#!/bin/bash
# =============================================================================
# NAS08 Weekly TRIM Report Script
# =============================================================================
# Runs fstrim on all mounted filesystems, captures results, gathers SMART
# health data for each SSD, and sends an HTML email report matching the
# KingDezigns backup monitor style.
#
# Triggered by: systemd fstrim.service (via drop-in override)
# Schedule: Weekly (Monday ~12:33 AM — managed by fstrim.timer)
#
# Script location: /usr/scripts/omv/fstrim-report.sh
# Log location: /var/log/fstrim/
# =============================================================================
set -uo pipefail
# -----------------------------------------------------------------------------
# CONFIGURATION
# -----------------------------------------------------------------------------
HOSTNAME_LABEL=$(hostname -s 2>/dev/null || echo "NAS08")
REPORT_TO="rufus.king@kingdezigns.com"
REPORT_FROM="trim-monitor@nas08.local" # Display only — OMV uses your configured SMTP relay
LOG_DIR="/var/log/fstrim"
KEEP_LOGS=90 # days to retain log files
DATE=$(date +%Y-%m-%d)
REPORT_TIME=$(date "+%Y-%m-%d %H:%M:%S")
LOG_FILE="${LOG_DIR}/fstrim-${DATE}.log"
# -----------------------------------------------------------------------------
# STATUS TRACKING
# OVERALL_STATUS: SUCCESS | WARNING | FAILED
# -----------------------------------------------------------------------------
OVERALL_STATUS="SUCCESS"
SECTION_ROWS_HTML=""
WARNING_COUNT=0
ERROR_COUNT=0
# Append a result row to the section table
# Usage: add_section_row "Section name" "STATUS" "Detail message"
# STATUS: OK | WARNING | ERROR | SKIPPED
add_section_row() {
local section="$1" status="$2" detail="$3"
case "$status" in
OK)
BADGE="✓ OK"
;;
WARNING)
BADGE="⚠ WARNING"
WARNING_COUNT=$((WARNING_COUNT + 1))
[[ "$OVERALL_STATUS" == "SUCCESS" ]] && OVERALL_STATUS="WARNING"
;;
ERROR)
BADGE="✗ ERROR"
ERROR_COUNT=$((ERROR_COUNT + 1))
OVERALL_STATUS="FAILED"
;;
SKIPPED)
BADGE="— SKIPPED"
;;
esac
SECTION_ROWS_HTML+="
| $section |
$BADGE |
$detail |
"
}
# -----------------------------------------------------------------------------
# LOGGING
# -----------------------------------------------------------------------------
mkdir -p "$LOG_DIR"
log() {
echo "[$(date '+%Y-%m-%d %H:%M:%S')] $1" | tee -a "$LOG_FILE"
}
log_section() {
echo "" | tee -a "$LOG_FILE"
echo "------------------------------------------------------------" | tee -a "$LOG_FILE"
log "$1"
echo "------------------------------------------------------------" | tee -a "$LOG_FILE"
}
# -----------------------------------------------------------------------------
# SECTION 1: RUN FSTRIM
# -----------------------------------------------------------------------------
log_section "Running fstrim on all mounted filesystems"
TRIM_RAW=$(fstrim -av 2>&1)
TRIM_EXIT=$?
