first commit
This commit is contained in:
398
alma-v2.1.sh
Executable file
398
alma-v2.1.sh
Executable file
@@ -0,0 +1,398 @@
|
||||
#!/bin/bash
|
||||
|
||||
# Disk Health Check Script for Alma Linux 9
|
||||
# Checks SSD TBW/lifespan and HDD health status
|
||||
|
||||
SCRIPT_NAME=$(basename "$0")
|
||||
VERSION="2.1"
|
||||
|
||||
# Color codes
|
||||
RED=$(tput setaf 1)
|
||||
GREEN=$(tput setaf 2)
|
||||
YELLOW=$(tput setaf 3)
|
||||
BLUE=$(tput setaf 4)
|
||||
CYAN=$(tput setaf 6)
|
||||
NC=$(tput sgr0)
|
||||
|
||||
# Function to print colored output
|
||||
print_color() {
|
||||
local color=$1
|
||||
local message=$2
|
||||
echo -e "${color}${message}${NC}"
|
||||
}
|
||||
|
||||
# Check if command exists
|
||||
command_exists() {
|
||||
command -v "$1" >/dev/null 2>&1
|
||||
}
|
||||
|
||||
# Check dependencies
|
||||
check_dependencies() {
|
||||
local missing=()
|
||||
|
||||
if ! command_exists smartctl; then
|
||||
missing+=("smartmontools")
|
||||
fi
|
||||
|
||||
if ! command_exists bc; then
|
||||
missing+=("bc")
|
||||
fi
|
||||
|
||||
if [[ ${#missing[@]} -gt 0 ]]; then
|
||||
print_color $RED "Error: Missing required packages: ${missing[*]}"
|
||||
echo "Install with: sudo dnf install ${missing[*]}"
|
||||
exit 1
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to get disk type
|
||||
get_disk_type() {
|
||||
local disk=$1
|
||||
local info=$(smartctl -i "$disk" 2>/dev/null)
|
||||
|
||||
if echo "$info" | grep -q "Solid State Device"; then
|
||||
echo "SSD"
|
||||
elif echo "$info" | grep -q "Rotation Rate"; then
|
||||
echo "HDD"
|
||||
else
|
||||
echo "UNKNOWN"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to calculate TBW for SSD
|
||||
calculate_tbw() {
|
||||
local raw_value=$1
|
||||
local sectors=$2
|
||||
|
||||
if [[ -n "$sectors" && "$sectors" != "0" ]]; then
|
||||
local bytes=$((sectors * 512))
|
||||
local tbw=$(echo "scale=2; $bytes / 1000 / 1000 / 1000 / 1000" | bc 2>/dev/null || echo "0")
|
||||
echo "$tbw"
|
||||
elif [[ -n "$raw_value" && "$raw_value" != "0" ]]; then
|
||||
local tbw=$(echo "scale=2; $raw_value * 32 / 1000 / 1000" | bc 2>/dev/null || echo "0")
|
||||
echo "$tbw"
|
||||
else
|
||||
echo "0"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to estimate SSD endurance based on model and capacity
|
||||
estimate_ssd_endurance() {
|
||||
local disk_model=$1
|
||||
local capacity_gb=$2
|
||||
|
||||
# Enterprise SSDs typically have higher endurance
|
||||
if echo "$disk_model" | grep -qi "MTFDDAK480TDS\|MICRON\|INTEL\|SAMSUNG\|KIOXIA"; then
|
||||
# Enterprise SSDs
|
||||
if [[ $capacity_gb -ge 1000 ]]; then
|
||||
echo "1200" # 1.2PB for 1TB enterprise
|
||||
elif [[ $capacity_gb -ge 480 ]]; then
|
||||
echo "600" # 600TB for 480GB enterprise
|
||||
elif [[ $capacity_gb -ge 240 ]]; then
|
||||
echo "300" # 300TB for 240GB enterprise
|
||||
else
|
||||
echo "150" # 150TB for smaller enterprise
|
||||
fi
|
||||
else
|
||||
# Consumer SSDs
|
||||
if [[ $capacity_gb -ge 1000 ]]; then
|
||||
echo "600" # 600TB for 1TB consumer
|
||||
elif [[ $capacity_gb -ge 480 ]]; then
|
||||
echo "300" # 300TB for 480GB consumer
|
||||
elif [[ $capacity_gb -ge 240 ]]; then
|
||||
echo "150" # 150TB for 240GB consumer
|
||||
elif [[ $capacity_gb -ge 120 ]]; then
|
||||
echo "80" # 80TB for 120GB consumer
|
||||
else
|
||||
echo "40" # 40TB for smaller drives
|
||||
fi
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to estimate SSD lifespan with TBW remaining
|
||||
estimate_ssd_lifespan() {
|
||||
local power_on_hours=$1
|
||||
local tbw_used=$2
|
||||
local disk_model=$3
|
||||
local capacity_gb=$4
|
||||
|
||||
if [[ -z "$power_on_hours" || "$power_on_hours" -eq 0 ]]; then
|
||||
echo "Unknown||Unknown"
|
||||
return
|
||||
fi
|
||||
|
||||
local estimated_endurance=$(estimate_ssd_endurance "$disk_model" "$capacity_gb")
|
||||
local tbw_remaining=$(echo "scale=2; $estimated_endurance - $tbw_used" | bc 2>/dev/null || echo "0")
|
||||
|
||||
if [[ $(echo "$tbw_used > 0" | bc 2>/dev/null) -eq 1 ]]; then
|
||||
local lifespan_used=$(echo "scale=1; $tbw_used * 100 / $estimated_endurance" | bc 2>/dev/null || echo "0")
|
||||
local lifespan_remaining=$(echo "scale=1; 100 - $lifespan_used" | bc 2>/dev/null || echo "100")
|
||||
|
||||
if [[ $(echo "$lifespan_used >= 80" | bc 2>/dev/null) -eq 1 ]]; then
|
||||
echo "${RED}${lifespan_remaining}%${NC}|${RED}${tbw_remaining} TB${NC}|High wear"
|
||||
elif [[ $(echo "$lifespan_used >= 50" | bc 2>/dev/null) -eq 1 ]]; then
|
||||
echo "${YELLOW}${lifespan_remaining}%${NC}|${YELLOW}${tbw_remaining} TB${NC}|Moderate wear"
|
||||
else
|
||||
echo "${GREEN}${lifespan_remaining}%${NC}|${GREEN}${tbw_remaining} TB${NC}|Healthy"
|
||||
fi
|
||||
else
|
||||
echo "Unknown|${estimated_endurance} TB|New"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to estimate HDD lifespan
|
||||
estimate_hdd_lifespan() {
|
||||
local power_on_hours=$1
|
||||
local reallocated_sectors=$2
|
||||
local pending_sectors=$3
|
||||
|
||||
if [[ -z "$power_on_hours" ]]; then
|
||||
echo "Unknown"
|
||||
return
|
||||
fi
|
||||
|
||||
# Convert to integers
|
||||
power_on_hours=${power_on_hours:-0}
|
||||
reallocated_sectors=${reallocated_sectors:-0}
|
||||
pending_sectors=${pending_sectors:-0}
|
||||
|
||||
if [[ "$pending_sectors" -gt 0 ]]; then
|
||||
echo "${RED}CRITICAL${NC} (Pending sectors: $pending_sectors)"
|
||||
elif [[ "$reallocated_sectors" -gt 100 ]]; then
|
||||
echo "${RED}< 6 months${NC} (High reallocated sectors: $reallocated_sectors)"
|
||||
elif [[ "$reallocated_sectors" -gt 10 ]]; then
|
||||
echo "${YELLOW}6-12 months${NC} (Reallocated sectors: $reallocated_sectors)"
|
||||
elif [[ "$power_on_hours" -gt 40000 ]]; then
|
||||
echo "${YELLOW}1-2 years${NC} (High usage: $power_on_hours hours)"
|
||||
elif [[ "$power_on_hours" -gt 25000 ]]; then
|
||||
echo "${GREEN}2-3 years${NC} (Moderate usage: $power_on_hours hours)"
|
||||
else
|
||||
echo "${GREEN}> 3 years${NC} (Low usage: $power_on_hours hours)"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to check disk SMART capabilities
|
||||
check_smart_capability() {
|
||||
local disk=$1
|
||||
local info=$(smartctl -i "$disk" 2>/dev/null)
|
||||
|
||||
if echo "$info" | grep -q "SMART support is: Available"; then
|
||||
if echo "$info" | grep -q "SMART support is: Enabled"; then
|
||||
echo "enabled"
|
||||
else
|
||||
echo "available"
|
||||
fi
|
||||
else
|
||||
echo "unavailable"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to check a single disk
|
||||
check_disk() {
|
||||
local disk=$1
|
||||
|
||||
print_color $CYAN "Checking disk: $disk"
|
||||
echo "=================================================="
|
||||
|
||||
# Check if disk exists and is accessible
|
||||
if [[ ! -b "$disk" ]]; then
|
||||
print_color $RED "Error: $disk is not a valid block device"
|
||||
echo ""
|
||||
return
|
||||
fi
|
||||
|
||||
# Check SMART capability
|
||||
local smart_status=$(check_smart_capability "$disk")
|
||||
if [[ "$smart_status" == "unavailable" ]]; then
|
||||
print_color $YELLOW "SMART not supported on $disk"
|
||||
echo ""
|
||||
return
|
||||
elif [[ "$smart_status" == "available" ]]; then
|
||||
print_color $YELLOW "SMART available but not enabled on $disk"
|
||||
echo "Enable with: smartctl -s on $disk"
|
||||
echo ""
|
||||
return
|
||||
fi
|
||||
|
||||
# Get basic disk information
|
||||
local info=$(smartctl -i "$disk" 2>/dev/null)
|
||||
local health=$(smartctl -H "$disk" 2>/dev/null)
|
||||
local attributes=$(smartctl -A "$disk" 2>/dev/null)
|
||||
|
||||
# Check if smartctl command succeeded
|
||||
if [[ $? -ne 0 ]]; then
|
||||
print_color $RED "Error: Cannot read SMART data from $disk"
|
||||
echo "You may need to run this script as root"
|
||||
echo ""
|
||||
return
|
||||
fi
|
||||
|
||||
# Extract disk information
|
||||
local model=$(echo "$info" | grep "Device Model:" | cut -d: -f2 | sed 's/^[ \t]*//')
|
||||
local serial=$(echo "$info" | grep "Serial Number:" | cut -d: -f2 | sed 's/^[ \t]*//')
|
||||
local capacity=$(echo "$info" | grep "User Capacity:" | cut -d: -f2 | sed 's/^[ \t]*//' | cut -d'[' -f1)
|
||||
local firmware=$(echo "$info" | grep "Firmware Version:" | cut -d: -f2 | sed 's/^[ \t]*//')
|
||||
|
||||
# Extract capacity in GB for endurance calculation
|
||||
local capacity_gb=0
|
||||
if echo "$capacity" | grep -qi "TB"; then
|
||||
capacity_gb=$(echo "$capacity" | grep -o '[0-9.]*' | head -1 | awk '{print $1 * 1000}' | bc 2>/dev/null | cut -d. -f1)
|
||||
else
|
||||
capacity_gb=$(echo "$capacity" | grep -o '[0-9.]*' | head -1 | cut -d. -f1)
|
||||
fi
|
||||
|
||||
local disk_type=$(get_disk_type "$disk")
|
||||
local health_status=$(echo "$health" | grep "result:" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
|
||||
# Extract SMART attributes
|
||||
local power_on_hours=$(echo "$attributes" | grep "Power_On_Hours" | awk '{print $10}' | head -1)
|
||||
local reallocated_sectors=$(echo "$attributes" | grep "Reallocated_Sector_Ct" | awk '{print $10}' | head -1)
|
||||
local pending_sectors=$(echo "$attributes" | grep "Current_Pending_Sector" | awk '{print $10}' | head -1)
|
||||
local total_written=$(echo "$attributes" | grep -E "Total_LBAs_Written|Host_Writes_32MiB" | awk '{print $10}' | head -1)
|
||||
local host_writes_32mib=$(echo "$attributes" | grep "Host_Writes_32MiB" | awk '{print $10}' | head -1)
|
||||
|
||||
# Display basic information
|
||||
echo "Model: ${model:-Unknown}"
|
||||
echo "Serial: ${serial:-Unknown}"
|
||||
echo "Type: $disk_type"
|
||||
echo "Capacity: ${capacity:-Unknown}"
|
||||
echo "Firmware: ${firmware:-Unknown}"
|
||||
echo "Health: ${health_status:-Unknown}"
|
||||
echo "Power On Hours: ${power_on_hours:-Unknown}"
|
||||
|
||||
# Disk type specific analysis
|
||||
if [[ "$disk_type" == "SSD" ]]; then
|
||||
local tbw_used=0
|
||||
if [[ -n "$total_written" && "$total_written" != "0" ]]; then
|
||||
tbw_used=$(calculate_tbw "" "$total_written")
|
||||
elif [[ -n "$host_writes_32mib" && "$host_writes_32mib" != "0" ]]; then
|
||||
tbw_used=$(calculate_tbw "$host_writes_32mib" "")
|
||||
fi
|
||||
|
||||
echo "TBW Used: ${tbw_used} TB"
|
||||
|
||||
local lifespan_info=$(estimate_ssd_lifespan "$power_on_hours" "$tbw_used" "$model" "$capacity_gb")
|
||||
local lifespan_percent=$(echo "$lifespan_info" | cut -d'|' -f1)
|
||||
local tbw_remaining=$(echo "$lifespan_info" | cut -d'|' -f2)
|
||||
local wear_status=$(echo "$lifespan_info" | cut -d'|' -f3)
|
||||
|
||||
echo "TBW Remaining: $tbw_remaining"
|
||||
echo "Lifespan: $lifespan_percent ($wear_status)"
|
||||
|
||||
elif [[ "$disk_type" == "HDD" ]]; then
|
||||
echo "Realloc Sectors: ${reallocated_sectors:-0}"
|
||||
echo "Pending Sectors: ${pending_sectors:-0}"
|
||||
|
||||
local lifespan=$(estimate_hdd_lifespan "$power_on_hours" "${reallocated_sectors:-0}" "${pending_sectors:-0}")
|
||||
echo "Lifespan: $lifespan"
|
||||
else
|
||||
print_color $YELLOW "Limited information available for this disk type"
|
||||
fi
|
||||
|
||||
echo ""
|
||||
}
|
||||
|
||||
# Function to detect all disks
|
||||
detect_disks() {
|
||||
local disks=()
|
||||
|
||||
# Check for SATA/SAS disks
|
||||
for disk in /dev/sd[a-z] /dev/sd[a-z][a-z]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
fi
|
||||
done
|
||||
|
||||
# Check for NVMe disks
|
||||
for disk in /dev/nvme[0-9]n[0-9] /dev/nvme[0-9]n[0-9]p[0-9]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
fi
|
||||
done
|
||||
|
||||
# Check for other disk types
|
||||
for disk in /dev/vd[a-z] /dev/xvd[a-z]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
fi
|
||||
done
|
||||
|
||||
echo "${disks[@]}"
|
||||
}
|
||||
|
||||
# Main function
|
||||
main() {
|
||||
print_color $BLUE "Disk Health Check Script v$VERSION for Alma Linux 9"
|
||||
print_color $BLUE "===================================================="
|
||||
echo ""
|
||||
|
||||
check_dependencies
|
||||
|
||||
local disks=()
|
||||
|
||||
# If specific disk provided, check only that disk
|
||||
if [[ $# -gt 0 ]]; then
|
||||
for disk in "$@"; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
else
|
||||
print_color $RED "Error: $disk is not a valid block device"
|
||||
fi
|
||||
done
|
||||
else
|
||||
# Auto-detect disks
|
||||
print_color $CYAN "Auto-detecting disks..."
|
||||
read -ra disks <<< "$(detect_disks)"
|
||||
fi
|
||||
|
||||
if [[ ${#disks[@]} -eq 0 ]]; then
|
||||
print_color $RED "No disks found or accessible"
|
||||
echo "Try running as root or specifying disk paths manually"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
print_color $GREEN "Found ${#disks[@]} disk(s) to check"
|
||||
echo ""
|
||||
|
||||
# Check if running as root, warn if not
|
||||
if [[ $EUID -ne 0 ]]; then
|
||||
print_color $YELLOW "Warning: Not running as root. Some disks may not be accessible."
|
||||
echo "For complete results, run as: sudo $0"
|
||||
echo ""
|
||||
fi
|
||||
|
||||
# Check each disk
|
||||
for disk in "${disks[@]}"; do
|
||||
check_disk "$disk"
|
||||
done
|
||||
|
||||
print_color $BLUE "Check completed!"
|
||||
}
|
||||
|
||||
# Usage information
|
||||
usage() {
|
||||
echo "Usage: $SCRIPT_NAME [DISK1 DISK2 ...]"
