13# Clean package cache (RHEL/CentOS)
14du -sh /var/cache/yum/

1# Add at end of script
2echo "# HELP disk_usage_percent Disk usage percentage" > /var/lib/node_exporter/disk_usage.prom
3df -h | grep -vE "$EXCLUDE_FS" | tail -n +2 | while read line; do
4    mount=$(echo $line | awk '{print $6}')
5    usage=$(echo $line | awk '{print $5}' | tr -d '%')
6    echo "disk_usage_percent{mountpoint=\"$mount\"} $usage" >> /var/lib/node_exporter/disk_usage.prom
7done

1# Add function for webhook alerts
2send_webhook_alert() {
3    local message="$1"
4    local webhook_url="YOUR_WEBHOOK_URL"
5
6    curl -X POST "$webhook_url" \
7        -H 'Content-Type: application/json' \
8        -d "{\"text\":\"$message\"}"
9}

1curl -O https://glyph.sh/scripts/disk-space-monitor.sh
2chmod +x disk-space-monitor.sh
3./disk-space-monitor.sh --help