TRIM_ROWS_HTML=""
TOTAL_BYTES=0
VOLUME_COUNT=0
while IFS= read -r line; do
if [[ "$line" =~ ^(/.+):[[:space:]]+(.+)[[:space:]]+\(([0-9]+)[[:space:]]bytes\) ]]; then
MOUNT="${BASH_REMATCH[1]}"
HUMAN="${BASH_REMATCH[2]}"
BYTES="${BASH_REMATCH[3]}"
TOTAL_BYTES=$((TOTAL_BYTES + BYTES))
VOLUME_COUNT=$((VOLUME_COUNT + 1))
log "Trimmed ${MOUNT}: ${HUMAN}"
TRIM_ROWS_HTML+="
| $MOUNT |
$HUMAN freed |
"
fi
done <<< "$TRIM_RAW"
# Convert total bytes to human readable
if [ $TOTAL_BYTES -ge 1073741824 ]; then
TOTAL_HUMAN=$(awk "BEGIN {printf \"%.1f GiB\", $TOTAL_BYTES/1073741824}")
elif [ $TOTAL_BYTES -ge 1048576 ]; then
TOTAL_HUMAN=$(awk "BEGIN {printf \"%.1f MiB\", $TOTAL_BYTES/1048576}")
else
TOTAL_HUMAN="${TOTAL_BYTES} bytes"
fi
if [ $TRIM_EXIT -eq 0 ]; then
log "fstrim completed — $VOLUME_COUNT volume(s) · $TOTAL_HUMAN total freed"
add_section_row "TRIM Execution" "OK" "$VOLUME_COUNT volume(s) processed · $TOTAL_HUMAN total freed"
else
log "ERROR: fstrim exited with code $TRIM_EXIT"
add_section_row "TRIM Execution" "ERROR" "fstrim exited with code $TRIM_EXIT"
fi
# -----------------------------------------------------------------------------
# SECTION 2: SSD SMART HEALTH
# -----------------------------------------------------------------------------
log_section "Gathering SSD SMART health data"
SMART_ROWS_HTML=""
SMART_WARN=0
SMART_ERR=0
for DISK in $(lsblk -dpno NAME | grep -E '^/dev/sd|^/dev/nvme'); do
MODEL=$(smartctl -i "$DISK" 2>/dev/null | awk -F': ' '/Device Model|Model Number/{gsub(/^[ \t]+|[ \t]+$/, "", $2); print $2; exit}')
SERIAL=$(smartctl -i "$DISK" 2>/dev/null | awk -F': ' '/Serial Number/{gsub(/^[ \t]+|[ \t]+$/, "", $2); print $2; exit}')
CAPACITY=$(lsblk -dno SIZE "$DISK" 2>/dev/null)
HEALTH=$(smartctl -H "$DISK" 2>/dev/null | awk '/overall-health|test result/{print $NF}')
# Temperature: +0 forces numeric conversion, handling "32 (Min/Max 25/56)" Crucial format
TEMP=$(smartctl -A "$DISK" 2>/dev/null | awk '/Temperature_Celsius|Temperature_Internal/{print $10+0; exit}')
POWER_ON=$(smartctl -A "$DISK" 2>/dev/null | awk '/Power_On_Hours/{print $10; exit}')
# Wear: PNY=SSD_Life_Left, Crucial=Percent_Lifetime_Remain, Samsung/WD=Wear_Leveling_Count, Intel=Media_Wearout_Indicator
WEAR=$(smartctl -A "$DISK" 2>/dev/null | awk '/SSD_Life_Left|Percent_Lifetime_Remain|Wear_Leveling_Count|Media_Wearout_Indicator/{print $10; exit}')
# Reallocated: PNY=Bad_Blk_Ct_Lat/Erl (RAW_VALUE is "0/88" — take first number), Crucial=Reallocate_NAND_Blk_Cnt, others=Reallocated_Sector_Ct
REALLOCATED=$(smartctl -A "$DISK" 2>/dev/null | awk '/Bad_Blk_Ct_Lat\/Erl/{split($10,a,"/"); print a[1]; exit} /Reallocate_NAND_Blk_Cnt|Reallocated_Sector_Ct|Reallocated_Sector/{print $10; exit}')
# Pending: Crucial=Current_Pending_ECC_Cnt, others=Current_Pending_Sector (PNY has no equivalent — will show N/A)
PENDING=$(smartctl -A "$DISK" 2>/dev/null | awk '/Current_Pending_ECC_Cnt|Current_Pending_Sector/{print $10; exit}')
# Uncorrectable: Crucial=Reported_Uncorrect, others=Offline_Uncorrectable (PNY has no equivalent — will show N/A)
UNCORRECTABLE=$(smartctl -A "$DISK" 2>/dev/null | awk '/Reported_Uncorrect|Offline_Uncorrectable/{print $10; exit}')