|
||||
echo ""
|
||||
echo "If no disks specified, auto-detects all available disks"
|
||||
echo ""
|
||||
echo "Examples:"
|
||||
echo " $SCRIPT_NAME # Check all auto-detected disks"
|
||||
echo " sudo $SCRIPT_NAME # Check all disks (as root)"
|
||||
echo " $SCRIPT_NAME /dev/sda # Check specific disk"
|
||||
echo " $SCRIPT_NAME /dev/sda /dev/nvme0n1 # Check multiple disks"
|
||||
}
|
||||
|
||||
# Parse command line arguments
|
||||
case "${1:-}" in
|
||||
-h|--help)
|
||||
usage
|
||||
exit 0
|
||||
;;
|
||||
-v|--version)
|
||||
echo "$SCRIPT_NAME version $VERSION"
|
||||
exit 0
|
||||
;;
|
||||
*)
|
||||
main "$@"
|
||||
;;
|
||||
esac
|
||||
536
alma-v2.4.sh
Executable file
536
alma-v2.4.sh
Executable file
@@ -0,0 +1,536 @@
|
||||
#!/bin/bash
|
||||
|
||||
# Disk Health Check Script for Alma Linux 9
|
||||
# Enhanced with SAS/PERC H730P controller support
|
||||
# Checks SSD TBW/lifespan and HDD health status
|
||||
|
||||
SCRIPT_NAME=$(basename "$0")
|
||||
VERSION="2.4"
|
||||
|
||||
# Color codes
|
||||
RED=$(tput setaf 1)
|
||||
GREEN=$(tput setaf 2)
|
||||
YELLOW=$(tput setaf 3)
|
||||
BLUE=$(tput setaf 4)
|
||||
CYAN=$(tput setaf 6)
|
||||
NC=$(tput sgr0)
|
||||
|
||||
# Function to print colored output
|
||||
print_color() {
|
||||
local color=$1
|
||||
local message=$2
|
||||
echo -e "${color}${message}${NC}"
|
||||
}
|
||||
|
||||
# Check if command exists
|
||||
command_exists() {
|
||||
command -v "$1" >/dev/null 2>&1
|
||||
}
|
||||
|
||||
# Check dependencies
|
||||
check_dependencies() {
|
||||
local missing=()
|
||||
|
||||
if ! command_exists smartctl; then
|
||||
missing+=("smartmontools")
|
||||
fi
|
||||
|
||||
if ! command_exists bc; then
|
||||
missing+=("bc")
|
||||
fi
|
||||
|
||||
if [[ ${#missing[@]} -gt 0 ]]; then
|
||||
print_color $RED "Error: Missing required packages: ${missing[*]}"
|
||||
echo "Install with: sudo dnf install ${missing[*]}"
|
||||
exit 1
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to test SMART access and get available data
|
||||
test_smart_access() {
|
||||
local disk=$1
|
||||
local controller=$2
|
||||
|
||||
local smart_cmd="smartctl"
|
||||
[[ -n "$controller" ]] && smart_cmd+=" -d $controller"
|
||||
|
||||
# Test basic SMART access
|
||||
if ! $smart_cmd -i "$disk" &>/dev/null; then
|
||||
echo "no_access"
|
||||
return
|
||||
fi
|
||||
|
||||
# Check if SMART is enabled (don't enable it, just check status)
|
||||
local smart_info=$($smart_cmd -i "$disk" 2>/dev/null)
|
||||
local smart_available=$(echo "$smart_info" | grep "SMART support is:" | awk '{print $4}')
|
||||
local smart_enabled=$(echo "$smart_info" | grep "SMART support is:" | awk '{print $6}')
|
||||
|
||||
if [[ "$smart_available" != "Available" ]]; then
|
||||
echo "not_available"
|
||||
return
|
||||
fi
|
||||
|
||||
if [[ "$smart_enabled" != "Enabled" ]]; then
|
||||
echo "disabled"
|
||||
return
|
||||
fi
|
||||
|
||||
# Test attribute reading
|
||||
local attributes=$($smart_cmd -A "$disk" 2>/dev/null)
|
||||
if [[ -z "$attributes" ]]; then
|
||||
echo "no_attributes"
|
||||
return
|
||||
fi
|
||||
|
||||
# Check if we have basic attributes
|
||||
local power_on_hours=$(echo "$attributes" | grep -i "Power_On_Hours\|Power-On" | awk '{print $10}' | head -1)
|
||||
if [[ -z "$power_on_hours" ]]; then
|
||||
echo "limited_attributes"
|
||||
return
|
||||
fi
|
||||
|
||||
echo "full_access"
|
||||
}
|
||||
|
||||
# Function to get disk information with enhanced SAS support
|
||||
get_disk_info() {
|
||||
local disk=$1
|
||||
local controller=$2
|
||||
|
||||
local smart_cmd="smartctl"
|
||||
[[ -n "$controller" ]] && smart_cmd+=" -d $controller"
|
||||
|
||||
local info=$($smart_cmd -i "$disk" 2>/dev/null)
|
||||
local attributes=$($smart_cmd -A "$disk" 2>/dev/null)
|
||||
local health=$($smart_cmd -H "$disk" 2>/dev/null)
|
||||
|
||||
# Extract information with multiple fallbacks for SAS drives
|
||||
local model=$(echo "$info" | grep -i "Device Model:\|Product:" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
local vendor=$(echo "$info" | grep -i "Vendor:" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
[[ -n "$vendor" && -n "$model" ]] && model="$vendor $model"
|
||||
|
||||
local serial=$(echo "$info" | grep -i "Serial Number:\|Serial number:" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
|
||||
local capacity=$(echo "$info" | grep -i "User Capacity:\|Total NVM Capacity:" | cut -d: -f2 | sed 's/^[ \t]*//' | cut -d'[' -f1 | head -1)
|
||||
|
||||
local firmware=$(echo "$info" | grep -i "Firmware Version:\|Firmware revision:\|Revision:" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
|
||||
local health_status=$(echo "$health" | grep -i "result:\|SMART overall-health" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
[[ -z "$health_status" ]] && health_status=$(echo "$health" | grep -i "SMART overall-health" | awk -F'[' '{print $2}' | cut -d']' -f1)
|
||||
|
||||
# Get disk type with SAS support
|
||||
local disk_type="UNKNOWN"
|
||||
if echo "$info" | grep -qi "Solid State Device"; then
|
||||
disk_type="SSD"
|
||||
elif echo "$info" | grep -qi "Rotation Rate"; then
|
||||
disk_type="HDD"
|
||||
elif echo "$info" | grep -qi "SCSI\|SAS"; then
|
||||
# SAS drives often don't specify, check rotation rate
|
||||
if echo "$info" | grep -qi "15000\|10000\|7200"; then
|
||||
disk_type="HDD"
|
||||
else
|
||||
disk_type="SSD"
|
||||
fi
|
||||
fi
|
||||
|
||||
# Extract SMART attributes with multiple field attempts for SAS
|
||||
local power_on_hours=$(echo "$attributes" | grep -i "Power_On_Hours\|Power-On" | awk '{print $NF}' | head -1)
|
||||
[[ -z "$power_on_hours" ]] && power_on_hours=$(echo "$attributes" | grep -i "Power_On_Hours_and_Msec" | awk '{print $10}' | head -1)
|
||||
|
||||
local reallocated_sectors=$(echo "$attributes" | grep -i "Reallocated_Sector_Ct\|Reallocated_Event_Count" | awk '{print $NF}' | head -1)
|
||||
|
||||
local pending_sectors=$(echo "$attributes" | grep -i "Current_Pending_Sector" | awk '{print $NF}' | head -1)
|
||||
|
||||
local total_written=$(echo "$attributes" | grep -i "Total_LBAs_Written\|Host_Writes_32MiB\|Lifetime_Writes" | awk '{print $NF}' | head -1)
|
||||
local host_writes_32mib=$(echo "$attributes" | grep -i "Host_Writes_32MiB" | awk '{print $NF}' | head -1)
|
||||
|
||||
# For SAS drives, try to get media wearout for SSDs
|
||||
local media_wearout=$(echo "$attributes" | grep -i "Media_Wearout_Indicator\|Wear_Leveling_Count" | awk '{print $NF}' | head -1)
|
||||
|
||||
echo "$model|$serial|$capacity|$firmware|$health_status|$disk_type|$power_on_hours|$reallocated_sectors|$pending_sectors|$total_written|$host_writes_32mib|$media_wearout"
|
||||
}
|
||||
|
||||
# Function to calculate TBW for SSD
|
||||
calculate_tbw() {
|
||||
local raw_value=$1
|
||||
local sectors=$2
|
||||
|
||||
if [[ -n "$sectors" && "$sectors" != "0" ]]; then
|
||||
local bytes=$((sectors * 512))
|
||||
local tbw=$(echo "scale=2; $bytes / 1000 / 1000 / 1000 / 1000" | bc 2>/dev/null || echo "0")
|
||||
echo "$tbw"
|
||||
elif [[ -n "$raw_value" && "$raw_value" != "0" ]]; then
|
||||
local tbw=$(echo "scale=2; $raw_value * 32 / 1000 / 1000" | bc 2>/dev/null || echo "0")
|
||||
echo "$tbw"
|
||||
else
|
||||
echo "0"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to estimate SSD endurance based on model and capacity
|
||||
estimate_ssd_endurance() {
|
||||
local disk_model=$1
|
||||
local capacity_gb=$2
|
||||
|
||||
# SAS SSDs typically have very high endurance
|
||||
if echo "$disk_model" | grep -qi "ST600MP\|SEAGATE.*SSD\|SAS.*SSD"; then
|
||||
# Enterprise SAS SSDs - very high endurance
|
||||
if [[ $capacity_gb -ge 1000 ]]; then
|
||||
echo "10000" # 10PB for 1TB+ enterprise SAS SSD
|
||||
elif [[ $capacity_gb -ge 600 ]]; then
|
||||
echo "6000" # 6PB for 600GB enterprise SAS SSD
|
||||
elif [[ $capacity_gb -ge 400 ]]; then
|
||||
echo "4000" # 4PB for 400GB enterprise SAS SSD
|
||||
else
|
||||
echo "2000" # 2PB for smaller enterprise SAS SSD
|
||||
fi
|
||||
elif echo "$disk_model" | grep -qi "MTFDDAK\|MICRON\|INTEL\|SAMSUNG\|KIOXIA"; then
|
||||
# Enterprise SATA SSDs
|
||||
if [[ $capacity_gb -ge 1000 ]]; then
|
||||
echo "1200" # 1.2PB for 1TB enterprise
|
||||
elif [[ $capacity_gb -ge 480 ]]; then
|
||||
echo "600" # 600TB for 480GB enterprise
|
||||
elif [[ $capacity_gb -ge 240 ]]; then
|
||||
echo "300" # 300TB for 240GB enterprise
|
||||
else
|
||||
echo "150" # 150TB for smaller enterprise
|
||||
fi
|
||||
else
|
||||
# Consumer SSDs
|
||||
if [[ $capacity_gb -ge 1000 ]]; then
|
||||
echo "600" # 600TB for 1TB consumer
|
||||
elif [[ $capacity_gb -ge 480 ]]; then
|
||||
echo "300" # 300TB for 480GB consumer
|
||||
elif [[ $capacity_gb -ge 240 ]]; then
|
||||
echo "150" # 150TB for 240GB consumer
|
||||
elif [[ $capacity_gb -ge 120 ]]; then
|
||||
echo "80" # 80TB for 120GB consumer
|
||||
else
|
||||
echo "40" # 40TB for smaller drives
|
||||
fi
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to estimate SSD lifespan with TBW remaining
|
||||
estimate_ssd_lifespan() {
|
||||
local power_on_hours=$1
|
||||
local tbw_used=$2
|
||||
local disk_model=$3
|
||||
local capacity_gb=$4
|
||||
local media_wearout=$5
|
||||
|
||||
if [[ -z "$power_on_hours" || "$power_on_hours" -eq 0 ]]; then
|
||||
echo "Unknown||Unknown||Unknown"
|
||||
return
|
||||
fi
|
||||
|
||||
local estimated_endurance=$(estimate_ssd_endurance "$disk_model" "$capacity_gb")
|
||||
local tbw_remaining=$(echo "scale=2; $estimated_endurance - $tbw_used" | bc 2>/dev/null || echo "0")
|
||||
|
||||
# If we have media wearout indicator, use it for more accurate estimation
|
||||
if [[ -n "$media_wearout" && "$media_wearout" != "0" ]]; then
|
||||
if [[ $media_wearout -le 10 ]]; then
|
||||
echo "${RED}10%${NC}|${RED}${tbw_remaining} TB${NC}|${RED}Critical wear${NC}|media_wearout"
|
||||
elif [[ $media_wearout -le 30 ]]; then
|
||||
echo "${YELLOW}${media_wearout}%${NC}|${YELLOW}${tbw_remaining} TB${NC}|${YELLOW}High wear${NC}|media_wearout"
|
||||
elif [[ $media_wearout -le 70 ]]; then
|
||||
echo "${YELLOW}${media_wearout}%${NC}|${YELLOW}${tbw_remaining} TB${NC}|${YELLOW}Moderate wear${NC}|media_wearout"
|
||||
else
|
||||
echo "${GREEN}${media_wearout}%${NC}|${GREEN}${tbw_remaining} TB${NC}|${GREEN}Healthy${NC}|media_wearout"
|
||||
fi
|
||||
return
|
||||
fi
|
||||
|
||||
if [[ $(echo "$tbw_used > 0" | bc 2>/dev/null) -eq 1 ]]; then
|
||||
local lifespan_used=$(echo "scale=1; $tbw_used * 100 / $estimated_endurance" | bc 2>/dev/null || echo "0")
|
||||
local lifespan_remaining=$(echo "scale=1; 100 - $lifespan_used" | bc 2>/dev/null || echo "100")
|
||||
|
||||
if [[ $(echo "$lifespan_used >= 80" | bc 2>/dev/null) -eq 1 ]]; then
|
||||
echo "${RED}${lifespan_remaining}%${NC}|${RED}${tbw_remaining} TB${NC}|${RED}High wear${NC}|tbw"
|
||||
elif [[ $(echo "$lifespan_used >= 50" | bc 2>/dev/null) -eq 1 ]]; then
|
||||
echo "${YELLOW}${lifespan_remaining}%${NC}|${YELLOW}${tbw_remaining} TB${NC}|${YELLOW}Moderate wear${NC}|tbw"
|
||||
else
|
||||
echo "${GREEN}${lifespan_remaining}%${NC}|${GREEN}${tbw_remaining} TB${NC}|${GREEN}Healthy${NC}|tbw"
|
||||
fi
|
||||
else
|
||||
echo "Unknown|${estimated_endurance} TB|New|estimated"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to estimate HDD lifespan
|
||||
estimate_hdd_lifespan() {
|
||||
local power_on_hours=$1
|
||||
local reallocated_sectors=$2
|
||||
local pending_sectors=$3
|
||||
|
||||
if [[ -z "$power_on_hours" ]]; then
|
||||
echo "Unknown"
|
||||
return
|
||||
fi
|
||||
|
||||
power_on_hours=${power_on_hours:-0}
|
||||
reallocated_sectors=${reallocated_sectors:-0}
|
||||
pending_sectors=${pending_sectors:-0}
|
||||
|
||||
if [[ "$pending_sectors" -gt 0 ]]; then
|
||||
echo "${RED}CRITICAL${NC} (Pending sectors: $pending_sectors)"
|
||||
elif [[ "$reallocated_sectors" -gt 100 ]]; then
|
||||
echo "${RED}< 6 months${NC} (High reallocated sectors: $reallocated_sectors)"
|
||||
elif [[ "$reallocated_sectors" -gt 10 ]]; then
|
||||
echo "${YELLOW}6-12 months${NC} (Reallocated sectors: $reallocated_sectors)"
|
||||
elif [[ "$power_on_hours" -gt 40000 ]]; then
|
||||
echo "${YELLOW}1-2 years${NC} (High usage: $power_on_hours hours)"
|
||||
elif [[ "$power_on_hours" -gt 25000 ]]; then
|
||||
echo "${GREEN}2-3 years${NC} (Moderate usage: $power_on_hours hours)"
|
||||
else
|
||||
echo "${GREEN}> 3 years${NC} (Low usage: $power_on_hours hours)"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to check a single disk with enhanced error handling
|
||||
check_disk() {
|
||||
local disk=$1
|
||||
local controller=$2
|
||||
|
||||
print_color $CYAN "Checking disk: $disk (Controller: ${controller:-direct})"
|
||||
echo "=================================================="
|
||||
|
||||
# Test SMART access level
|
||||
local access_level=$(test_smart_access "$disk" "$controller")
|
||||
|
||||
case $access_level in
|
||||
"no_access")
|
||||
print_color $RED "ERROR: Cannot access disk through controller"
|
||||
echo "Possible reasons:"
|
||||
echo " - Controller doesn't support SMART passthrough"
|
||||
echo " - Disk is part of a hardware RAID array"
|
||||
echo " - Insufficient permissions (try running as root)"
|
||||
echo " - Controller busy or offline"
|
||||
echo ""
|
||||
return
|
||||
;;
|
||||
"not_available")
|
||||
print_color $YELLOW "SMART not available on this disk"
|
||||
echo "This disk does not support SMART monitoring"
|
||||
echo ""
|
||||
return
|
||||
;;
|
||||
"disabled")
|
||||
print_color $YELLOW "SMART is disabled on this disk"
|
||||
echo "SMART is available but currently disabled"
|
||||
echo "To enable manually: smartctl -s on ${controller:+-d $controller} $disk"
|
||||
echo ""
|
||||
return
|
||||
;;
|
||||
"no_attributes")
|
||||
print_color $YELLOW "WARNING: Cannot read SMART attributes"
|
||||
echo "This is common with hardware RAID controllers like PERC H730P"
|
||||
echo "Try checking through the RAID management interface"
|
||||
echo ""
|
||||
return
|
||||
;;
|
||||
"limited_attributes")
|
||||
print_color $YELLOW "NOTE: Limited SMART data available"
|
||||
echo "Controller is filtering some SMART attributes"
|
||||
;;
|
||||
esac
|
||||
|
||||
# Get disk information
|
||||
local disk_info=$(get_disk_info "$disk" "$controller")
|
||||
IFS='|' read -r model serial capacity firmware health_status disk_type power_on_hours reallocated_sectors pending_sectors total_written host_writes_32mib media_wearout <<< "$disk_info"
|
||||
|
||||
# Display basic information
|
||||
echo "Model: ${model:-Unknown}"
|
||||
echo "Serial: ${serial:-Unknown}"
|
||||
echo "Type: $disk_type"
|
||||
echo "Capacity: ${capacity:-Unknown}"
|
||||
echo "Firmware: ${firmware:-Unknown}"
|
||||
echo "Health: ${health_status:-Unknown}"
|
||||
|
||||
# Only show power on hours if available
|
||||
if [[ -n "$power_on_hours" && "$power_on_hours" != "0" ]]; then
|
||||
echo "Power On Hours: $power_on_hours"
|
||||
else
|
||||
echo "Power On Hours: Unknown"
|
||||
fi
|
||||
|
||||
# Disk type specific analysis
|
||||
if [[ "$disk_type" == "SSD" ]]; then
|
||||
local tbw_used=0
|
||||
if [[ -n "$total_written" && "$total_written" != "0" ]]; then
|
||||
tbw_used=$(calculate_tbw "" "$total_written")
|
||||
elif [[ -n "$host_writes_32mib" && "$host_writes_32mib" != "0" ]]; then
|
||||
tbw_used=$(calculate_tbw "$host_writes_32mib" "")
|
||||
fi
|
||||
|
||||
if [[ $(echo "$tbw_used > 0" | bc 2>/dev/null) -eq 1 ]]; then
|
||||
echo "TBW Used: ${tbw_used} TB"
|
||||
fi
|
||||
|
||||
# Estimate capacity for endurance calculation
|
||||
local capacity_gb=0
|
||||
if echo "$capacity" | grep -qi "GB"; then
|
||||
capacity_gb=$(echo "$capacity" | grep -o '[0-9.]*' | head -1 | cut -d. -f1)
|
||||
elif echo "$capacity" | grep -qi "TB"; then
|
||||
capacity_gb=$(echo "$capacity" | grep -o '[0-9.]*' | head -1 | awk '{print $1 * 1000}' | bc 2>/dev/null | cut -d. -f1)
|
||||
fi
|
||||
|
||||
local lifespan_info=$(estimate_ssd_lifespan "$power_on_hours" "$tbw_used" "$model" "$capacity_gb" "$media_wearout")
|
||||
local lifespan_percent=$(echo "$lifespan_info" | cut -d'|' -f1)
|
||||
local tbw_remaining=$(echo "$lifespan_info" | cut -d'|' -f2)
|
||||
local wear_status=$(echo "$lifespan_info" | cut -d'|' -f3)
|
||||
local wear_source=$(echo "$lifespan_info" | cut -d'|' -f4)
|
||||
|
||||
if [[ "$wear_source" != "media_wearout" && $(echo "$tbw_used > 0" | bc 2>/dev/null) -eq 1 ]]; then
|
||||
echo "TBW Remaining: $tbw_remaining"
|
||||
fi
|
||||
|
||||
echo "Lifespan: $lifespan_percent ($wear_status)"
|
||||
|
||||
# Show wear source if available
|
||||
if [[ "$wear_source" == "media_wearout" ]]; then
|
||||
echo "Wear Source: Media Wearout Indicator"
|
||||
elif [[ "$wear_source" == "tbw" ]]; then
|
||||
echo "Wear Source: TBW Calculation"
|
||||
elif [[ "$wear_source" == "estimated" ]]; then
|
||||
echo "Wear Source: Estimated Endurance"
|
||||
fi
|
||||
|
||||
elif [[ "$disk_type" == "HDD" ]]; then
|
||||
if [[ -n "$reallocated_sectors" && "$reallocated_sectors" != "0" ]]; then
|
||||
echo "Realloc Sectors: $reallocated_sectors"
|
||||
fi
|
||||
if [[ -n "$pending_sectors" && "$pending_sectors" != "0" ]]; then
|
||||
echo "Pending Sectors: $pending_sectors"
|
||||
fi
|
||||
|
||||
local lifespan=$(estimate_hdd_lifespan "$power_on_hours" "${reallocated_sectors:-0}" "${pending_sectors:-0}")
|
||||
echo "Lifespan: $lifespan"
|
||||
else
|
||||
print_color $YELLOW "Limited information available for this disk type"
|
||||
echo "This is normal for hardware RAID configurations like PERC H730P"
|
||||
echo "For detailed SAS drive information, use controller management tools"
|
||||
fi
|
||||
|
||||
echo ""
|
||||
}
|
||||
|
||||
# Function to detect all disks with enhanced SAS support (no partitions)
|
||||
detect_disks() {
|
||||
local disks=()
|
||||
|
||||
# Check for SATA/SAS disks - only main devices, no partitions
|
||||
for disk in /dev/sd[a-z]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
fi
|
||||
done
|
||||
|
||||
# Check for NVMe disks - only main devices, no partitions
|
||||
for disk in /dev/nvme[0-9]n[0-9]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
fi
|
||||
done
|
||||
|
||||
# Check for SAS disks via SCSI generic - only main devices
|
||||
for disk in /dev/sg[0-9]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
fi
|
||||
done
|
||||
|
||||
# Check for other disk types - only main devices
|
||||
for disk in /dev/vd[a-z] /dev/xvd[a-z]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
fi
|
||||
done
|
||||
|
||||
echo "${disks[@]}"
|
||||
}
|
||||
|
||||
# Main function
|
||||
main() {
|
||||
print_color $BLUE "Alma Linux 9 Disk Health Check Script v$VERSION"
|
||||
print_color $BLUE "Enhanced with PERC H730P and SAS Support"
|
||||
print_color $BLUE "============================================"
|
||||
echo ""
|
||||
|
||||
check_dependencies
|
||||
|
||||
local disks=()
|
||||
|
||||
# If specific disk provided, check only that disk
|
||||
if [[ $# -gt 0 ]]; then
|
||||
for disk in "$@"; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
else
|
||||
print_color $RED "Error: $disk is not a valid block device"
|
||||
fi
|
||||
done
|
||||
else
|
||||
# Auto-detect disks
|
||||
print_color $CYAN "Auto-detecting disks (excluding partitions)..."
