10.1 Failed Services
10.2 Filesystems Above Threshold (80%) .
10.3 Inodes Above Threshold
10.4 Memory Pressure Анализ PSI (Pressure Stall Information).
10.5 Swap Usage and Activity (swap-in/swap-out)
10.6 System Load – compare with Load Average
10.7 Disk and Filesystem Errors (I/O, EXT4, XFS, Btrfs, ZFS, NVMe and etc.)
10.8 Network Errors (Link Down, Carrier Loss, NETDEV WATCHDOG, TX Timeout and etc.).
10.9 SMART Health
10.10 ZFS Health (degraded-pools, ro/rw errors ,scrub)
10.11 mdadm RAID Health (degraded, recovery, reshape and etc.)
10.12 Time Synchronization (NTP).
10.13 Health Summary (Overall Health).
Why Base64?
Base64 encoding allows complex multi-line Bash scripts to be safely delivered as a single command, avoiding issues with line breaks, quoting, and macro parsing while preserving the original script intact. Use: printf %s <base64_data> | base64 -d | bash
printf %s '#!/usr/bin/env bash

# ==================== 10. HEALTH CHECK ====================

echo "==================== 10. HEALTH CHECK ===================="

printf "\n"

OVERALL_STATUS="OK"

# --- 10.1 FAILED SERVICES ---

echo "10.1 FAILED SERVICES"

FAILED_SERVICES_COUNT=$(
    systemctl --failed \
        --type=service \
        --no-legend \
        --plain \
        2>/dev/null | \
        awk 'NF {count++} END {print count+0}'
)

if [ "${FAILED_SERVICES_COUNT:-0}" -gt 0 ]; then
    echo "CRITICAL: $FAILED_SERVICES_COUNT failed service(s) detected"

    systemctl --failed \
        --type=service \
        --no-pager \
        2>/dev/null

    OVERALL_STATUS="CRITICAL"
else
    echo "OK: No failed systemd services detected"
fi

printf "\n%.0s" {1..3}

# --- 10.2 FILESYSTEMS ABOVE THRESHOLD ---

echo "10.2 FILESYSTEMS ABOVE THRESHOLD"

DF_OUTPUT=$(timeout 8 df -hT 2>/dev/null)

if [ -n "$DF_OUTPUT" ]; then
    FS_OVER_LIMIT=$(
        printf "%s\n" "$DF_OUTPUT" | \
        awk '
            NR > 1 {
                use=$6
                gsub(/%/, "", use)
                if (use ~ /^[0-9]+$/ && use >= 80) {
                    print
                }
            }
        '
    )

    if [ -n "$FS_OVER_LIMIT" ]; then
        echo "WARNING: Filesystems at or above 80% usage:"
        printf "%s\n" "$DF_OUTPUT" | head -1
        printf "%s\n" "$FS_OVER_LIMIT"

        [ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
    else
        echo "OK: All filesystems are below 80% usage"
    fi
else
    FS_OVER_LIMIT=""
    echo "WARNING: Filesystem usage query unavailable or timed out"

    [ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
fi

printf "\n%.0s" {1..3}

# --- 10.3 INODES ABOVE THRESHOLD ---

echo "10.3 INODES ABOVE THRESHOLD"

DF_INODE_OUTPUT=$(timeout 8 df -ih 2>/dev/null)

if [ -n "$DF_INODE_OUTPUT" ]; then
    INODE_OVER_LIMIT=$(
        printf "%s\n" "$DF_INODE_OUTPUT" | \
        awk '
            NR > 1 {
                use=$5
                gsub(/%/, "", use)
                if (use ~ /^[0-9]+$/ && use >= 80) {
                    print
                }
            }
        '
    )

    if [ -n "$INODE_OVER_LIMIT" ]; then
        echo "WARNING: Filesystems at or above 80% inode usage:"
        printf "%s\n" "$DF_INODE_OUTPUT" | head -1
        printf "%s\n" "$INODE_OVER_LIMIT"

        [ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
    else
        echo "OK: All filesystems are below 80% inode usage"
    fi
else
    INODE_OVER_LIMIT=""
    echo "WARNING: Inode usage query unavailable or timed out"

    [ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
fi

printf "\n%.0s" {1..3}

# --- 10.4 MEMORY PRESSURE ---

echo "10.4 MEMORY PRESSURE"