[ -z "$MODEL" ] && MODEL="Unknown"
[ -z "$SERIAL" ] && SERIAL="Unknown"
[ -z "$HEALTH" ] && HEALTH="Unknown"
[ -z "$TEMP" ] && TEMP="N/A"
[ -z "$POWER_ON" ] && POWER_ON="N/A"
[ -z "$WEAR" ] && WEAR="N/A"
[ -z "$REALLOCATED" ] && REALLOCATED="N/A"
[ -z "$PENDING" ] && PENDING="N/A"
[ -z "$UNCORRECTABLE" ] && UNCORRECTABLE="N/A"
# Health colour
if [ "$HEALTH" = "PASSED" ]; then
HEALTH_COLOUR="#065f46"
HEALTH_BG="#d1fae5"
DISK_STATUS="OK"
else
HEALTH_COLOUR="#991b1b"
HEALTH_BG="#fee2e2"
DISK_STATUS="WARNING"
SMART_WARN=$((SMART_WARN + 1))
fi
# Flag non-zero reallocated/pending/uncorrectable sectors
SECTOR_NOTE=""
if [ "$REALLOCATED" != "N/A" ] && [ "$REALLOCATED" -gt 0 ] 2>/dev/null; then
SECTOR_NOTE+=" ⚠ Reallocated: $REALLOCATED"
SMART_WARN=$((SMART_WARN + 1))
[[ "$OVERALL_STATUS" == "SUCCESS" ]] && OVERALL_STATUS="WARNING"
fi
if [ "$PENDING" != "N/A" ] && [ "$PENDING" -gt 0 ] 2>/dev/null; then
SECTOR_NOTE+=" ⚠ Pending: $PENDING"
SMART_WARN=$((SMART_WARN + 1))
[[ "$OVERALL_STATUS" == "SUCCESS" ]] && OVERALL_STATUS="WARNING"
fi
if [ "$UNCORRECTABLE" != "N/A" ] && [ "$UNCORRECTABLE" -gt 0 ] 2>/dev/null; then
SECTOR_NOTE+=" 🚨 Uncorrectable: $UNCORRECTABLE"
SMART_ERR=$((SMART_ERR + 1))
OVERALL_STATUS="FAILED"
fi
log "Drive $DISK ($MODEL) — Health: $HEALTH | Temp: ${TEMP}°C | Wear: $WEAR | Power-on: ${POWER_ON}h${SECTOR_NOTE}"
SMART_ROWS_HTML+="
|
$DISK $MODEL · $CAPACITY · S/N: $SERIAL
|
| Overall health |
$HEALTH${SECTOR_NOTE:+$SECTOR_NOTE} |
| Temperature |
${TEMP}°C |
| Power-on hours |
${POWER_ON} hrs |
| Wear leveling |
$WEAR |
| Sectors (realloc / pending / uncorrectable) |
$REALLOCATED / $PENDING / $UNCORRECTABLE |
"
done
if [ $SMART_ERR -gt 0 ]; then
add_section_row "SSD SMART Health" "ERROR" "$SMART_ERR drive(s) with uncorrectable sectors — immediate attention required"
elif [ $SMART_WARN -gt 0 ]; then
add_section_row "SSD SMART Health" "WARNING" "$SMART_WARN warning(s) — review drive details below"
else
add_section_row "SSD SMART Health" "OK" "All drives PASSED · no sector errors detected"
fi
# -----------------------------------------------------------------------------
# SECTION 3: FILESYSTEM USAGE
# -----------------------------------------------------------------------------
log_section "Filesystem usage"
FS_ROWS_HTML=""
FS_WARN=0
while IFS= read -r line; do
USE_PCT=$(echo "$line" | awk '{print $5}' | tr -d '%')
MOUNT_PT=$(echo "$line" | awk '{print $6}')
USED=$(echo "$line" | awk '{print $3}')
SIZE=$(echo "$line" | awk '{print $2}')
if [ "$USE_PCT" -ge 90 ] 2>/dev/null; then
ROW_COLOUR="#fee2e2"; PCT_COLOUR="#991b1b"
FS_WARN=$((FS_WARN + 1))
[[ "$OVERALL_STATUS" == "SUCCESS" ]] && OVERALL_STATUS="WARNING"
elif [ "$USE_PCT" -ge 75 ] 2>/dev/null; then
ROW_COLOUR="#fef3c7"; PCT_COLOUR="#92400e"
FS_WARN=$((FS_WARN + 1))
[[ "$OVERALL_STATUS" == "SUCCESS" ]] && OVERALL_STATUS="WARNING"
else
ROW_COLOUR="transparent"; PCT_COLOUR="#065f46"
fi
log "Filesystem $MOUNT_PT: ${USED} used of ${SIZE} (${USE_PCT}%)"
FS_ROWS_HTML+="
| $MOUNT_PT |
$USED used of $SIZE |