|
||||
read -ra disks <<< "$(detect_disks)"
|
||||
fi
|
||||
|
||||
if [[ ${#disks[@]} -eq 0 ]]; then
|
||||
print_color $RED "No disks found or accessible"
|
||||
echo "Try running as root or specifying disk paths manually"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
print_color $GREEN "Found ${#disks[@]} disk(s) to check"
|
||||
echo ""
|
||||
|
||||
# Check if running as root, warn if not
|
||||
if [[ $EUID -ne 0 ]]; then
|
||||
print_color $YELLOW "Warning: Not running as root."
|
||||
print_color $YELLOW "Some disks/controllers may show limited information."
|
||||
echo "For complete results, run as: sudo $0"
|
||||
echo ""
|
||||
fi
|
||||
|
||||
# Check each disk
|
||||
for disk in "${disks[@]}"; do
|
||||
check_disk "$disk"
|
||||
done
|
||||
|
||||
print_color $BLUE "Check completed!"
|
||||
echo ""
|
||||
print_color $CYAN "Note: For PERC H730P controllers with SAS drives:"
|
||||
print_color $CYAN " - Install 'storcli' for detailed controller information"
|
||||
print_color $CYAN " - Use 'smartctl -d sat /dev/sgX' to try direct access"
|
||||
print_color $CYAN " - Hardware RAID controllers often limit SMART data access"
|
||||
}
|
||||
|
||||
# Usage information
|
||||
usage() {
|
||||
echo "Usage: $SCRIPT_NAME [DISK1 DISK2 ...]"
|
||||
echo ""
|
||||
echo "If no disks specified, auto-detects all available disks"
|
||||
echo ""
|
||||
echo "Examples:"
|
||||
echo " $SCRIPT_NAME # Check all auto-detected disks"
|
||||
echo " sudo $SCRIPT_NAME # Check all disks (as root)"
|
||||
echo " $SCRIPT_NAME /dev/sda # Check specific disk"
|
||||
echo " $SCRIPT_NAME /dev/sg0 # Check SAS disk directly"
|
||||
echo " $SCRIPT_NAME /dev/sda /dev/nvme0n1 # Check multiple disks"
|
||||
}
|
||||
|
||||
# Parse command line arguments
|
||||
case "${1:-}" in
|
||||
-h|--help)
|
||||
usage
|
||||
exit 0
|
||||
;;
|
||||
-v|--version)
|
||||
echo "$SCRIPT_NAME version $VERSION"
|
||||
exit 0
|
||||
;;
|
||||
*)
|
||||
main "$@"
|
||||
;;
|
||||
esac
|
||||
337
harvester-v2.1.sh
Executable file
337
harvester-v2.1.sh
Executable file
@@ -0,0 +1,337 @@
|
||||
#!/bin/bash
|
||||
|
||||
# Disk Health Check Script for Harvester OS
|
||||
# Checks SSD TBW/lifespan and HDD health status
|
||||
# Supports RAID controllers and direct disks
|
||||
|
||||
SCRIPT_NAME=$(basename "$0")
|
||||
VERSION="2.1"
|
||||
|
||||
# Color codes
|
||||
RED=$(tput setaf 1)
|
||||
GREEN=$(tput setaf 2)
|
||||
YELLOW=$(tput setaf 3)
|
||||
BLUE=$(tput setaf 4)
|
||||
CYAN=$(tput setaf 6)
|
||||
NC=$(tput sgr0)
|
||||
|
||||
# Function to print colored output
|
||||
print_color() {
|
||||
local color=$1
|
||||
local message=$2
|
||||
echo -e "${color}${message}${NC}"
|
||||
}
|
||||
|
||||
# Check if smartctl is installed
|
||||
command_exists() {
|
||||
command -v "$1" >/dev/null 2>&1
|
||||
}
|
||||
|
||||
if ! command_exists smartctl; then
|
||||
print_color $RED "Error: smartctl is not installed. Please install smartmontools package."
|
||||
exit 1
|
||||
fi
|
||||
|
||||
# Function to get disk type
|
||||
get_disk_type() {
|
||||
local disk=$1
|
||||
local controller=$2
|
||||
|
||||
local smart_cmd="smartctl"
|
||||
[[ -n "$controller" ]] && smart_cmd+=" -d $controller"
|
||||
smart_cmd+=" -i $disk"
|
||||
|
||||
local info=$($smart_cmd 2>/dev/null)
|
||||
|
||||
if echo "$info" | grep -q "Solid State Device"; then
|
||||
echo "SSD"
|
||||
elif echo "$info" | grep -q "Rotation Rate"; then
|
||||
echo "HDD"
|
||||
else
|
||||
echo "UNKNOWN"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to calculate TBW for SSD
|
||||
calculate_tbw() {
|
||||
local raw_value=$1
|
||||
local sectors=$2
|
||||
|
||||
if [[ -n "$sectors" && "$sectors" != "0" ]]; then
|
||||
# Calculate from sectors (most common)
|
||||
local bytes=$((sectors * 512))
|
||||
local tbw=$(echo "scale=2; $bytes / 1000 / 1000 / 1000 / 1000" | bc -l 2>/dev/null || echo "0")
|
||||
echo "$tbw"
|
||||
elif [[ -n "$raw_value" && "$raw_value" != "0" ]]; then
|
||||
# Try to calculate from raw value (varies by manufacturer)
|
||||
local tbw=$(echo "scale=2; $raw_value * 32 / 1000 / 1000" | bc -l 2>/dev/null || echo "0")
|
||||
echo "$tbw"
|
||||
else
|
||||
echo "0"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to estimate SSD endurance based on model and capacity
|
||||
estimate_ssd_endurance() {
|
||||
local disk_model=$1
|
||||
local capacity_gb=$2
|
||||
|
||||
# Enterprise SSDs typically have higher endurance
|
||||
if echo "$disk_model" | grep -qi "MTFDDAK480TDS\|MICRON\|INTEL\|SAMSUNG"; then
|
||||
# Enterprise SSDs
|
||||
if [[ $capacity_gb -ge 1000 ]]; then
|
||||
echo "1200" # 1.2PB for 1TB enterprise
|
||||
elif [[ $capacity_gb -ge 480 ]]; then
|
||||
echo "600" # 600TB for 480GB enterprise
|
||||
elif [[ $capacity_gb -ge 240 ]]; then
|
||||
echo "300" # 300TB for 240GB enterprise
|
||||
else
|
||||
echo "150" # 150TB for smaller enterprise
|
||||
fi
|
||||
else
|
||||
# Consumer SSDs
|
||||
if [[ $capacity_gb -ge 1000 ]]; then
|
||||
echo "600" # 600TB for 1TB consumer
|
||||
elif [[ $capacity_gb -ge 480 ]]; then
|
||||
echo "300" # 300TB for 480GB consumer
|
||||
elif [[ $capacity_gb -ge 240 ]]; then
|
||||
echo "150" # 150TB for 240GB consumer
|
||||
elif [[ $capacity_gb -ge 120 ]]; then
|
||||
echo "80" # 80TB for 120GB consumer
|
||||
else
|
||||
echo "40" # 40TB for smaller drives
|
||||
fi
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to estimate SSD lifespan with TBW remaining
|
||||
estimate_ssd_lifespan() {
|
||||
local power_on_hours=$1
|
||||
local tbw_used=$2
|
||||
local disk_model=$3
|
||||
local capacity_gb=$4
|
||||
|
||||
if [[ -z "$power_on_hours" || "$power_on_hours" -eq 0 ]]; then
|
||||
echo "Unknown||Unknown"
|
||||
return
|
||||
fi
|
||||
|
||||
local estimated_endurance=$(estimate_ssd_endurance "$disk_model" "$capacity_gb")
|
||||
local tbw_remaining=$(echo "scale=2; $estimated_endurance - $tbw_used" | bc -l 2>/dev/null || echo "0")
|
||||
|
||||
if [[ $(echo "$tbw_used > 0" | bc -l 2>/dev/null) -eq 1 ]]; then
|
||||
local lifespan_used=$(echo "scale=1; $tbw_used * 100 / $estimated_endurance" | bc -l 2>/dev/null || echo "0")
|
||||
local lifespan_remaining=$(echo "scale=1; 100 - $lifespan_used" | bc -l 2>/dev/null || echo "100")
|
||||
|
||||
if [[ $(echo "$lifespan_used >= 80" | bc -l) -eq 1 ]]; then
|
||||
echo "${RED}${lifespan_remaining}%${NC}|${RED}${tbw_remaining} TB${NC}|High wear"
|
||||
elif [[ $(echo "$lifespan_used >= 50" | bc -l) -eq 1 ]]; then
|
||||
echo "${YELLOW}${lifespan_remaining}%${NC}|${YELLOW}${tbw_remaining} TB${NC}|Moderate wear"
|
||||
else
|
||||
echo "${GREEN}${lifespan_remaining}%${NC}|${GREEN}${tbw_remaining} TB${NC}|Healthy"
|
||||
fi
|
||||
else
|
||||
echo "Unknown|${estimated_endurance} TB|New"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to estimate HDD lifespan
|
||||
estimate_hdd_lifespan() {
|
||||
local power_on_hours=$1
|
||||
local reallocated_sectors=$2
|
||||
local pending_sectors=$3
|
||||
|
||||
if [[ -z "$power_on_hours" ]]; then
|
||||
echo "Unknown"
|
||||
return
|
||||
fi
|
||||
|
||||
# HDD lifespan estimation based on common failure patterns
|
||||
if [[ "$pending_sectors" -gt 0 ]]; then
|
||||
echo "${RED}CRITICAL${NC} (Pending sectors: $pending_sectors)"
|
||||
elif [[ "$reallocated_sectors" -gt 100 ]]; then
|
||||
echo "${RED}< 6 months${NC} (High reallocated sectors: $reallocated_sectors)"
|
||||
elif [[ "$reallocated_sectors" -gt 10 ]]; then
|
||||
echo "${YELLOW}6-12 months${NC} (Reallocated sectors: $reallocated_sectors)"
|
||||
elif [[ "$power_on_hours" -gt 40000 ]]; then
|
||||
echo "${YELLOW}1-2 years${NC} (High usage: $power_on_hours hours)"
|
||||
elif [[ "$power_on_hours" -gt 25000 ]]; then
|
||||
echo "${GREEN}2-3 years${NC} (Moderate usage: $power_on_hours hours)"
|
||||
else
|
||||
echo "${GREEN}> 3 years${NC} (Low usage: $power_on_hours hours)"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to check a single disk
|
||||
check_disk() {
|
||||
local disk=$1
|
||||
local controller=$2
|
||||
|
||||
local smart_cmd="smartctl"
|
||||
[[ -n "$controller" ]] && smart_cmd+=" -d $controller"
|
||||
|
||||
print_color $CYAN "Checking disk: $disk (Controller: ${controller:-direct})"
|
||||
echo "=================================================="
|
||||
|
||||
# Get basic disk information
|
||||
local info=$($smart_cmd -i "$disk" 2>/dev/null)
|
||||
local health=$($smart_cmd -H "$disk" 2>/dev/null)
|
||||
local attributes=$($smart_cmd -A "$disk" 2>/dev/null)
|
||||
|
||||
# Extract disk information
|
||||
local model=$(echo "$info" | grep "Device Model:" | cut -d: -f2 | sed 's/^[ \t]*//')
|
||||
local serial=$(echo "$info" | grep "Serial Number:" | cut -d: -f2 | sed 's/^[ \t]*//')
|
||||
local capacity=$(echo "$info" | grep "User Capacity:" | cut -d: -f2 | sed 's/^[ \t]*//' | cut -d'[' -f1)
|
||||
local firmware=$(echo "$info" | grep "Firmware Version:" | cut -d: -f2 | sed 's/^[ \t]*//')
|
||||
|
||||
# Extract capacity in GB for endurance calculation
|
||||
local capacity_gb=0
|
||||
if [[ $capacity =~ \[([0-9.]+)\s+GB\] ]]; then
|
||||
capacity_gb=${BASH_REMATCH[1]}
|
||||
elif [[ $capacity =~ \[([0-9.]+)\s+TB\] ]]; then
|
||||
capacity_gb=$(echo "${BASH_REMATCH[1]} * 1000" | bc -l 2>/dev/null | cut -d. -f1)
|
||||
fi
|
||||
|
||||
local disk_type=$(get_disk_type "$disk" "$controller")
|
||||
local health_status=$(echo "$health" | grep "result:" | cut -d: -f2 | sed 's/^[ \t]*//')
|
||||
|
||||
# Extract SMART attributes
|
||||
local power_on_hours=$(echo "$attributes" | grep "Power_On_Hours" | awk '{print $10}')
|
||||
local reallocated_sectors=$(echo "$attributes" | grep "Reallocated_Sector_Ct" | awk '{print $10}')
|
||||
local pending_sectors=$(echo "$attributes" | grep "Current_Pending_Sector" | awk '{print $10}')
|
||||
local total_written=$(echo "$attributes" | grep -E "Total_LBAs_Written|Host_Writes_32MiB" | awk '{print $10}')
|
||||
|
||||
# For SSDs with Host_Writes_32MiB
|
||||
local host_writes_32mib=$(echo "$attributes" | grep "Host_Writes_32MiB" | awk '{print $10}')
|
||||
|
||||
# Display basic information
|
||||
echo "Model: $model"
|
||||
echo "Serial: $serial"
|
||||
echo "Type: $disk_type"
|
||||
echo "Capacity: $capacity"
|
||||
echo "Firmware: $firmware"
|
||||
echo "Health: $health_status"
|
||||
echo "Power On Hours: $power_on_hours"
|
||||
|
||||
# Disk type specific analysis
|
||||
if [[ "$disk_type" == "SSD" ]]; then
|
||||
local tbw_used=0
|
||||
if [[ -n "$total_written" && "$total_written" != "0" ]]; then
|
||||
tbw_used=$(calculate_tbw "" "$total_written")
|
||||
elif [[ -n "$host_writes_32mib" && "$host_writes_32mib" != "0" ]]; then
|
||||
tbw_used=$(calculate_tbw "$host_writes_32mib" "")
|
||||
fi
|
||||
|
||||
echo "TBW Used: ${tbw_used} TB"
|
||||
|
||||
local lifespan_info=$(estimate_ssd_lifespan "$power_on_hours" "$tbw_used" "$model" "$capacity_gb")
|
||||
local lifespan_percent=$(echo "$lifespan_info" | cut -d'|' -f1)
|
||||
local tbw_remaining=$(echo "$lifespan_info" | cut -d'|' -f2)
|
||||
local wear_status=$(echo "$lifespan_info" | cut -d'|' -f3)
|
||||
|
||||
echo "TBW Remaining: $tbw_remaining"
|
||||
echo "Lifespan: $lifespan_percent ($wear_status)"
|
||||
|
||||
elif [[ "$disk_type" == "HDD" ]]; then
|
||||
echo "Realloc Sectors: ${reallocated_sectors:-0}"
|
||||
echo "Pending Sectors: ${pending_sectors:-0}"
|
||||
|
||||
local lifespan=$(estimate_hdd_lifespan "$power_on_hours" "${reallocated_sectors:-0}" "${pending_sectors:-0}")
|
||||
echo "Lifespan: $lifespan"
|
||||
else
|
||||
print_color $YELLOW "Unknown disk type - limited information available"
|
||||
fi
|
||||
|
||||
echo ""
|
||||
}
|
||||
|
||||
# Function to detect RAID controllers and disks
|
||||
detect_raid_disks() {
|
||||
local controllers=("megaraid" "cciss" "areca" "3ware" "hpt")
|
||||
local disks=()
|
||||
|
||||
# Check for direct disks first
|
||||
for disk in /dev/sd[a-z] /dev/nvme[0-9]n[0-9]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk:direct")
|
||||
fi
|
||||
done
|
||||
|
||||
# Check for RAID controllers
|
||||
for controller in "${controllers[@]}"; do
|
||||
for i in {0..15}; do
|
||||
if smartctl -d "$controller,$i" -i /dev/sda >/dev/null 2>&1; then
|
||||
disks+=("/dev/sda:$controller,$i")
|
||||
fi
|
||||
done
|
||||
done
|
||||
|
||||
echo "${disks[@]}"
|
||||
}
|
||||
|
||||
# Main function
|
||||
main() {
|
||||
print_color $BLUE "Disk Health Check Script v$VERSION for Harvester OS"
|
||||
print_color $BLUE "===================================================="
|
||||
echo ""
|
||||
|
||||
local disks=()
|
||||
|
||||
# If specific disk provided, check only that disk
|
||||
if [[ $# -gt 0 ]]; then
|
||||
for disk in "$@"; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk:direct")
|
||||
else
|
||||
print_color $RED "Error: $disk is not a valid block device"
|
||||
fi
|
||||
done
|
||||
else
|
||||
# Auto-detect disks
|
||||
print_color $CYAN "Auto-detecting disks..."