MEM_PRESS_AVG=""

if [ -f /proc/pressure/memory ]; then
    MEM_PRESS_AVG=$(
        awk '
            $1 == "some" {
                for (i=1; i<=NF; i++) {
                    if ($i ~ /^avg10=/) {
                        split($i, value, "=")
                        print value[2]
                        exit
                    }
                }
            }
        ' /proc/pressure/memory
    )

    if awk -v value="${MEM_PRESS_AVG:-0}" \
        'BEGIN {exit !(value > 10.0)}'
    then
        echo "WARNING: Memory PSI avg10 is high: ${MEM_PRESS_AVG:-unknown}"
        cat /proc/pressure/memory

        [ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
    else
        echo "OK: Memory PSI avg10 is within threshold: ${MEM_PRESS_AVG:-0}"
    fi
else
    echo "INFO: Memory PSI is not available on this kernel"
fi

printf "\n%.0s" {1..3}

# --- 10.5 SWAP USAGE AND ACTIVITY ---

echo "10.5 SWAP USAGE AND ACTIVITY"

SWAP_TOTAL_METRIC=$(
    free -b 2>/dev/null | \
    awk '/^Swap:/ {print $2}'
)

SWAP_USED_METRIC=$(
    free -b 2>/dev/null | \
    awk '/^Swap:/ {print $3}'
)

SWAP_PCT_METRIC=0
SWAP_ACTIVITY=0

if [ "${SWAP_TOTAL_METRIC:-0}" -gt 0 ]; then
    SWAP_PCT_METRIC=$(
        awk \
            -v used="${SWAP_USED_METRIC:-0}" \
            -v total="${SWAP_TOTAL_METRIC:-1}" \
            'BEGIN {printf "%.0f", (used/total)*100}'
    )

    SWAP_ACTIVITY=$(
        vmstat 1 2 2>/dev/null | \
        tail -1 | \
        awk '{print $7+$8}'
    )

    echo "Swap usage    : ${SWAP_PCT_METRIC}%"
    echo "Swap activity : ${SWAP_ACTIVITY:-0} pages/s combined si+so"

    if [ "${SWAP_ACTIVITY:-0}" -gt 0 ]; then
        echo "WARNING: Active swap-in or swap-out activity detected"
        free -h

        [ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
    elif [ "${SWAP_PCT_METRIC:-0}" -gt 50 ]; then
        echo "INFO: Swap usage is above 50%, but no current swap activity was detected"
    else
        echo "OK: Swap usage and activity are within expected limits"
    fi
else
    echo "INFO: Swap is not configured"
fi

printf "\n%.0s" {1..3}

# --- 10.6 SYSTEM LOAD ---

echo "10.6 SYSTEM LOAD"

read -r LOAD_ONE_MIN LOAD_FIVE_MIN LOAD_FIFTEEN_MIN _ < /proc/loadavg

CPU_CORES_COUNT=$(
    nproc 2>/dev/null || \
    grep -c '^processor' /proc/cpuinfo 2>/dev/null || \
    echo 1
)

LOAD_PER_CORE=$(
    awk \
        -v load="${LOAD_ONE_MIN:-0}" \
        -v cores="${CPU_CORES_COUNT:-1}" \
        'BEGIN {
            if (cores < 1) {
                cores=1
            }
            printf "%.2f", load/cores
        }'
)

echo "Load average : ${LOAD_ONE_MIN:-0} ${LOAD_FIVE_MIN:-0} ${LOAD_FIFTEEN_MIN:-0}"
echo "CPU cores    : ${CPU_CORES_COUNT:-1}"
echo "Load/core    : $LOAD_PER_CORE"

if awk -v value="$LOAD_PER_CORE" \
    'BEGIN {exit !(value >= 1.0)}'
then
    echo "CRITICAL: One-minute load is at or above 1.00 per CPU core"
    uptime

    OVERALL_STATUS="CRITICAL"
elif awk -v value="$LOAD_PER_CORE" \
    'BEGIN {exit !(value >= 0.70)}'
then
    echo "WARNING: One-minute load is at or above 0.70 per CPU core"
    uptime