${USE_PCT}% |
"
done < <(df -h | grep -E '^/dev/')
if [ $FS_WARN -gt 0 ]; then
add_section_row "Filesystem Usage" "WARNING" "$FS_WARN volume(s) above 75% — review recommended"
else
add_section_row "Filesystem Usage" "OK" "All volumes within normal range"
fi
# -----------------------------------------------------------------------------
# ROTATE OLD LOGS
# -----------------------------------------------------------------------------
find "$LOG_DIR" -name "fstrim-*.log" -mtime +${KEEP_LOGS} -delete
# -----------------------------------------------------------------------------
# SYSTEM CONTEXT
# -----------------------------------------------------------------------------
OMV_VER=$(dpkg -l openmediavault 2>/dev/null | awk '/^ii/{print $3}' || echo "Unknown")
KERNEL_VER=$(uname -r)
HOST_IP=$(hostname -I | awk '{print $1}')
UPTIME_STR=$(uptime -p)
LOAD_STR=$(uptime | awk -F'load average:' '{print $2}' | xargs)
MEM_USED=$(free -h | awk '/^Mem:/{print $3}')
MEM_TOTAL=$(free -h | awk '/^Mem:/{print $2}')
NEXT_RUN=$(systemctl show fstrim.timer --property=NextElapseUSecRealtime 2>/dev/null \
| awk -F= '{print $2}' || echo "See: systemctl status fstrim.timer")
log_section "Report complete — Overall: $OVERALL_STATUS"
# -----------------------------------------------------------------------------
# BUILD HTML EMAIL (style: KingDezigns monitor)
# -----------------------------------------------------------------------------
# Overall banner & badge
case "$OVERALL_STATUS" in
SUCCESS)
BADGE_HTML="✅ SUCCESS
"
BANNER_HTML="| ✅ TRIM completed successfully — no action required. |
"
;;
WARNING)
BADGE_HTML="⚠️ WARNING
"
BANNER_HTML="| ⚠️ TRIM completed with warnings — review recommended. |
"
;;
FAILED)
BADGE_HTML="🚨 FAILED
"
BANNER_HTML="| 🚨 TRIM run failed — immediate attention required! |
"
;;
esac
HTML_BODY=$(cat <
|
SSD Maintenance Monitor
💽 $HOSTNAME_LABEL
$REPORT_TIME | Weekly TRIM · $WARNING_COUNT warning(s) · $ERROR_COUNT error(s)
|
$BADGE_HTML |
|
$BANNER_HTML
📋 Run Summary
| Section |
Status |
Detail |
$SECTION_ROWS_HTML
|
✂️ TRIM Results — $TOTAL_HUMAN Total Freed
| Mount point |
Space freed |
$TRIM_ROWS_HTML
|
|
|
💾 Filesystem Usage
| Mount point |
Usage |
Used % |
$FS_ROWS_HTML
|
🖥️ System Info
| OMV |
${OMV_VER} |
Kernel |
${KERNEL_VER} |
|
| IP Address |
${HOST_IP} |
Uptime |
${UPTIME_STR} |
|
| Load average |
${LOAD_STR} |
Memory |
${MEM_USED} / ${MEM_TOTAL} |
|
| Log file |
${LOG_FILE} |
|
| Next TRIM run |
${NEXT_RUN} |
|
|
SSD Maintenance Monitor | $HOSTNAME_LABEL | KingDezigns Infrastructure
|
|
EOF
)
# -----------------------------------------------------------------------------
# SEND EMAIL
# -----------------------------------------------------------------------------
SUBJECT="[TRIM] $HOSTNAME_LABEL — $OVERALL_STATUS | $REPORT_TIME"
{
echo "To: $REPORT_TO"
echo "From: $REPORT_FROM"
echo "Subject: $SUBJECT"
echo "MIME-Version: 1.0"
echo "Content-Type: text/html; charset=UTF-8"
echo ""
echo "$HTML_BODY"
} | sendmail -t
log "Email sent to $REPORT_TO — Overall status: $OVERALL_STATUS"
log "===== fstrim-report.sh finished ====="