|
||||
read -ra disks <<< "$(detect_raid_disks)"
|
||||
fi
|
||||
|
||||
if [[ ${#disks[@]} -eq 0 ]]; then
|
||||
print_color $RED "No disks found or accessible"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
print_color $GREEN "Found ${#disks[@]} disk(s) to check"
|
||||
echo ""
|
||||
|
||||
# Check each disk
|
||||
for disk_info in "${disks[@]}"; do
|
||||
IFS=':' read -r disk controller <<< "$disk_info"
|
||||
check_disk "$disk" "$controller"
|
||||
done
|
||||
|
||||
print_color $BLUE "Check completed!"
|
||||
}
|
||||
|
||||
# Usage information
|
||||
usage() {
|
||||
echo "Usage: $SCRIPT_NAME [DISK1 DISK2 ...]"
|
||||
echo ""
|
||||
echo "If no disks specified, auto-detects all available disks and RAID arrays"
|
||||
echo ""
|
||||
echo "Examples:"
|
||||
echo " $SCRIPT_NAME # Check all auto-detected disks"
|
||||
echo " $SCRIPT_NAME /dev/sda # Check specific disk"
|
||||
echo " $SCRIPT_NAME /dev/sda /dev/sdb # Check multiple disks"
|
||||
}
|
||||
|
||||
# Parse command line arguments
|
||||
case "${1:-}" in
|
||||
-h|--help)
|
||||
usage
|
||||
exit 0
|
||||
;;
|
||||
-v|--version)
|
||||
echo "$SCRIPT_NAME version $VERSION"
|
||||
exit 0
|
||||
;;
|
||||
*)
|
||||
main "$@"
|
||||
;;
|
||||
esac
|
||||
542
harvester-v2.4.sh
Executable file
542
harvester-v2.4.sh
Executable file
@@ -0,0 +1,542 @@
|
||||
#!/bin/bash
|
||||
|
||||
# Disk Health Check Script for Harvester OS
|
||||
# Enhanced with SAS/PERC H730P controller support
|
||||
# Checks SSD TBW/lifespan and HDD health status
|
||||
|
||||
SCRIPT_NAME=$(basename "$0")
|
||||
VERSION="2.4"
|
||||
|
||||
# Color codes
|
||||
RED=$(tput setaf 1)
|
||||
GREEN=$(tput setaf 2)
|
||||
YELLOW=$(tput setaf 3)
|
||||
BLUE=$(tput setaf 4)
|
||||
CYAN=$(tput setaf 6)
|
||||
NC=$(tput sgr0)
|
||||
|
||||
# Function to print colored output
|
||||
print_color() {
|
||||
local color=$1
|
||||
local message=$2
|
||||
echo -e "${color}${message}${NC}"
|
||||
}
|
||||
|
||||
# Check if command exists
|
||||
command_exists() {
|
||||
command -v "$1" >/dev/null 2>&1
|
||||
}
|
||||
|
||||
if ! command_exists smartctl; then
|
||||
print_color $RED "Error: smartctl is not installed. Please install smartmontools package."
|
||||
exit 1
|
||||
fi
|
||||
|
||||
# Function to test SMART access and get available data
|
||||
test_smart_access() {
|
||||
local disk=$1
|
||||
local controller=$2
|
||||
|
||||
local smart_cmd="smartctl"
|
||||
[[ -n "$controller" ]] && smart_cmd+=" -d $controller"
|
||||
|
||||
# Test basic SMART access
|
||||
if ! $smart_cmd -i "$disk" &>/dev/null; then
|
||||
echo "no_access"
|
||||
return
|
||||
fi
|
||||
|
||||
# Check if SMART is enabled (don't enable it, just check status)
|
||||
local smart_info=$($smart_cmd -i "$disk" 2>/dev/null)
|
||||
local smart_available=$(echo "$smart_info" | grep "SMART support is:" | awk '{print $4}')
|
||||
local smart_enabled=$(echo "$smart_info" | grep "SMART support is:" | awk '{print $6}')
|
||||
|
||||
if [[ "$smart_available" != "Available" ]]; then
|
||||
echo "not_available"
|
||||
return
|
||||
fi
|
||||
|
||||
if [[ "$smart_enabled" != "Enabled" ]]; then
|
||||
echo "disabled"
|
||||
return
|
||||
fi
|
||||
|
||||
# Test attribute reading
|
||||
local attributes=$($smart_cmd -A "$disk" 2>/dev/null)
|
||||
if [[ -z "$attributes" ]]; then
|
||||
echo "no_attributes"
|
||||
return
|
||||
fi
|
||||
|
||||
# Check if we have basic attributes
|
||||
local power_on_hours=$(echo "$attributes" | grep -i "Power_On_Hours\|Power-On" | awk '{print $10}' | head -1)
|
||||
if [[ -z "$power_on_hours" ]]; then
|
||||
echo "limited_attributes"
|
||||
return
|
||||
fi
|
||||
|
||||
echo "full_access"
|
||||
}
|
||||
|
||||
# Function to get disk information with enhanced SAS support
|
||||
get_disk_info() {
|
||||
local disk=$1
|
||||
local controller=$2
|
||||
|
||||
local smart_cmd="smartctl"
|
||||
[[ -n "$controller" ]] && smart_cmd+=" -d $controller"
|
||||
|
||||
local info=$($smart_cmd -i "$disk" 2>/dev/null)
|
||||
local attributes=$($smart_cmd -A "$disk" 2>/dev/null)
|
||||
local health=$($smart_cmd -H "$disk" 2>/dev/null)
|
||||
|
||||
# Extract information with multiple fallbacks for SAS drives
|
||||
local model=$(echo "$info" | grep -i "Device Model:\|Product:" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
local vendor=$(echo "$info" | grep -i "Vendor:" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
[[ -n "$vendor" && -n "$model" ]] && model="$vendor $model"
|
||||
|
||||
local serial=$(echo "$info" | grep -i "Serial Number:\|Serial number:" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
|
||||
local capacity=$(echo "$info" | grep -i "User Capacity:\|Total NVM Capacity:" | cut -d: -f2 | sed 's/^[ \t]*//' | cut -d'[' -f1 | head -1)
|
||||
|
||||
local firmware=$(echo "$info" | grep -i "Firmware Version:\|Firmware revision:\|Revision:" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
|
||||
local health_status=$(echo "$health" | grep -i "result:\|SMART overall-health" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
[[ -z "$health_status" ]] && health_status=$(echo "$health" | grep -i "SMART overall-health" | awk -F'[' '{print $2}' | cut -d']' -f1)
|
||||
|
||||
# Get disk type with SAS support
|
||||
local disk_type="UNKNOWN"
|
||||
if echo "$info" | grep -qi "Solid State Device"; then
|
||||
disk_type="SSD"
|
||||
elif echo "$info" | grep -qi "Rotation Rate"; then
|
||||
disk_type="HDD"
|
||||
elif echo "$info" | grep -qi "SCSI\|SAS"; then
|
||||
# SAS drives often don't specify, check rotation rate
|
||||
if echo "$info" | grep -qi "15000\|10000\|7200"; then
|
||||
disk_type="HDD"
|
||||
else
|
||||
disk_type="SSD"
|
||||
fi
|
||||
fi
|
||||
|
||||
# Extract SMART attributes with multiple field attempts for SAS
|
||||
local power_on_hours=$(echo "$attributes" | grep -i "Power_On_Hours\|Power-On" | awk '{print $NF}' | head -1)
|
||||
[[ -z "$power_on_hours" ]] && power_on_hours=$(echo "$attributes" | grep -i "Power_On_Hours_and_Msec" | awk '{print $10}' | head -1)
|
||||
|
||||
local reallocated_sectors=$(echo "$attributes" | grep -i "Reallocated_Sector_Ct\|Reallocated_Event_Count" | awk '{print $NF}' | head -1)
|
||||
|
||||
local pending_sectors=$(echo "$attributes" | grep -i "Current_Pending_Sector" | awk '{print $NF}' | head -1)
|
||||
|
||||
local total_written=$(echo "$attributes" | grep -i "Total_LBAs_Written\|Host_Writes_32MiB\|Lifetime_Writes" | awk '{print $NF}' | head -1)
|
||||
local host_writes_32mib=$(echo "$attributes" | grep -i "Host_Writes_32MiB" | awk '{print $NF}' | head -1)
|
||||
|
||||
# For SAS drives, try to get media wearout for SSDs
|
||||
local media_wearout=$(echo "$attributes" | grep -i "Media_Wearout_Indicator\|Wear_Leveling_Count" | awk '{print $NF}' | head -1)
|
||||
|
||||
echo "$model|$serial|$capacity|$firmware|$health_status|$disk_type|$power_on_hours|$reallocated_sectors|$pending_sectors|$total_written|$host_writes_32mib|$media_wearout"
|
||||
}
|
||||
|
||||
# Function to calculate TBW for SSD
|
||||
calculate_tbw() {
|
||||
local raw_value=$1
|
||||
local sectors=$2
|
||||
|
||||
if [[ -n "$sectors" && "$sectors" != "0" ]]; then
|
||||
local bytes=$((sectors * 512))
|
||||
local tbw=$(echo "scale=2; $bytes / 1000 / 1000 / 1000 / 1000" | bc -l 2>/dev/null || echo "0")
|
||||
echo "$tbw"
|
||||
elif [[ -n "$raw_value" && "$raw_value" != "0" ]]; then
|
||||
local tbw=$(echo "scale=2; $raw_value * 32 / 1000 / 1000" | bc -l 2>/dev/null || echo "0")
|
||||
echo "$tbw"
|
||||
else
|
||||
echo "0"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to estimate SSD endurance based on model and capacity
|
||||
estimate_ssd_endurance() {
|
||||
local disk_model=$1
|
||||
local capacity_gb=$2
|
||||
|
||||
# SAS SSDs typically have very high endurance
|
||||
if echo "$disk_model" | grep -qi "ST600MP\|SEAGATE.*SSD\|SAS.*SSD"; then
|
||||
# Enterprise SAS SSDs - very high endurance
|
||||
if [[ $capacity_gb -ge 1000 ]]; then
|
||||
echo "10000" # 10PB for 1TB+ enterprise SAS SSD
|
||||
elif [[ $capacity_gb -ge 600 ]]; then
|
||||
echo "6000" # 6PB for 600GB enterprise SAS SSD
|
||||
elif [[ $capacity_gb -ge 400 ]]; then
|
||||
echo "4000" # 4PB for 400GB enterprise SAS SSD
|
||||
else
|
||||
echo "2000" # 2PB for smaller enterprise SAS SSD
|
||||
fi
|
||||
elif echo "$disk_model" | grep -qi "MTFDDAK\|MICRON\|INTEL\|SAMSUNG\|KIOXIA"; then
|
||||
# Enterprise SATA SSDs
|
||||
if [[ $capacity_gb -ge 1000 ]]; then
|
||||
echo "1200" # 1.2PB for 1TB enterprise
|
||||
elif [[ $capacity_gb -ge 480 ]]; then
|
||||
echo "600" # 600TB for 480GB enterprise
|
||||
elif [[ $capacity_gb -ge 240 ]]; then
|
||||
echo "300" # 300TB for 240GB enterprise
|
||||
else
|
||||
echo "150" # 150TB for smaller enterprise
|
||||
fi
|
||||
else
|
||||
# Consumer SSDs
|
||||
if [[ $capacity_gb -ge 1000 ]]; then
|
||||
echo "600" # 600TB for 1TB consumer
|
||||
elif [[ $capacity_gb -ge 480 ]]; then
|
||||
echo "300" # 300TB for 480GB consumer
|
||||
elif [[ $capacity_gb -ge 240 ]]; then
|
||||
echo "150" # 150TB for 240GB consumer
|
||||
elif [[ $capacity_gb -ge 120 ]]; then
|
||||
echo "80" # 80TB for 120GB consumer
|
||||
else
|
||||
echo "40" # 40TB for smaller drives
|
||||
fi
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to estimate SSD lifespan with TBW remaining
|
||||
estimate_ssd_lifespan() {
|
||||
local power_on_hours=$1
|
||||
local tbw_used=$2
|
||||
local disk_model=$3
|
||||
local capacity_gb=$4
|
||||
local media_wearout=$5
|
||||
|
||||
if [[ -z "$power_on_hours" || "$power_on_hours" -eq 0 ]]; then
|
||||
echo "Unknown||Unknown||Unknown"
|
||||
return
|
||||
fi
|
||||
|
||||
local estimated_endurance=$(estimate_ssd_endurance "$disk_model" "$capacity_gb")
|
||||
local tbw_remaining=$(echo "scale=2; $estimated_endurance - $tbw_used" | bc -l 2>/dev/null || echo "0")
|
||||
|
||||
# If we have media wearout indicator, use it for more accurate estimation
|
||||
if [[ -n "$media_wearout" && "$media_wearout" != "0" ]]; then
|
||||
if [[ $media_wearout -le 10 ]]; then
|
||||
echo "${RED}10%${NC}|${RED}${tbw_remaining} TB${NC}|${RED}Critical wear${NC}|media_wearout"
|
||||
elif [[ $media_wearout -le 30 ]]; then
|
||||
echo "${YELLOW}${media_wearout}%${NC}|${YELLOW}${tbw_remaining} TB${NC}|${YELLOW}High wear${NC}|media_wearout"
|
||||
elif [[ $media_wearout -le 70 ]]; then
|
||||
echo "${YELLOW}${media_wearout}%${NC}|${YELLOW}${tbw_remaining} TB${NC}|${YELLOW}Moderate wear${NC}|media_wearout"
|
||||
else
|
||||
echo "${GREEN}${media_wearout}%${NC}|${GREEN}${tbw_remaining} TB${NC}|${GREEN}Healthy${NC}|media_wearout"
|
||||
fi
|
||||
return
|
||||
fi
|
||||
|
||||
if [[ $(echo "$tbw_used > 0" | bc -l 2>/dev/null) -eq 1 ]]; then
|
||||
local lifespan_used=$(echo "scale=1; $tbw_used * 100 / $estimated_endurance" | bc -l 2>/dev/null || echo "0")
|
||||
local lifespan_remaining=$(echo "scale=1; 100 - $lifespan_used" | bc -l 2>/dev/null || echo "100")
|
||||
|
||||
if [[ $(echo "$lifespan_used >= 80" | bc -l) -eq 1 ]]; then
|
||||
echo "${RED}${lifespan_remaining}%${NC}|${RED}${tbw_remaining} TB${NC}|${RED}High wear${NC}|tbw"
|
||||
elif [[ $(echo "$lifespan_used >= 50" | bc -l) -eq 1 ]]; then
|
||||
echo "${YELLOW}${lifespan_remaining}%${NC}|${YELLOW}${tbw_remaining} TB${NC}|${YELLOW}Moderate wear${NC}|tbw"
|
||||
else
|
||||
echo "${GREEN}${lifespan_remaining}%${NC}|${GREEN}${tbw_remaining} TB${NC}|${GREEN}Healthy${NC}|tbw"
|
||||
fi
|
||||
else
|
||||
echo "Unknown|${estimated_endurance} TB|New|estimated"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to estimate HDD lifespan
|
||||
estimate_hdd_lifespan() {
|
||||
local power_on_hours=$1
|
||||
local reallocated_sectors=$2
|
||||
local pending_sectors=$3
|
||||
|
||||
if [[ -z "$power_on_hours" ]]; then
|
||||
echo "Unknown"
|
||||
return
|
||||
fi
|
||||
|
||||
power_on_hours=${power_on_hours:-0}
|
||||
reallocated_sectors=${reallocated_sectors:-0}
|
||||
pending_sectors=${pending_sectors:-0}
|
||||
|
||||
if [[ "$pending_sectors" -gt 0 ]]; then
|
||||
echo "${RED}CRITICAL${NC} (Pending sectors: $pending_sectors)"
|
||||
elif [[ "$reallocated_sectors" -gt 100 ]]; then
|
||||
echo "${RED}< 6 months${NC} (High reallocated sectors: $reallocated_sectors)"
|
||||
elif [[ "$reallocated_sectors" -gt 10 ]]; then
|
||||
echo "${YELLOW}6-12 months${NC} (Reallocated sectors: $reallocated_sectors)"
|
||||
elif [[ "$power_on_hours" -gt 40000 ]]; then
|
||||
echo "${YELLOW}1-2 years${NC} (High usage: $power_on_hours hours)"
|
||||
elif [[ "$power_on_hours" -gt 25000 ]]; then
|
||||
echo "${GREEN}2-3 years${NC} (Moderate usage: $power_on_hours hours)"
|
||||
else
|
||||
echo "${GREEN}> 3 years${NC} (Low usage: $power_on_hours hours)"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to check a single disk with enhanced error handling
|
||||
check_disk() {
|
||||
local disk=$1
|
||||
local controller=$2
|
||||
|
||||
print_color $CYAN "Checking disk: $disk (Controller: ${controller:-direct})"
|
||||
echo "=================================================="
|
||||
|
||||
# Test SMART access level
|
||||
local access_level=$(test_smart_access "$disk" "$controller")
|
||||
|
||||
case $access_level in
|
||||
"no_access")
|
||||
print_color $RED "ERROR: Cannot access disk through controller"
|
||||
echo "Possible reasons:"
|
||||
echo " - Controller doesn't support SMART passthrough"
|
||||
echo " - Disk is part of a hardware RAID array"
|
||||
echo " - Insufficient permissions (try running as root)"
|
||||
echo " - Controller busy or offline"
|
||||
echo ""
|
||||
return
|
||||
;;
|
||||
"not_available")
|
||||
print_color $YELLOW "SMART not available on this disk"
|
||||
echo "This disk does not support SMART monitoring"
|
||||
echo ""
|
||||
return
|
||||
;;
|
||||
"disabled")
|
||||
print_color $YELLOW "SMART is disabled on this disk"
|
||||
echo "SMART is available but currently disabled"
|
||||
echo "To enable manually: smartctl -s on ${controller:+-d $controller} $disk"
|
||||
echo ""
|
||||
return
|
||||
;;
|
||||
"no_attributes")
|
||||
print_color $YELLOW "WARNING: Cannot read SMART attributes"
|
||||
echo "This is common with hardware RAID controllers like PERC H730P"
|
||||
echo "Try checking through the RAID management interface"
|
||||
echo ""
|
||||
return
|
||||
;;
|
||||
"limited_attributes")
|
||||
print_color $YELLOW "NOTE: Limited SMART data available"
|
||||
echo "Controller is filtering some SMART attributes"
|
||||
;;
|
||||
esac
|
||||
|
||||
# Get disk information
|
||||
local disk_info=$(get_disk_info "$disk" "$controller")