    [ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
else
    echo "OK: Normalized load is within threshold"
fi

printf "\n%.0s" {1..3}

# --- 10.7 DISK AND FILESYSTEM ERRORS ---

echo "10.7 DISK AND FILESYSTEM ERRORS"

if command -v journalctl >/dev/null 2>&1; then
    DISK_ERR_HUNT=$(
        journalctl \
            -k \
            -b 0 \
            --no-pager \
            2>/dev/null | \
        grep -Ei \
            'I/O error|buffer I/O|blk_update_request|critical medium error|uncorrectable|EXT4-fs error|XFS.*error|BTRFS.*error|ZFS.*error|nvme.*error|bad sector|read-only file system' | \
        tail -15
    )
else
    DISK_ERR_HUNT=$(
        dmesg 2>/dev/null | \
        grep -Ei \
            'I/O error|buffer I/O|blk_update_request|critical medium error|uncorrectable|EXT4-fs error|XFS.*error|BTRFS.*error|ZFS.*error|nvme.*error|bad sector|read-only file system' | \
        tail -15
    )
fi

if [ -n "$DISK_ERR_HUNT" ]; then
    echo "CRITICAL: Disk or filesystem errors detected:"
    printf "%s\n" "$DISK_ERR_HUNT"

    OVERALL_STATUS="CRITICAL"
else
    echo "OK: No disk or filesystem errors detected in the current boot"
fi

printf "\n%.0s" {1..3}

# --- 10.8 NETWORK ERRORS ---

echo "10.8 NETWORK ERRORS"

if command -v journalctl >/dev/null 2>&1; then
    NET_ERR_HUNT=$(
        journalctl \
            -k \
            -b 0 \
            --no-pager \
            2>/dev/null | \
        grep -Ei \
            'link is down|link down|lost carrier|carrier lost|NETDEV WATCHDOG|transmit queue timed out|tx timeout|reset adapter|firmware failed' | \
        tail -15
    )
else
    NET_ERR_HUNT=$(
        dmesg 2>/dev/null | \
        grep -Ei \
            'link is down|link down|lost carrier|carrier lost|NETDEV WATCHDOG|transmit queue timed out|tx timeout|reset adapter|firmware failed' | \
        tail -15
    )
fi

if [ -n "$NET_ERR_HUNT" ]; then
    echo "WARNING: Network link or adapter errors detected:"
    printf "%s\n" "$NET_ERR_HUNT"

    [ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
else
    echo "OK: No network link or adapter errors detected in the current boot"
fi

printf "\n%.0s" {1..3}

# --- 10.9 SMART HEALTH ---

echo "10.9 SMART HEALTH"

SMART_ERR_FLAG=0
SMART_UNKNOWN_FLAG=0
SMART_DEVICE_COUNT=0

if command -v smartctl >/dev/null 2>&1; then
    while IFS= read -r DEV_ID; do
        [ -z "$DEV_ID" ] && continue

        SMART_DEVICE_COUNT=$((SMART_DEVICE_COUNT + 1))

        SMART_OUTPUT=$(
            smartctl -H "$DEV_ID" 2>/dev/null
        )

        if printf "%s\n" "$SMART_OUTPUT" | \
            grep -Eqi 'PASSED|SMART Health Status: OK|overall-health.*PASSED'
        then
            echo "OK: SMART health passed on $DEV_ID"
        elif printf "%s\n" "$SMART_OUTPUT" | \
            grep -Eqi 'FAILED|BAD|CRITICAL'
        then
            echo "CRITICAL: SMART health failure reported on $DEV_ID"
            printf "%s\n" "$SMART_OUTPUT"

            SMART_ERR_FLAG=1
            OVERALL_STATUS="CRITICAL"
        else
            echo "INFO: SMART health could not be determined for $DEV_ID"

            SMART_UNKNOWN_FLAG=1
        fi
    done < <(
        smartctl --scan 2>/dev/null | \
        awk '{print $1}'
    )

    if [ "$SMART_DEVICE_COUNT" -eq 0 ]; then
        echo "INFO: No SMART-capable devices discovered by smartctl --scan"
    elif [ "$SMART_ERR_FLAG" -eq 0 ] && [ "$SMART_UNKNOWN_FLAG" -eq 0 ]; then
        echo "OK: All discovered SMART devices reported healthy"
    fi
else
    echo "INFO: smartctl is not installed"
fi

printf "\n%.0s" {1..3}

# --- 10.10 ZFS HEALTH ---

echo "10.10 ZFS HEALTH"