|
||||
IFS='|' read -r model serial capacity firmware health_status disk_type power_on_hours reallocated_sectors pending_sectors total_written host_writes_32mib media_wearout <<< "$disk_info"
|
||||
|
||||
# Display basic information
|
||||
echo "Model: ${model:-Unknown}"
|
||||
echo "Serial: ${serial:-Unknown}"
|
||||
echo "Type: $disk_type"
|
||||
echo "Capacity: ${capacity:-Unknown}"
|
||||
echo "Firmware: ${firmware:-Unknown}"
|
||||
echo "Health: ${health_status:-Unknown}"
|
||||
|
||||
# Only show power on hours if available
|
||||
if [[ -n "$power_on_hours" && "$power_on_hours" != "0" ]]; then
|
||||
echo "Power On Hours: $power_on_hours"
|
||||
else
|
||||
echo "Power On Hours: Unknown"
|
||||
fi
|
||||
|
||||
# Disk type specific analysis
|
||||
if [[ "$disk_type" == "SSD" ]]; then
|
||||
local tbw_used=0
|
||||
if [[ -n "$total_written" && "$total_written" != "0" ]]; then
|
||||
tbw_used=$(calculate_tbw "" "$total_written")
|
||||
elif [[ -n "$host_writes_32mib" && "$host_writes_32mib" != "0" ]]; then
|
||||
tbw_used=$(calculate_tbw "$host_writes_32mib" "")
|
||||
fi
|
||||
|
||||
if [[ $(echo "$tbw_used > 0" | bc -l 2>/dev/null) -eq 1 ]]; then
|
||||
echo "TBW Used: ${tbw_used} TB"
|
||||
fi
|
||||
|
||||
# Estimate capacity for endurance calculation
|
||||
local capacity_gb=0
|
||||
if echo "$capacity" | grep -qi "GB"; then
|
||||
capacity_gb=$(echo "$capacity" | grep -o '[0-9.]*' | head -1 | cut -d. -f1)
|
||||
elif echo "$capacity" | grep -qi "TB"; then
|
||||
capacity_gb=$(echo "$capacity" | grep -o '[0-9.]*' | head -1 | awk '{print $1 * 1000}' | bc 2>/dev/null | cut -d. -f1)
|
||||
fi
|
||||
|
||||
local lifespan_info=$(estimate_ssd_lifespan "$power_on_hours" "$tbw_used" "$model" "$capacity_gb" "$media_wearout")
|
||||
local lifespan_percent=$(echo "$lifespan_info" | cut -d'|' -f1)
|
||||
local tbw_remaining=$(echo "$lifespan_info" | cut -d'|' -f2)
|
||||
local wear_status=$(echo "$lifespan_info" | cut -d'|' -f3)
|
||||
local wear_source=$(echo "$lifespan_info" | cut -d'|' -f4)
|
||||
|
||||
if [[ "$wear_source" != "media_wearout" && $(echo "$tbw_used > 0" | bc -l 2>/dev/null) -eq 1 ]]; then
|
||||
echo "TBW Remaining: $tbw_remaining"
|
||||
fi
|
||||
|
||||
echo "Lifespan: $lifespan_percent ($wear_status)"
|
||||
|
||||
# Show wear source if available
|
||||
if [[ "$wear_source" == "media_wearout" ]]; then
|
||||
echo "Wear Source: Media Wearout Indicator"
|
||||
elif [[ "$wear_source" == "tbw" ]]; then
|
||||
echo "Wear Source: TBW Calculation"
|
||||
elif [[ "$wear_source" == "estimated" ]]; then
|
||||
echo "Wear Source: Estimated Endurance"
|
||||
fi
|
||||
|
||||
elif [[ "$disk_type" == "HDD" ]]; then
|
||||
if [[ -n "$reallocated_sectors" && "$reallocated_sectors" != "0" ]]; then
|
||||
echo "Realloc Sectors: $reallocated_sectors"
|
||||
fi
|
||||
if [[ -n "$pending_sectors" && "$pending_sectors" != "0" ]]; then
|
||||
echo "Pending Sectors: $pending_sectors"
|
||||
fi
|
||||
|
||||
local lifespan=$(estimate_hdd_lifespan "$power_on_hours" "${reallocated_sectors:-0}" "${pending_sectors:-0}")
|
||||
echo "Lifespan: $lifespan"
|
||||
else
|
||||
print_color $YELLOW "Limited information available for this disk type"
|
||||
echo "This is normal for hardware RAID configurations like PERC H730P"
|
||||
echo "For detailed SAS drive information, use controller management tools"
|
||||
fi
|
||||
|
||||
echo ""
|
||||
}
|
||||
|
||||
# Function to detect RAID controllers and disks with PERC H730P support
|
||||
detect_raid_disks() {
|
||||
local controllers=("megaraid" "cciss" "areca" "3ware" "hpt" "auto")
|
||||
local disks=()
|
||||
|
||||
# Check for direct disks first - only main devices, no partitions
|
||||
for disk in /dev/sd[a-z]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk:direct")
|
||||
fi
|
||||
done
|
||||
|
||||
# Check for NVMe disks - only main devices, no partitions
|
||||
for disk in /dev/nvme[0-9]n[0-9]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk:direct")
|
||||
fi
|
||||
done
|
||||
|
||||
# Check for SAS disks directly via SCSI generic
|
||||
for disk in /dev/sg[0-9]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk:direct")
|
||||
fi
|
||||
done
|
||||
|
||||
# Check for RAID controllers with enhanced detection
|
||||
for controller in "${controllers[@]}"; do
|
||||
print_color $BLUE "Scanning for $controller controllers..."
|
||||
for i in {0..31}; do
|
||||
# Try different disk devices for each controller
|
||||
for base_disk in "/dev/sda" "/dev/sdb" "/dev/sdc" "/dev/sg0" "/dev/sg1"; do
|
||||
if [[ -b "$base_disk" ]]; then
|
||||
if smartctl -d "$controller,$i" -i "$base_disk" &>/dev/null; then
|
||||
disks+=("$base_disk:$controller,$i")
|
||||
print_color $GREEN " Found $controller,$i on $base_disk"
|
||||
break
|
||||
fi
|
||||
fi
|
||||
done
|
||||
done
|
||||
done
|
||||
|
||||
# Special detection for PERC H730P
|
||||
print_color $BLUE "Scanning for PERC H730P controllers..."
|
||||
if command_exists storcli; then
|
||||
print_color $GREEN " storcli detected - checking PERC H730P"
|
||||
disks+=("/dev/sda:perc-h730p")
|
||||
fi
|
||||
|
||||
echo "${disks[@]}"
|
||||
}
|
||||
|
||||
# Main function
|
||||
main() {
|
||||
print_color $BLUE "Harvester OS Disk Health Check Script v$VERSION"
|
||||
print_color $BLUE "Enhanced with PERC H730P and SAS Support"
|
||||
print_color $BLUE "============================================"
|
||||
echo ""
|
||||
|
||||
if ! command_exists smartctl; then
|
||||
print_color $RED "Error: smartctl is not installed. Please install smartmontools package."
|
||||
exit 1
|
||||
fi
|
||||
|
||||
local disks=()
|
||||
|
||||
# If specific disk provided, check only that disk
|
||||
if [[ $# -gt 0 ]]; then
|
||||
for disk in "$@"; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk:direct")
|
||||
else
|
||||
print_color $RED "Error: $disk is not a valid block device"
|
||||
fi
|
||||
done
|
||||
else
|
||||
# Auto-detect disks
|
||||
print_color $CYAN "Auto-detecting disks and RAID controllers..."
|
||||
read -ra disks <<< "$(detect_raid_disks)"
|
||||
fi
|
||||
|
||||
if [[ ${#disks[@]} -eq 0 ]]; then
|
||||
print_color $RED "No disks found or accessible"
|
||||
echo "Try running as root: sudo $0"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
print_color $GREEN "Found ${#disks[@]} disk(s) to check"
|
||||
echo ""
|
||||
|
||||
# Check if running as root
|
||||
if [[ $EUID -ne 0 ]]; then
|
||||
print_color $YELLOW "Warning: Not running as root."
|
||||
print_color $YELLOW "Some disks/controllers may show limited information."
|
||||
echo ""
|
||||
fi
|
||||
|
||||
# Check each disk
|
||||
for disk_info in "${disks[@]}"; do
|
||||
IFS=':' read -r disk controller <<< "$disk_info"
|
||||
check_disk "$disk" "$controller"
|
||||
done
|
||||
|
||||
print_color $BLUE "Check completed!"
|
||||
echo ""
|
||||
print_color $CYAN "Note: For PERC H730P controllers with SAS drives:"
|
||||
print_color $CYAN " - Use 'storcli /c0 show all' for detailed information"
|
||||
print_color $CYAN " - Use 'storcli /c0/eall/sall show' for physical disk status"
|
||||
print_color $CYAN " - Hardware RAID controllers often limit SMART data access"
|
||||
}
|
||||
|
||||
# Usage information
|
||||
usage() {
|
||||
echo "Usage: $SCRIPT_NAME [DISK1 DISK2 ...]"
|
||||
echo ""
|
||||
echo "If no disks specified, auto-detects all available disks and RAID arrays"
|
||||
echo ""
|
||||
echo "Examples:"
|
||||
echo " sudo $SCRIPT_NAME # Check all disks (recommended)"
|
||||
echo " $SCRIPT_NAME /dev/sda # Check specific disk"
|
||||
echo " $SCRIPT_NAME /dev/sg0 # Check SAS disk directly"
|
||||
echo " $SCRIPT_NAME /dev/sda /dev/sdb # Check multiple disks"
|
||||
}
|
||||
|
||||
# Parse command line arguments
|
||||
case "${1:-}" in
|
||||
-h|--help)
|
||||
usage
|
||||
exit 0
|
||||
;;
|
||||
-v|--version)
|
||||
echo "$SCRIPT_NAME version $VERSION"
|
||||
exit 0
|
||||
;;
|
||||
*)
|
||||
main "$@"
|
||||
;;
|
||||
esac
|
||||
615
ubuntu-v2.5.sh
Executable file
615
ubuntu-v2.5.sh
Executable file
@@ -0,0 +1,615 @@
|
||||
#!/bin/bash
|
||||
|
||||
# Disk Health Check Script for Ubuntu 24.04
|
||||
# Enhanced with SAS/PERC H730P controller support
|
||||
# Checks SSD TBW/lifespan and HDD health status
|
||||
|
||||
SCRIPT_NAME=$(basename "$0")
|
||||
VERSION="2.5"
|
||||
|
||||
# Color codes
|
||||
RED=$(tput setaf 1)
|
||||
GREEN=$(tput setaf 2)
|
||||
YELLOW=$(tput setaf 3)
|
||||
BLUE=$(tput setaf 4)
|
||||
CYAN=$(tput setaf 6)
|
||||
NC=$(tput sgr0)
|
||||
|
||||
# Function to print colored output
|
||||
print_color() {
|
||||
local color=$1
|
||||
local message=$2
|
||||
echo -e "${color}${message}${NC}"
|
||||
}
|
||||
|
||||
# Check if command exists
|
||||
command_exists() {
|
||||
command -v "$1" >/dev/null 2>&1
|
||||
}
|
||||
|
||||
# Check dependencies
|
||||
check_dependencies() {
|
||||
local missing=()
|
||||
|
||||
if ! command_exists smartctl; then
|
||||
missing+=("smartmontools")
|
||||
fi
|
||||
|
||||
if ! command_exists bc; then
|
||||
missing+=("bc")
|
||||
fi
|
||||
|
||||
if [[ ${#missing[@]} -gt 0 ]]; then
|
||||
print_color $RED "Error: Missing required packages: ${missing[*]}"
|
||||
echo "Install with: sudo apt update && sudo apt install ${missing[*]}"
|
||||
exit 1
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to test SMART access and get available data - FIXED VERSION
|
||||
test_smart_access() {
|
||||
local disk=$1
|
||||
local controller=$2
|
||||
|
||||
local smart_cmd="smartctl"
|
||||
[[ -n "$controller" ]] && smart_cmd+=" -d $controller"
|
||||
|
||||
# Test basic SMART access
|
||||
if ! $smart_cmd -i "$disk" &>/dev/null; then
|
||||
echo "no_access"
|
||||
return
|
||||
fi
|
||||
|
||||
# Get SMART information
|
||||
local smart_info=$($smart_cmd -i "$disk" 2>/dev/null)
|
||||
|
||||
# Check if SMART is available - FIXED PARSING
|
||||
if ! echo "$smart_info" | grep -q "SMART support is:"; then
|
||||
echo "not_available"
|
||||
return
|
||||
fi
|
||||
|
||||
# Extract SMART status - FIXED LOGIC
|
||||
local smart_support_line=$(echo "$smart_info" | grep "SMART support is:")
|
||||
local smart_available=$(echo "$smart_support_line" | grep -q "Available" && echo "Available" || echo "")
|
||||
local smart_enabled=$(echo "$smart_support_line" | grep -q "Enabled" && echo "Enabled" || echo "")
|
||||
|
||||
if [[ -z "$smart_available" ]]; then
|
||||
echo "not_available"
|
||||
return
|
||||
fi
|
||||
|
||||
if [[ -z "$smart_enabled" ]]; then
|
||||
echo "disabled"
|
||||
return
|
||||
fi
|
||||
|
||||
# Test attribute reading
|
||||
local attributes=$($smart_cmd -A "$disk" 2>/dev/null)
|
||||
if [[ -z "$attributes" ]] || ! echo "$attributes" | grep -q "ATTRIBUTE_NAME"; then
|
||||
echo "no_attributes"
|
||||
return
|
||||
fi
|
||||
|
||||
echo "full_access"
|
||||
}
|
||||
|
||||
# Function to get disk information with enhanced SAS support
|
||||
get_disk_info() {
|
||||
local disk=$1
|
||||
local controller=$2
|
||||
|
||||
local smart_cmd="smartctl"
|
||||
[[ -n "$controller" ]] && smart_cmd+=" -d $controller"
|
||||
|
||||
local info=$($smart_cmd -i "$disk" 2>/dev/null)
|
||||
local attributes=$($smart_cmd -A "$disk" 2>/dev/null)
|
||||
local health=$($smart_cmd -H "$disk" 2>/dev/null)
|
||||
|
||||
# Extract information with multiple fallbacks for SAS drives
|
||||
local model=$(echo "$info" | grep -i "Device Model:\|Product:" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
local vendor=$(echo "$info" | grep -i "Vendor:" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
[[ -n "$vendor" && -n "$model" ]] && model="$vendor $model"
|
||||
|
||||
local serial=$(echo "$info" | grep -i "Serial Number:\|Serial number:" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
|
||||
local capacity=$(echo "$info" | grep -i "User Capacity:\|Total NVM Capacity:" | cut -d: -f2 | sed 's/^[ \t]*//' | cut -d'[' -f1 | head -1)
|
||||
|
||||
local firmware=$(echo "$info" | grep -i "Firmware Version:\|Firmware revision:\|Revision:" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
|
||||
local health_status=$(echo "$health" | grep -i "result:\|SMART overall-health" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
[[ -z "$health_status" ]] && health_status=$(echo "$health" | grep -i "SMART overall-health" | awk -F'[' '{print $2}' | cut -d']' -f1)
|
||||
|
||||
# Get disk type with SAS support
|
||||
local disk_type="UNKNOWN"
|
||||
if echo "$info" | grep -qi "Solid State Device"; then
|
||||
disk_type="SSD"
|
||||
elif echo "$info" | grep -qi "Rotation Rate"; then
|
||||
disk_type="HDD"
|
||||
elif echo "$info" | grep -qi "SCSI\|SAS"; then
|
||||
# SAS drives often don't specify, check rotation rate
|
||||
if echo "$info" | grep -qi "15000\|10000\|7200"; then
|
||||
disk_type="HDD"
|
||||
else
|
||||
disk_type="SSD"
|
||||
fi
|
||||
fi
|
||||
|
||||
# Extract SMART attributes with multiple field attempts
|
||||
local power_on_hours=$(echo "$attributes" | grep -i "Power_On_Hours" | awk '{print $10}' | head -1)
|
||||
|
||||
local reallocated_sectors=$(echo "$attributes" | grep -i "Reallocated_Sector_Ct" | awk '{print $10}' | head -1)
|
||||
|
||||
local pending_sectors=$(echo "$attributes" | grep -i "Current_Pending_Sector" | awk '{print $10}' | head -1)
|
||||
|
||||
# For Kingston and other SSDs with different attribute names
|
||||
local total_written=$(echo "$attributes" | grep -i "Total_LBAs_Written\|Lifetime_Writes_GiB\|Host_Writes_32MiB\|Flash_Writes_GiB" | awk '{print $10}' | head -1)
|
||||
local host_writes_32mib=$(echo "$attributes" | grep -i "Host_Writes_32MiB" | awk '{print $10}' | head -1)
|
||||
|
||||
# For wear leveling indicators
|
||||
local media_wearout=$(echo "$attributes" | grep -i "Media_Wearout_Indicator\|Wear_Leveling_Count\|SSD_Life_Left" | awk '{print $10}' | head -1)
|
||||
|
||||
echo "$model|$serial|$capacity|$firmware|$health_status|$disk_type|$power_on_hours|$reallocated_sectors|$pending_sectors|$total_written|$host_writes_32mib|$media_wearout"
|
||||
}
|
||||
|
||||
# Function to calculate TBW for SSD - ENHANCED FOR KINGSTON
|
||||
calculate_tbw() {
|
||||
local raw_value=$1
|
||||
local sectors=$2
|
||||
local disk_model=$3
|
||||
|
||||
# Kingston SSDs use Lifetime_Writes_GiB and Flash_Writes_GiB
|
||||
if echo "$disk_model" | grep -qi "KINGSTON"; then
|
||||
if [[ -n "$raw_value" && "$raw_value" != "0" ]]; then
|
||||
# Convert from GiB to TB
|
||||
local tbw=$(echo "scale=2; $raw_value / 1000" | bc 2>/dev/null || echo "0")
|
||||
echo "$tbw"
|
||||
return
|
||||
fi
|
||||
fi
|
||||
|
||||
if [[ -n "$sectors" && "$sectors" != "0" ]]; then
|
||||
local bytes=$((sectors * 512))
|
||||
local tbw=$(echo "scale=2; $bytes / 1000 / 1000 / 1000 / 1000" | bc 2>/dev/null || echo "0")
|
||||
echo "$tbw"
|
||||
elif [[ -n "$raw_value" && "$raw_value" != "0" ]]; then
|
||||
local tbw=$(echo "scale=2; $raw_value * 32 / 1000 / 1000" | bc 2>/dev/null || echo "0")
|
||||
echo "$tbw"
|
||||
else
|
||||
echo "0"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to estimate SSD endurance based on model and capacity
|
||||
estimate_ssd_endurance() {
|
||||
local disk_model=$1
|
||||
local capacity_gb=$2
|
||||
|
||||
# Kingston consumer SSDs
|
||||
if echo "$disk_model" | grep -qi "KINGSTON.*SA400"; then
|
||||
if [[ $capacity_gb -ge 960 ]]; then
|
||||
echo "300" # 300TB for 960GB Kingston SA400
|
||||
elif [[ $capacity_gb -ge 480 ]]; then
|
||||
echo "150" # 150TB for 480GB Kingston
|
||||
else
|
||||
echo "80" # 80TB for smaller Kingston
|
||||
fi
|
||||
# SAS SSDs typically have very high endurance
|
||||
elif echo "$disk_model" | grep -qi "ST600MP\|SEAGATE.*SSD\|SAS.*SSD"; then
|
||||
# Enterprise SAS SSDs - very high endurance
|
||||
if [[ $capacity_gb -ge 1000 ]]; then
|
||||
echo "10000" # 10PB for 1TB+ enterprise SAS SSD
|
||||
elif [[ $capacity_gb -ge 600 ]]; then
|
||||
echo "6000" # 6PB for 600GB enterprise SAS SSD
|
||||
elif [[ $capacity_gb -ge 400 ]]; then
|
||||
echo "4000" # 4PB for 400GB enterprise SAS SSD
|
||||
else
|
||||
echo "2000" # 2PB for smaller enterprise SAS SSD
|
||||
fi
|
||||
elif echo "$disk_model" | grep -qi "MTFDDAK\|MICRON\|INTEL\|SAMSUNG\|KIOXIA\|WDC\|WESTERN DIGITAL"; then
|
||||
# Enterprise SATA/NVMe SSDs
|
||||
if [[ $capacity_gb -ge 1000 ]]; then
|
||||
echo "1200" # 1.2PB for 1TB enterprise
|
||||
elif [[ $capacity_gb -ge 480 ]]; then
|
||||
echo "600" # 600TB for 480GB enterprise
|
||||
elif [[ $capacity_gb -ge 240 ]]; then
|
||||
echo "300" # 300TB for 240GB enterprise
|
||||
else
|
||||
echo "150" # 150TB for smaller enterprise
|
||||
fi
|
||||
else
|
||||
# Consumer SSDs
|
||||
if [[ $capacity_gb -ge 1000 ]]; then
|
||||
echo "600" # 600TB for 1TB consumer
|
||||
elif [[ $capacity_gb -ge 480 ]]; then
|
||||
echo "300" # 300TB for 480GB consumer
|
||||
elif [[ $capacity_gb -ge 240 ]]; then
|
||||
echo "150" # 150TB for 240GB consumer
|
||||
elif [[ $capacity_gb -ge 120 ]]; then
|
||||
echo "80" # 80TB for 120GB consumer
|
||||
else
|
||||
echo "40" # 40TB for smaller drives
|
||||
fi
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to estimate SSD lifespan with TBW remaining
|
||||
estimate_ssd_lifespan() {
|
||||
local power_on_hours=$1
|
||||
local tbw_used=$2
|
||||
local disk_model=$3
|
||||
local capacity_gb=$4
|
||||
local media_wearout=$5
|
||||
|
||||
if [[ -z "$power_on_hours" || "$power_on_hours" -eq 0 ]]; then
|
||||
echo "Unknown||Unknown||Unknown"
|
||||
return
|
||||
fi
|
||||
|
||||
local estimated_endurance=$(estimate_ssd_endurance "$disk_model" "$capacity_gb")
|
||||
local tbw_remaining=$(echo "scale=2; $estimated_endurance - $tbw_used" | bc 2>/dev/null || echo "0")
|
||||
|
||||
# If we have media wearout indicator, use it for more accurate estimation
|
||||
if [[ -n "$media_wearout" && "$media_wearout" != "0" ]]; then
|
||||
# For Kingston, SSD_Life_Left is already a percentage
|
||||
if echo "$disk_model" | grep -qi "KINGSTON"; then
|
||||
if [[ $media_wearout -le 10 ]]; then
|
||||
echo "${RED}${media_wearout}%${NC}|${RED}${tbw_remaining} TB${NC}|${RED}Critical wear${NC}|media_wearout"
|
||||
elif [[ $media_wearout -le 30 ]]; then
|
||||
echo "${YELLOW}${media_wearout}%${NC}|${YELLOW}${tbw_remaining} TB${NC}|${YELLOW}High wear${NC}|media_wearout"
|
||||
elif [[ $media_wearout -le 70 ]]; then
|
||||
echo "${YELLOW}${media_wearout}%${NC}|${YELLOW}${tbw_remaining} TB${NC}|${YELLOW}Moderate wear${NC}|media_wearout"
|
||||
else
|
||||
echo "${GREEN}${media_wearout}%${NC}|${GREEN}${tbw_remaining} TB${NC}|${GREEN}Healthy${NC}|media_wearout"
|
||||
fi
|
||||
else
|
||||
# For other drives, media_wearout might be countdown from 100
|
||||
local wear_percent=$media_wearout
|
||||
if [[ $media_wearout -le 10 ]]; then
|
||||
echo "${RED}${wear_percent}%${NC}|${RED}${tbw_remaining} TB${NC}|${RED}Critical wear${NC}|media_wearout"
|
||||
elif [[ $media_wearout -le 30 ]]; then
|
||||
echo "${YELLOW}${wear_percent}%${NC}|${YELLOW}${tbw_remaining} TB${NC}|${YELLOW}High wear${NC}|media_wearout"
|
||||
elif [[ $media_wearout -le 70 ]]; then
|
||||
echo "${YELLOW}${wear_percent}%${NC}|${YELLOW}${tbw_remaining} TB${NC}|${YELLOW}Moderate wear${NC}|media_wearout"
|
||||
else
|
||||
echo "${GREEN}${wear_percent}%${NC}|${GREEN}${tbw_remaining} TB${NC}|${GREEN}Healthy${NC}|media_wearout"
|
||||
fi
|
||||
fi
|
||||
return
|
||||
fi
|
||||
|
||||
if [[ $(echo "$tbw_used > 0" | bc 2>/dev/null) -eq 1 ]]; then
|
||||
local lifespan_used=$(echo "scale=1; $tbw_used * 100 / $estimated_endurance" | bc 2>/dev/null || echo "0")
|
||||
local lifespan_remaining=$(echo "scale=1; 100 - $lifespan_used" | bc 2>/dev/null || echo "100")
|
||||
|
||||
if [[ $(echo "$lifespan_used >= 80" | bc 2>/dev/null) -eq 1 ]]; then
|
||||
echo "${RED}${lifespan_remaining}%${NC}|${RED}${tbw_remaining} TB${NC}|${RED}High wear${NC}|tbw"
|
||||
elif [[ $(echo "$lifespan_used >= 50" | bc 2>/dev/null) -eq 1 ]]; then
|
||||
echo "${YELLOW}${lifespan_remaining}%${NC}|${YELLOW}${tbw_remaining} TB${NC}|${YELLOW}Moderate wear${NC}|tbw"
|
||||
else
|
||||
echo "${GREEN}${lifespan_remaining}%${NC}|${GREEN}${tbw_remaining} TB${NC}|${GREEN}Healthy${NC}|tbw"
|
||||
fi
|
||||
else
|
||||
echo "Unknown|${estimated_endurance} TB|New|estimated"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to estimate HDD lifespan
|
||||
estimate_hdd_lifespan() {
|
||||
local power_on_hours=$1
|
||||
local reallocated_sectors=$2
|
||||
local pending_sectors=$3
|
||||
|
||||
if [[ -z "$power_on_hours" ]]; then
|
||||
echo "Unknown"
|
||||
return
|
||||
fi
|
||||
|
||||
power_on_hours=${power_on_hours:-0}
|
||||
reallocated_sectors=${reallocated_sectors:-0}
|
||||
pending_sectors=${pending_sectors:-0}
|
||||
|
||||
if [[ "$pending_sectors" -gt 0 ]]; then
|
||||
echo "${RED}CRITICAL${NC} (Pending sectors: $pending_sectors)"
|
||||
elif [[ "$reallocated_sectors" -gt 100 ]]; then
|
||||
echo "${RED}< 6 months${NC} (High reallocated sectors: $reallocated_sectors)"
|
||||
elif [[ "$reallocated_sectors" -gt 10 ]]; then
|
||||
echo "${YELLOW}6-12 months${NC} (Reallocated sectors: $reallocated_sectors)"
|
||||
elif [[ "$power_on_hours" -gt 40000 ]]; then
|
||||
echo "${YELLOW}1-2 years${NC} (High usage: $power_on_hours hours)"
|
||||
elif [[ "$power_on_hours" -gt 25000 ]]; then
|
||||
echo "${GREEN}2-3 years${NC} (Moderate usage: $power_on_hours hours)"
|
||||
else
|
||||
echo "${GREEN}> 3 years${NC} (Low usage: $power_on_hours hours)"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to check a single disk with enhanced error handling
|
||||
check_disk() {
|
||||
local disk=$1
|
||||
local controller=$2
|
||||
|
||||
print_color $CYAN "Checking disk: $disk (Controller: ${controller:-direct})"
|
||||
echo "=================================================="
|
||||
|
||||
# Test SMART access level
|
||||
local access_level=$(test_smart_access "$disk" "$controller")
|
||||
|
||||
case $access_level in
|
||||
"no_access")
|
||||
print_color $RED "ERROR: Cannot access disk through controller"
|
||||
echo "Possible reasons:"
|
||||
echo " - Controller doesn't support SMART passthrough"
|
||||
echo " - Disk is part of a hardware RAID array"
|
||||
echo " - Insufficient permissions (try running as root)"
|
||||
echo " - Controller busy or offline"
|
||||
echo ""
|
||||
return
|
||||
;;
|
||||
"not_available")
|
||||
print_color $YELLOW "SMART not available on this disk"
|
||||
echo "This disk does not support SMART monitoring"
|
||||
echo ""
|
||||
return
|
||||
;;
|
||||
"disabled")
|
||||
print_color $YELLOW "SMART is disabled on this disk"
|
||||
echo "SMART is available but currently disabled"
|
||||
echo "To enable manually: smartctl -s on ${controller:+-d $controller} $disk"
|
||||
echo ""
|
||||
return
|
||||
;;
|
||||
"no_attributes")
|
||||
print_color $YELLOW "WARNING: Cannot read SMART attributes"
|
||||
echo "This is common with hardware RAID controllers like PERC H730P"
|
||||
echo "Try checking through the RAID management interface"
|
||||
echo ""
|
||||
return
|
||||
;;
|
||||
"limited_attributes")
|
||||
print_color $YELLOW "NOTE: Limited SMART data available"
|
||||
echo "Controller is filtering some SMART attributes"
|
||||
;;
|
||||
esac
|
||||
|
||||
# Get disk information
|
||||
local disk_info=$(get_disk_info "$disk" "$controller")
|
||||
IFS='|' read -r model serial capacity firmware health_status disk_type power_on_hours reallocated_sectors pending_sectors total_written host_writes_32mib media_wearout <<< "$disk_info"
|
||||
|
||||
# Display basic information
|
||||
echo "Model: ${model:-Unknown}"
|
||||
echo "Serial: ${serial:-Unknown}"
|
||||
echo "Type: $disk_type"
|
||||
echo "Capacity: ${capacity:-Unknown}"
|
||||
echo "Firmware: ${firmware:-Unknown}"
|
||||
echo "Health: ${health_status:-Unknown}"
|
||||
|
||||
# Only show power on hours if available
|
||||
if [[ -n "$power_on_hours" && "$power_on_hours" != "0" ]]; then
|
||||
echo "Power On Hours: $power_on_hours"
|
||||
else
|
||||
echo "Power On Hours: Unknown"
|
||||
fi
|
||||
|
||||
# Disk type specific analysis
|
||||
if [[ "$disk_type" == "SSD" ]]; then
|
||||
local tbw_used=0
|
||||
if [[ -n "$total_written" && "$total_written" != "0" ]]; then
|
||||
tbw_used=$(calculate_tbw "" "$total_written" "$model")
|
||||
elif [[ -n "$host_writes_32mib" && "$host_writes_32mib" != "0" ]]; then
|
||||
tbw_used=$(calculate_tbw "$host_writes_32mib" "" "$model")
|
||||
fi
|
||||
|
||||
if [[ $(echo "$tbw_used > 0" | bc 2>/dev/null) -eq 1 ]]; then
|
||||
echo "TBW Used: ${tbw_used} TB"
|
||||
fi
|
||||
|
||||
# Estimate capacity for endurance calculation
|
||||
local capacity_gb=0
|
||||
if echo "$capacity" | grep -qi "GB"; then
|
||||
capacity_gb=$(echo "$capacity" | grep -o '[0-9.]*' | head -1 | cut -d. -f1)
|
||||
elif echo "$capacity" | grep -qi "TB"; then
|
||||
capacity_gb=$(echo "$capacity" | grep -o '[0-9.]*' | head -1 | awk '{print $1 * 1000}' | bc 2>/dev/null | cut -d. -f1)
|
||||
fi
|
||||
|
||||
local lifespan_info=$(estimate_ssd_lifespan "$power_on_hours" "$tbw_used" "$model" "$capacity_gb" "$media_wearout")
|
||||
local lifespan_percent=$(echo "$lifespan_info" | cut -d'|' -f1)
|
||||
local tbw_remaining=$(echo "$lifespan_info" | cut -d'|' -f2)
|
||||
local wear_status=$(echo "$lifespan_info" | cut -d'|' -f3)
|
||||
local wear_source=$(echo "$lifespan_info" | cut -d'|' -f4)
|
||||
|
||||
if [[ "$wear_source" != "media_wearout" && $(echo "$tbw_used > 0" | bc 2>/dev/null) -eq 1 ]]; then
|
||||
echo "TBW Remaining: $tbw_remaining"
|
||||
fi
|
||||
|
||||
echo "Lifespan: $lifespan_percent ($wear_status)"
|
||||
|
||||
# Show wear source if available
|
||||
if [[ "$wear_source" == "media_wearout" ]]; then
|
||||
echo "Wear Source: Media Wearout Indicator"
|
||||
elif [[ "$wear_source" == "tbw" ]]; then
|
||||
echo "Wear Source: TBW Calculation"
|
||||
elif [[ "$wear_source" == "estimated" ]]; then
|
||||
echo "Wear Source: Estimated Endurance"
|
||||
fi
|
||||
|
||||
elif [[ "$disk_type" == "HDD" ]]; then
|
||||
if [[ -n "$reallocated_sectors" && "$reallocated_sectors" != "0" ]]; then
|
||||
echo "Realloc Sectors: $reallocated_sectors"
|
||||
fi
|
||||
if [[ -n "$pending_sectors" && "$pending_sectors" != "0" ]]; then
|
||||
echo "Pending Sectors: $pending_sectors"
|
||||
fi
|
||||
|
||||
local lifespan=$(estimate_hdd_lifespan "$power_on_hours" "${reallocated_sectors:-0}" "${pending_sectors:-0}")
|
||||
echo "Lifespan: $lifespan"
|
||||
else
|
||||
print_color $YELLOW "Limited information available for this disk type"
|
||||
echo "This is normal for hardware RAID configurations like PERC H730P"
|
||||
echo "For detailed SAS drive information, use controller management tools"
|
||||
fi
|
||||
|
||||
echo ""
|
||||
}
|
||||
|
||||
# Function to detect all disks with enhanced SAS support (no partitions) - FIXED
|
||||
detect_disks() {
|
||||
local disks=()
|
||||
|
||||
# Check for SATA/SAS disks - only main devices, no partitions
|
||||
for disk in /dev/sd[a-z]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
fi
|
||||
done
|
||||
|
||||
# Check for NVMe disks - only main devices, no partitions
|
||||
for disk in /dev/nvme[0-9]n[0-9]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
fi
|
||||
done
|
||||
|
||||
# Check for SAS disks via SCSI generic - only main devices
|
||||
for disk in /dev/sg[0-9]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
fi
|
||||
done
|
||||
|
||||
# Check for other disk types - only main devices
|
||||
for disk in /dev/vd[a-z] /dev/xvd[a-z]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
fi
|
||||
done
|
||||
|
||||
echo "${disks[@]}"
|
||||
}
|
||||
|
||||
# Function to detect RAID controllers (Ubuntu specific) - FIXED
|
||||
detect_raid_controllers() {
|
||||
local controllers=("megaraid" "cciss" "areca" "3ware" "hpt")
|
||||
local raid_disks=()
|
||||
|
||||
# Check for RAID controllers
|
||||
for controller in "${controllers[@]}"; do
|
||||
for i in {0..31}; do
|
||||
# Try different disk devices for each controller
|
||||
for base_disk in "/dev/sda" "/dev/sdb" "/dev/sdc" "/dev/sg0" "/dev/sg1"; do
|
||||
if [[ -b "$base_disk" ]]; then
|
||||
if smartctl -d "$controller,$i" -i "$base_disk" &>/dev/null; then
|
||||
raid_disks+=("$base_disk:$controller,$i")
|
||||
break
|
||||
fi
|
||||
fi
|
||||
done
|
||||
done
|
||||
done
|
||||
|
||||
echo "${raid_disks[@]}"
|
||||
}
|
||||
|
||||
# Main function - FIXED
|
||||
main() {
|
||||
print_color $BLUE "Ubuntu 24.04 Disk Health Check Script v$VERSION"
|
||||
print_color $BLUE "Enhanced with PERC H730P and SAS Support"
|
||||
print_color $BLUE "============================================"
|
||||
echo ""
|
||||
|
||||
check_dependencies
|
||||
|
||||
local disks=()
|
||||
|
||||
# If specific disk provided, check only that disk
|
||||
if [[ $# -gt 0 ]]; then
|
||||
for disk in "$@"; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
else
|
||||
print_color $RED "Error: $disk is not a valid block device"
|
||||
fi
|
||||
done
|
||||
else
|
||||
# Auto-detect disks - FIXED: don't mix output with disk detection
|
||||
print_color $CYAN "Auto-detecting disks (excluding partitions)..."