ZFS_HEALTH_STATUS="NOT INSTALLED"

if command -v zpool >/dev/null 2>&1; then
    ZPOOL_FAULT_MAP=$(
        zpool status -x 2>/dev/null
    )

    if [ -z "$ZPOOL_FAULT_MAP" ]; then
        echo "INFO: No ZFS pools detected"
        ZFS_HEALTH_STATUS="NO POOLS"
    elif printf "%s\n" "$ZPOOL_FAULT_MAP" | \
        grep -qi 'all pools are healthy'
    then
        echo "OK: All ZFS pools are healthy"
        ZFS_HEALTH_STATUS="OK"
    else
        echo "CRITICAL: ZFS pool issue detected:"
        printf "%s\n" "$ZPOOL_FAULT_MAP"

        ZFS_HEALTH_STATUS="CRITICAL"
        OVERALL_STATUS="CRITICAL"
    fi
else
    echo "INFO: ZFS tools are not installed"
fi

printf "\n%.0s" {1..3}

# --- 10.11 MDADM RAID HEALTH ---

echo "10.11 MDADM RAID HEALTH"

MDADM_HEALTH_STATUS="NOT CONFIGURED"

if [ -r /proc/mdstat ] && grep -q '^md[0-9]' /proc/mdstat; then
    MDADM_HEALTH_STATUS="OK"

    MDADM_PROBLEMS=$(
        grep -E \
            'inactive|recovery|resync|reshape|check|\[[U_]*_[U_]*\]' \
            /proc/mdstat \
            2>/dev/null
    )

    if [ -n "$MDADM_PROBLEMS" ]; then
        echo "WARNING: mdadm array is degraded, inactive, or running maintenance:"
        cat /proc/mdstat

        MDADM_HEALTH_STATUS="WARNING"

        [ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
    else
        echo "OK: Active mdadm arrays appear healthy"
    fi
else
    echo "INFO: No active mdadm arrays detected"
fi

printf "\n%.0s" {1..3}

# --- 10.12 TIME SYNCHRONIZATION ---

echo "10.12 TIME SYNCHRONIZATION"

TIME_SYNC_STATUS="UNKNOWN"

if command -v timedatectl >/dev/null 2>&1; then
    TIME_SYNC_STATUS=$(
        timedatectl show \
            -p NTPSynchronized \
            --value \
            2>/dev/null
    )

    if [ "$TIME_SYNC_STATUS" = "yes" ]; then
        echo "OK: System clock is synchronized"
        TIME_SYNC_STATUS="OK"
    else
        echo "WARNING: System clock is not synchronized"
        timedatectl status 2>/dev/null

        TIME_SYNC_STATUS="WARNING"

        [ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
    fi
else
    echo "INFO: timedatectl is not available"
fi

printf "\n%.0s" {1..3}

# --- 10.13 SUMMARY ---

echo "10.13 SUMMARY"

echo "--- Core Global Infrastructure Health Matrix ---"

printf "Platform Node Instance...%s\n" \
    "$(hostname 2>/dev/null || echo unresolved)"

if [ "${FAILED_SERVICES_COUNT:-0}" -eq 0 ]; then
    echo "Systemd Services..........OK"
else
    echo "Systemd Services..........CRITICAL ($FAILED_SERVICES_COUNT failed)"
fi

if [ -z "$FS_OVER_LIMIT" ]; then
    echo "Filesystem Capacity.......OK"
else
    echo "Filesystem Capacity.......WARNING"
fi

if [ -z "$INODE_OVER_LIMIT" ]; then
    echo "Inode Capacity............OK"
else
    echo "Inode Capacity............WARNING"
fi

if [ -z "$DISK_ERR_HUNT" ]; then
    echo "Kernel Disk Health........OK"
else
    echo "Kernel Disk Health........CRITICAL"
fi

if [ -z "$NET_ERR_HUNT" ]; then
    echo "Network Link Health.......OK"
else
    echo "Network Link Health.......WARNING"
fi