|
||||
local direct_disks=()
|
||||
read -ra direct_disks <<< "$(detect_disks)"
|
||||
|
||||
print_color $CYAN "Scanning for RAID controllers..."
|
||||
local raid_disks=()
|
||||
read -ra raid_disks <<< "$(detect_raid_controllers)"
|
||||
|
||||
# Combine both lists
|
||||
disks=("${direct_disks[@]}" "${raid_disks[@]}")
|
||||
fi
|
||||
|
||||
if [[ ${#disks[@]} -eq 0 ]]; then
|
||||
print_color $RED "No disks found or accessible"
|
||||
echo "Try running as root or specifying disk paths manually"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
print_color $GREEN "Found ${#disks[@]} disk(s) to check"
|
||||
echo ""
|
||||
|
||||
# Check if running as root, warn if not
|
||||
if [[ $EUID -ne 0 ]]; then
|
||||
print_color $YELLOW "Warning: Not running as root."
|
||||
print_color $YELLOW "Some disks/controllers may show limited information."
|
||||
echo "For complete results, run as: sudo $0"
|
||||
echo ""
|
||||
fi
|
||||
|
||||
# Check each disk
|
||||
for disk_info in "${disks[@]}"; do
|
||||
# Check if this is a RAID disk (has controller specified)
|
||||
if [[ "$disk_info" == *":"* ]]; then
|
||||
IFS=':' read -r disk controller <<< "$disk_info"
|
||||
check_disk "$disk" "$controller"
|
||||
else
|
||||
check_disk "$disk_info"
|
||||
fi
|
||||
done
|
||||
|
||||
print_color $BLUE "Check completed!"
|
||||
echo ""
|
||||
print_color $CYAN "Note: For PERC H730P controllers with SAS drives:"
|
||||
print_color $CYAN " - Install 'storcli' for detailed controller information"
|
||||
print_color $CYAN " - Use 'smartctl -d sat /dev/sgX' to try direct access"
|
||||
print_color $CYAN " - Hardware RAID controllers often limit SMART data access"
|
||||
echo ""
|
||||
print_color $CYAN "Ubuntu-specific tips:"
|
||||
print_color $CYAN " - Use 'lsblk' to see all available block devices"
|
||||
print_color $CYAN " - Use 'lshw -class disk' for detailed disk information"
|
||||
}
|
||||
|
||||
# Usage information
|
||||
usage() {
|
||||
echo "Usage: $SCRIPT_NAME [DISK1 DISK2 ...]"
|
||||
echo ""
|
||||
echo "If no disks specified, auto-detects all available disks"
|
||||
echo ""
|
||||
echo "Examples:"
|
||||
echo " $SCRIPT_NAME # Check all auto-detected disks"
|
||||
echo " sudo $SCRIPT_NAME # Check all disks (as root)"
|
||||
echo " $SCRIPT_NAME /dev/sda # Check specific disk"
|
||||
echo " $SCRIPT_NAME /dev/nvme0n1 # Check NVMe disk"
|
||||
echo " $SCRIPT_NAME /dev/sg0 # Check SAS disk directly"
|
||||
echo " $SCRIPT_NAME /dev/sda /dev/nvme0n1 # Check multiple disks"
|
||||
}
|
||||
|
||||
# Parse command line arguments
|
||||
case "${1:-}" in
|
||||
-h|--help)
|
||||
usage
|
||||
exit 0
|
||||
;;
|
||||
-v|--version)
|
||||
echo "$SCRIPT_NAME version $VERSION"
|
||||
exit 0
|
||||
;;
|
||||
*)
|
||||
main "$@"
|
||||
;;
|
||||
esac
|
||||
650
ubuntu-v2.6.sh
Executable file
650
ubuntu-v2.6.sh
Executable file
@@ -0,0 +1,650 @@
|
||||
#!/bin/bash
|
||||
|
||||
# Disk Health Check Script for Ubuntu 24.04
|
||||
# Enhanced with SAS/PERC H730P controller support
|
||||
# Checks SSD TBW/lifespan and HDD health status
|
||||
|
||||
SCRIPT_NAME=$(basename "$0")
|
||||
VERSION="2.6"
|
||||
|
||||
# Color codes
|
||||
RED=$(tput setaf 1)
|
||||
GREEN=$(tput setaf 2)
|
||||
YELLOW=$(tput setaf 3)
|
||||
BLUE=$(tput setaf 4)
|
||||
CYAN=$(tput setaf 6)
|
||||
NC=$(tput sgr0)
|
||||
|
||||
# Function to print colored output
|
||||
print_color() {
|
||||
local color=$1
|
||||
local message=$2
|
||||
echo -e "${color}${message}${NC}"
|
||||
}
|
||||
|
||||
# Check if command exists
|
||||
command_exists() {
|
||||
command -v "$1" >/dev/null 2>&1
|
||||
}
|
||||
|
||||
# Check dependencies
|
||||
check_dependencies() {
|
||||
local missing=()
|
||||
|
||||
if ! command_exists smartctl; then
|
||||
missing+=("smartmontools")
|
||||
fi
|
||||
|
||||
if ! command_exists bc; then
|
||||
missing+=("bc")
|
||||
fi
|
||||
|
||||
if [[ ${#missing[@]} -gt 0 ]]; then
|
||||
print_color $RED "Error: Missing required packages: ${missing[*]}"
|
||||
echo "Install with: sudo apt update && sudo apt install ${missing[*]}"
|
||||
exit 1
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to test SMART access and get available data - ENHANCED FOR NVMe
|
||||
test_smart_access() {
|
||||
local disk=$1
|
||||
local controller=$2
|
||||
|
||||
local smart_cmd="smartctl"
|
||||
[[ -n "$controller" ]] && smart_cmd+=" -d $controller"
|
||||
|
||||
# Test basic SMART access
|
||||
if ! $smart_cmd -i "$disk" &>/dev/null; then
|
||||
echo "no_access"
|
||||
return
|
||||
fi
|
||||
|
||||
# Get SMART information
|
||||
local smart_info=$($smart_cmd -i "$disk" 2>/dev/null)
|
||||
|
||||
# Check if this is an NVMe drive
|
||||
if echo "$smart_info" | grep -qi "NVMe"; then
|
||||
# NVMe drives have different SMART implementation
|
||||
if $smart_cmd -H "$disk" &>/dev/null; then
|
||||
echo "full_access"
|
||||
else
|
||||
echo "no_attributes"
|
||||
fi
|
||||
return
|
||||
fi
|
||||
|
||||
# Check if SMART is available for SATA/SAS
|
||||
if ! echo "$smart_info" | grep -q "SMART support is:"; then
|
||||
echo "not_available"
|
||||
return
|
||||
fi
|
||||
|
||||
# Extract SMART status
|
||||
local smart_support_line=$(echo "$smart_info" | grep "SMART support is:")
|
||||
local smart_available=$(echo "$smart_support_line" | grep -q "Available" && echo "Available" || echo "")
|
||||
local smart_enabled=$(echo "$smart_support_line" | grep -q "Enabled" && echo "Enabled" || echo "")
|
||||
|
||||
if [[ -z "$smart_available" ]]; then
|
||||
echo "not_available"
|
||||
return
|
||||
fi
|
||||
|
||||
if [[ -z "$smart_enabled" ]]; then
|
||||
echo "disabled"
|
||||
return
|
||||
fi
|
||||
|
||||
# Test attribute reading
|
||||
local attributes=$($smart_cmd -A "$disk" 2>/dev/null)
|
||||
if [[ -z "$attributes" ]] || ! echo "$attributes" | grep -q "ATTRIBUTE_NAME"; then
|
||||
echo "no_attributes"
|
||||
return
|
||||
fi
|
||||
|
||||
echo "full_access"
|
||||
}
|
||||
|
||||
# Function to get disk information with enhanced SAS and NVMe support
|
||||
get_disk_info() {
|
||||
local disk=$1
|
||||
local controller=$2
|
||||
|
||||
local smart_cmd="smartctl"
|
||||
[[ -n "$controller" ]] && smart_cmd+=" -d $controller"
|
||||
|
||||
local info=$($smart_cmd -i "$disk" 2>/dev/null)
|
||||
local attributes=$($smart_cmd -A "$disk" 2>/dev/null)
|
||||
local health=$($smart_cmd -H "$disk" 2>/dev/null)
|
||||
|
||||
# Extract information with multiple fallbacks
|
||||
local model=$(echo "$info" | grep -i "Device Model:\|Product:\|Model Number:" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
local vendor=$(echo "$info" | grep -i "Vendor:" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
[[ -n "$vendor" && -n "$model" ]] && model="$vendor $model"
|
||||
|
||||
local serial=$(echo "$info" | grep -i "Serial Number:\|Serial number:" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
|
||||
local capacity=$(echo "$info" | grep -i "User Capacity:\|Total NVM Capacity:\|Namespace 1 Size/Capacity:" | cut -d: -f2 | sed 's/^[ \t]*//' | cut -d'[' -f1 | head -1)
|
||||
|
||||
local firmware=$(echo "$info" | grep -i "Firmware Version:\|Firmware revision:\|Revision:" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
|
||||
local health_status=$(echo "$health" | grep -i "result:\|SMART overall-health\|Health Status:" | cut -d: -f2 | sed 's/^[ \t]*//' | head -1)
|
||||
[[ -z "$health_status" ]] && health_status=$(echo "$health" | grep -i "SMART overall-health" | awk -F'[' '{print $2}' | cut -d']' -f1)
|
||||
|
||||
# Get disk type
|
||||
local disk_type="UNKNOWN"
|
||||
if echo "$info" | grep -qi "Solid State Device\|NVMe"; then
|
||||
disk_type="SSD"
|
||||
elif echo "$info" | grep -qi "Rotation Rate"; then
|
||||
disk_type="HDD"
|
||||
elif echo "$info" | grep -qi "SCSI\|SAS"; then
|
||||
if echo "$info" | grep -qi "15000\|10000\|7200"; then
|
||||
disk_type="HDD"
|
||||
else
|
||||
disk_type="SSD"
|
||||
fi
|
||||
fi
|
||||
|
||||
# Extract SMART attributes with multiple field attempts
|
||||
local power_on_hours=$(echo "$attributes" | grep -i "Power_On_Hours" | awk '{print $10}' | head -1 | sed 's/[^0-9]//g')
|
||||
|
||||
local reallocated_sectors=$(echo "$attributes" | grep -i "Reallocated_Sector_Ct" | awk '{print $10}' | head -1)
|
||||
|
||||
local pending_sectors=$(echo "$attributes" | grep -i "Current_Pending_Sector" | awk '{print $10}' | head -1)
|
||||
|
||||
# For Kingston and other SSDs with different attribute names
|
||||
local total_written=$(echo "$attributes" | grep -i "Total_LBAs_Written\|Lifetime_Writes_GiB\|Host_Writes_32MiB\|Flash_Writes_GiB\|Data Units Written" | awk '{print $10}' | head -1)
|
||||
local host_writes_32mib=$(echo "$attributes" | grep -i "Host_Writes_32MiB" | awk '{print $10}' | head -1)
|
||||
|
||||
# For wear leveling indicators
|
||||
local media_wearout=$(echo "$attributes" | grep -i "Media_Wearout_Indicator\|Wear_Leveling_Count\|SSD_Life_Left\|Percentage Used\|Available Spare" | awk '{print $10}' | head -1)
|
||||
|
||||
echo "$model|$serial|$capacity|$firmware|$health_status|$disk_type|$power_on_hours|$reallocated_sectors|$pending_sectors|$total_written|$host_writes_32mib|$media_wearout"
|
||||
}
|
||||
|
||||
# Function to calculate TBW for SSD - ENHANCED FOR KINGSTON AND NVMe
|
||||
calculate_tbw() {
|
||||
local raw_value=$1
|
||||
local sectors=$2
|
||||
local disk_model=$3
|
||||
local attribute_name=$4
|
||||
|
||||
# Kingston SSDs use Lifetime_Writes_GiB and Flash_Writes_GiB
|
||||
if echo "$disk_model" | grep -qi "KINGSTON"; then
|
||||
if [[ -n "$raw_value" && "$raw_value" != "0" ]]; then
|
||||
# Convert from GiB to TB
|
||||
local tbw=$(echo "scale=2; $raw_value / 1000" | bc 2>/dev/null || echo "0")
|
||||
echo "$tbw"
|
||||
return
|
||||
fi
|
||||
fi
|
||||
|
||||
# NVMe drives use Data Units Written (1 unit = 1,000,000 bytes for NVMe 1.0+, 512,000 bytes for older)
|
||||
if echo "$attribute_name" | grep -qi "Data Units Written"; then
|
||||
if [[ -n "$raw_value" && "$raw_value" != "0" ]]; then
|
||||
# Convert from data units to TB (assuming 1,000,000 bytes per unit)
|
||||
local bytes=$(echo "$raw_value * 1000000" | bc 2>/dev/null)
|
||||
local tbw=$(echo "scale=2; $bytes / 1000 / 1000 / 1000 / 1000" | bc 2>/dev/null || echo "0")
|
||||
echo "$tbw"
|
||||
return
|
||||
fi
|
||||
fi
|
||||
|
||||
if [[ -n "$sectors" && "$sectors" != "0" ]]; then
|
||||
local bytes=$((sectors * 512))
|
||||
local tbw=$(echo "scale=2; $bytes / 1000 / 1000 / 1000 / 1000" | bc 2>/dev/null || echo "0")
|
||||
echo "$tbw"
|
||||
elif [[ -n "$raw_value" && "$raw_value" != "0" ]]; then
|
||||
local tbw=$(echo "scale=2; $raw_value * 32 / 1000 / 1000" | bc 2>/dev/null || echo "0")
|
||||
echo "$tbw"
|
||||
else
|
||||
echo "0"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to estimate SSD endurance based on model and capacity
|
||||
estimate_ssd_endurance() {
|
||||
local disk_model=$1
|
||||
local capacity_gb=$2
|
||||
|
||||
# Kingston consumer SSDs
|
||||
if echo "$disk_model" | grep -qi "KINGSTON.*SA400"; then
|
||||
if [[ $capacity_gb -ge 960 ]]; then
|
||||
echo "300" # 300TB for 960GB Kingston SA400
|
||||
elif [[ $capacity_gb -ge 480 ]]; then
|
||||
echo "150" # 150TB for 480GB Kingston
|
||||
else
|
||||
echo "80" # 80TB for smaller Kingston
|
||||
fi
|
||||
# NVMe SSDs typically have higher endurance
|
||||
elif echo "$disk_model" | grep -qi "NVMe"; then
|
||||
if [[ $capacity_gb -ge 2000 ]]; then
|
||||
echo "1200" # 1.2PB for 2TB+ NVMe
|
||||
elif [[ $capacity_gb -ge 1000 ]]; then
|
||||
echo "600" # 600TB for 1TB NVMe
|
||||
elif [[ $capacity_gb -ge 500 ]]; then
|
||||
echo "300" # 300TB for 500GB NVMe
|
||||
else
|
||||
echo "150" # 150TB for smaller NVMe
|
||||
fi
|
||||
# SAS SSDs typically have very high endurance
|
||||
elif echo "$disk_model" | grep -qi "ST600MP\|SEAGATE.*SSD\|SAS.*SSD"; then
|
||||
if [[ $capacity_gb -ge 1000 ]]; then
|
||||
echo "10000" # 10PB for 1TB+ enterprise SAS SSD
|
||||
elif [[ $capacity_gb -ge 600 ]]; then
|
||||
echo "6000" # 6PB for 600GB enterprise SAS SSD
|
||||
elif [[ $capacity_gb -ge 400 ]]; then
|
||||
echo "4000" # 4PB for 400GB enterprise SAS SSD
|
||||
else
|
||||
echo "2000" # 2PB for smaller enterprise SAS SSD
|
||||
fi
|
||||
elif echo "$disk_model" | grep -qi "MTFDDAK\|MICRON\|INTEL\|SAMSUNG\|KIOXIA\|WDC\|WESTERN DIGITAL"; then
|
||||
# Enterprise SATA/NVMe SSDs
|
||||
if [[ $capacity_gb -ge 1000 ]]; then
|
||||
echo "1200" # 1.2PB for 1TB enterprise
|
||||
elif [[ $capacity_gb -ge 480 ]]; then
|
||||
echo "600" # 600TB for 480GB enterprise
|
||||
elif [[ $capacity_gb -ge 240 ]]; then
|
||||
echo "300" # 300TB for 240GB enterprise
|
||||
else
|
||||
echo "150" # 150TB for smaller enterprise
|
||||
fi
|
||||
else
|
||||
# Consumer SSDs
|
||||
if [[ $capacity_gb -ge 1000 ]]; then
|
||||
echo "600" # 600TB for 1TB consumer
|
||||
elif [[ $capacity_gb -ge 480 ]]; then
|
||||
echo "300" # 300TB for 480GB consumer
|
||||
elif [[ $capacity_gb -ge 240 ]]; then
|
||||
echo "150" # 150TB for 240GB consumer
|
||||
elif [[ $capacity_gb -ge 120 ]]; then
|
||||
echo "80" # 80TB for 120GB consumer
|
||||
else
|
||||
echo "40" # 40TB for smaller drives
|
||||
fi
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to estimate SSD lifespan with TBW remaining - ENHANCED
|
||||
estimate_ssd_lifespan() {
|
||||
local power_on_hours=$1
|
||||
local tbw_used=$2
|
||||
local disk_model=$3
|
||||
local capacity_gb=$4
|
||||
local media_wearout=$5
|
||||
|
||||
if [[ -z "$power_on_hours" || "$power_on_hours" -eq 0 ]]; then
|
||||
echo "Unknown||Unknown||Unknown"
|
||||
return
|
||||
fi
|
||||
|
||||
local estimated_endurance=$(estimate_ssd_endurance "$disk_model" "$capacity_gb")
|
||||
local tbw_remaining=$(echo "scale=2; $estimated_endurance - $tbw_used" | bc 2>/dev/null || echo "0")
|
||||
|
||||
# If we have media wearout indicator, use it for more accurate estimation
|
||||
if [[ -n "$media_wearout" && "$media_wearout" != "0" ]]; then
|
||||
# For Kingston, SSD_Life_Left is already a percentage
|
||||
if echo "$disk_model" | grep -qi "KINGSTON"; then
|
||||
if [[ $media_wearout -le 10 ]]; then
|
||||