echo "ZFS Health................$ZFS_HEALTH_STATUS"
echo "mdadm Health..............$MDADM_HEALTH_STATUS"
echo "Time Synchronization......$TIME_SYNC_STATUS"
echo "Normalized Load/Core......$LOAD_PER_CORE"
echo "Overall Health............$OVERALL_STATUS"
' | base64 -d | bash
# ==================== 10. HEALTH CHECK ====================
echo "==================== 10. HEALTH CHECK ===================="
printf "\n"
OVERALL_STATUS="OK"
# --- 10.1 FAILED SERVICES ---
echo "10.1 FAILED SERVICES"
FAILED_SERVICES_COUNT=$(
systemctl --failed \
--type=service \
--no-legend \
--plain \
2>/dev/null | \
awk 'NF {count++} END {print count+0}'
)
if [ "${FAILED_SERVICES_COUNT:-0}" -gt 0 ]; then
echo "CRITICAL: $FAILED_SERVICES_COUNT failed service(s) detected"
systemctl --failed \
--type=service \
--no-pager \
2>/dev/null
OVERALL_STATUS="CRITICAL"
else
echo "OK: No failed systemd services detected"
fi
printf "\n%.0s" {1..3}
# --- 10.2 FILESYSTEMS ABOVE THRESHOLD ---
echo "10.2 FILESYSTEMS ABOVE THRESHOLD"
DF_OUTPUT=$(timeout 8 df -hT 2>/dev/null)
if [ -n "$DF_OUTPUT" ]; then
FS_OVER_LIMIT=$(
printf "%s\n" "$DF_OUTPUT" | \
awk '
NR > 1 {
use=$6
gsub(/%/, "", use)
if (use ~ /^[0-9]+$/ && use >= 80) {
print
}
}
'
)
if [ -n "$FS_OVER_LIMIT" ]; then
echo "WARNING: Filesystems at or above 80% usage:"
printf "%s\n" "$DF_OUTPUT" | head -1
printf "%s\n" "$FS_OVER_LIMIT"
[ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
else
echo "OK: All filesystems are below 80% usage"
fi
else
FS_OVER_LIMIT=""
echo "WARNING: Filesystem usage query unavailable or timed out"
[ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
fi
printf "\n%.0s" {1..3}
# --- 10.3 INODES ABOVE THRESHOLD ---
echo "10.3 INODES ABOVE THRESHOLD"
DF_INODE_OUTPUT=$(timeout 8 df -ih 2>/dev/null)
if [ -n "$DF_INODE_OUTPUT" ]; then
INODE_OVER_LIMIT=$(
printf "%s\n" "$DF_INODE_OUTPUT" | \
awk '
NR > 1 {
use=$5
gsub(/%/, "", use)
if (use ~ /^[0-9]+$/ && use >= 80) {
print
}
}
'
)
if [ -n "$INODE_OVER_LIMIT" ]; then
echo "WARNING: Filesystems at or above 80% inode usage:"
printf "%s\n" "$DF_INODE_OUTPUT" | head -1
printf "%s\n" "$INODE_OVER_LIMIT"
[ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
else
echo "OK: All filesystems are below 80% inode usage"
fi
else
INODE_OVER_LIMIT=""
echo "WARNING: Inode usage query unavailable or timed out"
[ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
fi
printf "\n%.0s" {1..3}
# --- 10.4 MEMORY PRESSURE ---
echo "10.4 MEMORY PRESSURE"
MEM_PRESS_AVG=""
if [ -f /proc/pressure/memory ]; then
MEM_PRESS_AVG=$(
awk '
$1 == "some" {
for (i=1; i<=NF; i++) {
if ($i ~ /^avg10=/) {
split($i, value, "=")
print value[2]
exit
}
}
}
' /proc/pressure/memory
)
if awk -v value="${MEM_PRESS_AVG:-0}" \
'BEGIN {exit !(value > 10.0)}'
then
echo "WARNING: Memory PSI avg10 is high: ${MEM_PRESS_AVG:-unknown}"
cat /proc/pressure/memory
[ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
else
echo "OK: Memory PSI avg10 is within threshold: ${MEM_PRESS_AVG:-0}"
fi
else
echo "INFO: Memory PSI is not available on this kernel"