echo "${RED}${media_wearout}%${NC}|${RED}${tbw_remaining} TB${NC}|${RED}Critical wear${NC}|media_wearout"
|
||||
elif [[ $media_wearout -le 30 ]]; then
|
||||
echo "${YELLOW}${media_wearout}%${NC}|${YELLOW}${tbw_remaining} TB${NC}|${YELLOW}High wear${NC}|media_wearout"
|
||||
elif [[ $media_wearout -le 70 ]]; then
|
||||
echo "${YELLOW}${media_wearout}%${NC}|${YELLOW}${tbw_remaining} TB${NC}|${YELLOW}Moderate wear${NC}|media_wearout"
|
||||
else
|
||||
echo "${GREEN}${media_wearout}%${NC}|${GREEN}${tbw_remaining} TB${NC}|${GREEN}Healthy${NC}|media_wearout"
|
||||
fi
|
||||
else
|
||||
# For other drives, media_wearout might be countdown from 100
|
||||
local wear_percent=$media_wearout
|
||||
if [[ $media_wearout -le 10 ]]; then
|
||||
echo "${RED}${wear_percent}%${NC}|${RED}${tbw_remaining} TB${NC}|${RED}Critical wear${NC}|media_wearout"
|
||||
elif [[ $media_wearout -le 30 ]]; then
|
||||
echo "${YELLOW}${wear_percent}%${NC}|${YELLOW}${tbw_remaining} TB${NC}|${YELLOW}High wear${NC}|media_wearout"
|
||||
elif [[ $media_wearout -le 70 ]]; then
|
||||
echo "${YELLOW}${wear_percent}%${NC}|${YELLOW}${tbw_remaining} TB${NC}|${YELLOW}Moderate wear${NC}|media_wearout"
|
||||
else
|
||||
echo "${GREEN}${wear_percent}%${NC}|${GREEN}${tbw_remaining} TB${NC}|${GREEN}Healthy${NC}|media_wearout"
|
||||
fi
|
||||
fi
|
||||
return
|
||||
fi
|
||||
|
||||
if [[ $(echo "$tbw_used > 0" | bc 2>/dev/null) -eq 1 ]]; then
|
||||
local lifespan_used=$(echo "scale=1; $tbw_used * 100 / $estimated_endurance" | bc 2>/dev/null || echo "0")
|
||||
local lifespan_remaining=$(echo "scale=1; 100 - $lifespan_used" | bc 2>/dev/null || echo "100")
|
||||
|
||||
if [[ $(echo "$lifespan_used >= 80" | bc 2>/dev/null) -eq 1 ]]; then
|
||||
echo "${RED}${lifespan_remaining}%${NC}|${RED}${tbw_remaining} TB${NC}|${RED}High wear${NC}|tbw"
|
||||
elif [[ $(echo "$lifespan_used >= 50" | bc 2>/dev/null) -eq 1 ]]; then
|
||||
echo "${YELLOW}${lifespan_remaining}%${NC}|${YELLOW}${tbw_remaining} TB${NC}|${YELLOW}Moderate wear${NC}|tbw"
|
||||
else
|
||||
echo "${GREEN}${lifespan_remaining}%${NC}|${GREEN}${tbw_remaining} TB${NC}|${GREEN}Healthy${NC}|tbw"
|
||||
fi
|
||||
else
|
||||
echo "Unknown|${estimated_endurance} TB|New|estimated"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to estimate HDD lifespan - FIXED POWER_ON_HOURS PARSING
|
||||
estimate_hdd_lifespan() {
|
||||
local power_on_hours=$1
|
||||
local reallocated_sectors=$2
|
||||
local pending_sectors=$3
|
||||
|
||||
# Clean power_on_hours to extract just the numeric part
|
||||
local clean_hours=$(echo "$power_on_hours" | sed 's/[^0-9].*//')
|
||||
clean_hours=${clean_hours:-0}
|
||||
|
||||
if [[ -z "$clean_hours" || "$clean_hours" -eq 0 ]]; then
|
||||
echo "Unknown"
|
||||
return
|
||||
fi
|
||||
|
||||
reallocated_sectors=${reallocated_sectors:-0}
|
||||
pending_sectors=${pending_sectors:-0}
|
||||
|
||||
if [[ "$pending_sectors" -gt 0 ]]; then
|
||||
echo "${RED}CRITICAL${NC} (Pending sectors: $pending_sectors)"
|
||||
elif [[ "$reallocated_sectors" -gt 100 ]]; then
|
||||
echo "${RED}< 6 months${NC} (High reallocated sectors: $reallocated_sectors)"
|
||||
elif [[ "$reallocated_sectors" -gt 10 ]]; then
|
||||
echo "${YELLOW}6-12 months${NC} (Reallocated sectors: $reallocated_sectors)"
|
||||
elif [[ "$clean_hours" -gt 40000 ]]; then
|
||||
echo "${YELLOW}1-2 years${NC} (High usage: $clean_hours hours)"
|
||||
elif [[ "$clean_hours" -gt 25000 ]]; then
|
||||
echo "${GREEN}2-3 years${NC} (Moderate usage: $clean_hours hours)"
|
||||
else
|
||||
echo "${GREEN}> 3 years${NC} (Low usage: $clean_hours hours)"
|
||||
fi
|
||||
}
|
||||
|
||||
# Function to check a single disk with enhanced error handling
|
||||
check_disk() {
|
||||
local disk=$1
|
||||
local controller=$2
|
||||
|
||||
print_color $CYAN "Checking disk: $disk (Controller: ${controller:-direct})"
|
||||
echo "=================================================="
|
||||
|
||||
# Test SMART access level
|
||||
local access_level=$(test_smart_access "$disk" "$controller")
|
||||
|
||||
case $access_level in
|
||||
"no_access")
|
||||
print_color $RED "ERROR: Cannot access disk through controller"
|
||||
echo "Possible reasons:"
|
||||
echo " - Controller doesn't support SMART passthrough"
|
||||
echo " - Disk is part of a hardware RAID array"
|
||||
echo " - Insufficient permissions (try running as root)"
|
||||
echo " - Controller busy or offline"
|
||||
echo ""
|
||||
return
|
||||
;;
|
||||
"not_available")
|
||||
print_color $YELLOW "SMART not available on this disk"
|
||||
echo "This disk does not support SMART monitoring"
|
||||
echo ""
|
||||
return
|
||||
;;
|
||||
"disabled")
|
||||
print_color $YELLOW "SMART is disabled on this disk"
|
||||
echo "SMART is available but currently disabled"
|
||||
echo "To enable manually: smartctl -s on ${controller:+-d $controller} $disk"
|
||||
echo ""
|
||||
return
|
||||
;;
|
||||
"no_attributes")
|
||||
print_color $YELLOW "WARNING: Cannot read SMART attributes"
|
||||
echo "This is common with hardware RAID controllers like PERC H730P"
|
||||
echo "Try checking through the RAID management interface"
|
||||
echo ""
|
||||
return
|
||||
;;
|
||||
"limited_attributes")
|
||||
print_color $YELLOW "NOTE: Limited SMART data available"
|
||||
echo "Controller is filtering some SMART attributes"
|
||||
;;
|
||||
esac
|
||||
|
||||
# Get disk information
|
||||
local disk_info=$(get_disk_info "$disk" "$controller")
|
||||
IFS='|' read -r model serial capacity firmware health_status disk_type power_on_hours reallocated_sectors pending_sectors total_written host_writes_32mib media_wearout <<< "$disk_info"
|
||||
|
||||
# Display basic information
|
||||
echo "Model: ${model:-Unknown}"
|
||||
echo "Serial: ${serial:-Unknown}"
|
||||
echo "Type: $disk_type"
|
||||
echo "Capacity: ${capacity:-Unknown}"
|
||||
echo "Firmware: ${firmware:-Unknown}"
|
||||
echo "Health: ${health_status:-Unknown}"
|
||||
|
||||
# Only show power on hours if available
|
||||
if [[ -n "$power_on_hours" && "$power_on_hours" != "0" ]]; then
|
||||
echo "Power On Hours: $power_on_hours"
|
||||
else
|
||||
echo "Power On Hours: Unknown"
|
||||
fi
|
||||
|
||||
# Disk type specific analysis
|
||||
if [[ "$disk_type" == "SSD" ]]; then
|
||||
# Get the actual attribute name for TBW calculation
|
||||
local attributes=$(smartctl -A "$disk" 2>/dev/null)
|
||||
local tbw_attribute_name=$(echo "$attributes" | grep -i "Lifetime_Writes_GiB\|Flash_Writes_GiB\|Data Units Written" | head -1 | awk '{print $2}')
|
||||
|
||||
local tbw_used=0
|
||||
if [[ -n "$total_written" && "$total_written" != "0" ]]; then
|
||||
tbw_used=$(calculate_tbw "" "$total_written" "$model" "$tbw_attribute_name")
|
||||
elif [[ -n "$host_writes_32mib" && "$host_writes_32mib" != "0" ]]; then
|
||||
tbw_used=$(calculate_tbw "$host_writes_32mib" "" "$model" "$tbw_attribute_name")
|
||||
fi
|
||||
|
||||
# Always show TBW information for SSDs
|
||||
echo "TBW Used: ${tbw_used} TB"
|
||||
|
||||
# Estimate capacity for endurance calculation
|
||||
local capacity_gb=0
|
||||
if echo "$capacity" | grep -qi "GB"; then
|
||||
capacity_gb=$(echo "$capacity" | grep -o '[0-9.]*' | head -1 | cut -d. -f1)
|
||||
elif echo "$capacity" | grep -qi "TB"; then
|
||||
capacity_gb=$(echo "$capacity" | grep -o '[0-9.]*' | head -1 | awk '{print $1 * 1000}' | bc 2>/dev/null | cut -d. -f1)
|
||||
fi
|
||||
|
||||
local lifespan_info=$(estimate_ssd_lifespan "$power_on_hours" "$tbw_used" "$model" "$capacity_gb" "$media_wearout")
|
||||
local lifespan_percent=$(echo "$lifespan_info" | cut -d'|' -f1)
|
||||
local tbw_remaining=$(echo "$lifespan_info" | cut -d'|' -f2)
|
||||
local wear_status=$(echo "$lifespan_info" | cut -d'|' -f3)
|
||||
local wear_source=$(echo "$lifespan_info" | cut -d'|' -f4)
|
||||
|
||||
echo "TBW Remaining: $tbw_remaining"
|
||||
echo "Lifespan: $lifespan_percent ($wear_status)"
|
||||
|
||||
# Show wear source if available
|
||||
if [[ "$wear_source" == "media_wearout" ]]; then
|
||||
echo "Wear Source: Media Wearout Indicator"
|
||||
elif [[ "$wear_source" == "tbw" ]]; then
|
||||
echo "Wear Source: TBW Calculation"
|
||||
elif [[ "$wear_source" == "estimated" ]]; then
|
||||
echo "Wear Source: Estimated Endurance"
|
||||
fi
|
||||
|
||||
elif [[ "$disk_type" == "HDD" ]]; then
|
||||
if [[ -n "$reallocated_sectors" && "$reallocated_sectors" != "0" ]]; then
|
||||
echo "Realloc Sectors: $reallocated_sectors"
|
||||
fi
|
||||
if [[ -n "$pending_sectors" && "$pending_sectors" != "0" ]]; then
|
||||
echo "Pending Sectors: $pending_sectors"
|
||||
fi
|
||||
|
||||
local lifespan=$(estimate_hdd_lifespan "$power_on_hours" "${reallocated_sectors:-0}" "${pending_sectors:-0}")
|
||||
echo "Lifespan: $lifespan"
|
||||
else
|
||||
print_color $YELLOW "Limited information available for this disk type"
|
||||
echo "This is normal for hardware RAID configurations like PERC H730P"
|
||||
echo "For detailed SAS drive information, use controller management tools"
|
||||
fi
|
||||
|
||||
echo ""
|
||||
}
|
||||
|
||||
# Function to detect all disks with enhanced SAS support (no partitions)
|
||||
detect_disks() {
|
||||
local disks=()
|
||||
|
||||
# Check for SATA/SAS disks - only main devices, no partitions
|
||||
for disk in /dev/sd[a-z]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
fi
|
||||
done
|
||||
|
||||
# Check for NVMe disks - only main devices, no partitions
|
||||
for disk in /dev/nvme[0-9]n[0-9]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
fi
|
||||
done
|
||||
|
||||
# Check for SAS disks via SCSI generic - only main devices
|
||||
for disk in /dev/sg[0-9]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
fi
|
||||
done
|
||||
|
||||
# Check for other disk types - only main devices
|
||||
for disk in /dev/vd[a-z] /dev/xvd[a-z]; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
fi
|
||||
done
|
||||
|
||||
echo "${disks[@]}"
|
||||
}
|
||||
|
||||
# Function to detect RAID controllers (Ubuntu specific)
|
||||
detect_raid_controllers() {
|
||||
local controllers=("megaraid" "cciss" "areca" "3ware" "hpt")
|
||||
local raid_disks=()
|
||||
|
||||
# Check for RAID controllers
|
||||
for controller in "${controllers[@]}"; do
|
||||
for i in {0..31}; do
|
||||
# Try different disk devices for each controller
|
||||
for base_disk in "/dev/sda" "/dev/sdb" "/dev/sdc" "/dev/sg0" "/dev/sg1"; do
|
||||
if [[ -b "$base_disk" ]]; then
|
||||
if smartctl -d "$controller,$i" -i "$base_disk" &>/dev/null; then
|
||||
raid_disks+=("$base_disk:$controller,$i")
|
||||
break
|
||||
fi
|
||||
fi
|
||||
done
|
||||
done
|
||||
done
|
||||
|
||||
echo "${raid_disks[@]}"
|
||||
}
|
||||
|
||||
# Main function
|
||||
main() {
|
||||
print_color $BLUE "Ubuntu 24.04 Disk Health Check Script v$VERSION"
|
||||
print_color $BLUE "Enhanced with PERC H730P and SAS Support"
|
||||
print_color $BLUE "============================================"
|
||||
echo ""
|
||||
|
||||
check_dependencies
|
||||
|
||||
local disks=()
|
||||
|
||||
# If specific disk provided, check only that disk
|
||||
if [[ $# -gt 0 ]]; then
|
||||
for disk in "$@"; do
|
||||
if [[ -b "$disk" ]]; then
|
||||
disks+=("$disk")
|
||||
else
|
||||
print_color $RED "Error: $disk is not a valid block device"
|
||||
fi
|
||||
done
|
||||
else
|
||||
# Auto-detect disks
|
||||
print_color $CYAN "Auto-detecting disks (excluding partitions)..."
|
||||
local direct_disks=()
|
||||
read -ra direct_disks <<< "$(detect_disks)"
|
||||
|
||||
print_color $CYAN "Scanning for RAID controllers..."
|
||||
local raid_disks=()
|
||||
read -ra raid_disks <<< "$(detect_raid_controllers)"
|
||||
|
||||
# Combine both lists
|
||||
disks=("${direct_disks[@]}" "${raid_disks[@]}")
|
||||
fi
|
||||
|
||||
if [[ ${#disks[@]} -eq 0 ]]; then
|
||||
print_color $RED "No disks found or accessible"
|
||||
echo "Try running as root or specifying disk paths manually"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
print_color $GREEN "Found ${#disks[@]} disk(s) to check"
|
||||
echo ""
|
||||
|
||||
# Check if running as root, warn if not
|
||||
if [[ $EUID -ne 0 ]]; then
|
||||
print_color $YELLOW "Warning: Not running as root."
|
||||
print_color $YELLOW "Some disks/controllers may show limited information."
|
||||
echo "For complete results, run as: sudo $0"
|
||||
echo ""
|
||||
fi
|
||||
|
||||
# Check each disk
|
||||
for disk_info in "${disks[@]}"; do
|
||||
# Check if this is a RAID disk (has controller specified)
|
||||
if [[ "$disk_info" == *":"* ]]; then
|
||||
IFS=':' read -r disk controller <<< "$disk_info"
|
||||
check_disk "$disk" "$controller"
|
||||
else
|
||||
check_disk "$disk_info"
|
||||
fi
|
||||
done
|
||||
|
||||
print_color $BLUE "Check completed!"
|
||||
echo ""
|
||||
print_color $CYAN "Note: For PERC H730P controllers with SAS drives:"
|
||||
print_color $CYAN " - Install 'storcli' for detailed controller information"
|
||||
print_color $CYAN " - Use 'smartctl -d sat /dev/sgX' to try direct access"
|
||||
print_color $CYAN " - Hardware RAID controllers often limit SMART data access"
|
||||
echo ""
|
||||
print_color $CYAN "Ubuntu-specific tips:"
|
||||
print_color $CYAN " - Use 'lsblk' to see all available block devices"
|
||||
print_color $CYAN " - Use 'lshw -class disk' for detailed disk information"
|
||||
}
|
||||
|
||||
# Usage information
|
||||
usage() {
|
||||
echo "Usage: $SCRIPT_NAME [DISK1 DISK2 ...]"
|
||||
echo ""
|
||||
echo "If no disks specified, auto-detects all available disks"
|
||||
echo ""
|
||||
echo "Examples:"
|
||||
echo " $SCRIPT_NAME # Check all auto-detected disks"
|
||||
echo " sudo $SCRIPT_NAME # Check all disks (as root)"
|
||||
echo " $SCRIPT_NAME /dev/sda # Check specific disk"
|
||||
echo " $SCRIPT_NAME /dev/nvme0n1 # Check NVMe disk"
|
||||
echo " $SCRIPT_NAME /dev/sg0 # Check SAS disk directly"
|
||||
echo " $SCRIPT_NAME /dev/sda /dev/nvme0n1 # Check multiple disks"
|
||||
}
|
||||
|
||||
# Parse command line arguments
|
||||
case "${1:-}" in
|
||||
-h|--help)
|
||||
usage
|
||||
exit 0
|
||||
;;
|
||||
-v|--version)
|
||||
echo "$SCRIPT_NAME version $VERSION"
|
||||
exit 0
|
||||
;;
|
||||
*)
|
||||
main "$@"
|
||||
;;
|
||||
esac
|
||||
Reference in New Issue
Block a user