fi
printf "\n%.0s" {1..3}
# --- 10.5 SWAP USAGE AND ACTIVITY ---
echo "10.5 SWAP USAGE AND ACTIVITY"
SWAP_TOTAL_METRIC=$(
free -b 2>/dev/null | \
awk '/^Swap:/ {print $2}'
)
SWAP_USED_METRIC=$(
free -b 2>/dev/null | \
awk '/^Swap:/ {print $3}'
)
SWAP_PCT_METRIC=0
SWAP_ACTIVITY=0
if [ "${SWAP_TOTAL_METRIC:-0}" -gt 0 ]; then
SWAP_PCT_METRIC=$(
awk \
-v used="${SWAP_USED_METRIC:-0}" \
-v total="${SWAP_TOTAL_METRIC:-1}" \
'BEGIN {printf "%.0f", (used/total)*100}'
)
SWAP_ACTIVITY=$(
vmstat 1 2 2>/dev/null | \
tail -1 | \
awk '{print $7+$8}'
)
echo "Swap usage : ${SWAP_PCT_METRIC}%"
echo "Swap activity : ${SWAP_ACTIVITY:-0} pages/s combined si+so"
if [ "${SWAP_ACTIVITY:-0}" -gt 0 ]; then
echo "WARNING: Active swap-in or swap-out activity detected"
free -h
[ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
elif [ "${SWAP_PCT_METRIC:-0}" -gt 50 ]; then
echo "INFO: Swap usage is above 50%, but no current swap activity was detected"
else
echo "OK: Swap usage and activity are within expected limits"
fi
else
echo "INFO: Swap is not configured"
fi
printf "\n%.0s" {1..3}
# --- 10.6 SYSTEM LOAD ---
echo "10.6 SYSTEM LOAD"
read -r LOAD_ONE_MIN LOAD_FIVE_MIN LOAD_FIFTEEN_MIN _ < /proc/loadavg
CPU_CORES_COUNT=$(
nproc 2>/dev/null || \
grep -c '^processor' /proc/cpuinfo 2>/dev/null || \
echo 1
)
LOAD_PER_CORE=$(
awk \
-v load="${LOAD_ONE_MIN:-0}" \
-v cores="${CPU_CORES_COUNT:-1}" \
'BEGIN {
if (cores < 1) {
cores=1
}
printf "%.2f", load/cores
}'
)
echo "Load average : ${LOAD_ONE_MIN:-0} ${LOAD_FIVE_MIN:-0} ${LOAD_FIFTEEN_MIN:-0}"
echo "CPU cores : ${CPU_CORES_COUNT:-1}"
echo "Load/core : $LOAD_PER_CORE"
if awk -v value="$LOAD_PER_CORE" \
'BEGIN {exit !(value >= 1.0)}'
then
echo "CRITICAL: One-minute load is at or above 1.00 per CPU core"
uptime
OVERALL_STATUS="CRITICAL"
elif awk -v value="$LOAD_PER_CORE" \
'BEGIN {exit !(value >= 0.70)}'
then
echo "WARNING: One-minute load is at or above 0.70 per CPU core"
uptime
[ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
else
echo "OK: Normalized load is within threshold"
fi
printf "\n%.0s" {1..3}
# --- 10.7 DISK AND FILESYSTEM ERRORS ---
echo "10.7 DISK AND FILESYSTEM ERRORS"
if command -v journalctl >/dev/null 2>&1; then
DISK_ERR_HUNT=$(
journalctl \
-k \
-b 0 \
--no-pager \
2>/dev/null | \
grep -Ei \
'I/O error|buffer I/O|blk_update_request|critical medium error|uncorrectable|EXT4-fs error|XFS.*error|BTRFS.*error|ZFS.*error|nvme.*error|bad sector|read-only file system' | \
tail -15
)
else
DISK_ERR_HUNT=$(
dmesg 2>/dev/null | \
grep -Ei \
'I/O error|buffer I/O|blk_update_request|critical medium error|uncorrectable|EXT4-fs error|XFS.*error|BTRFS.*error|ZFS.*error|nvme.*error|bad sector|read-only file system' | \
tail -15
)
fi
if [ -n "$DISK_ERR_HUNT" ]; then
echo "CRITICAL: Disk or filesystem errors detected:"
printf "%s\n" "$DISK_ERR_HUNT"
OVERALL_STATUS="CRITICAL"
else
echo "OK: No disk or filesystem errors detected in the current boot"
fi
printf "\n%.0s" {1..3}
# --- 10.8 NETWORK ERRORS ---
echo "10.8 NETWORK ERRORS"
if command -v journalctl >/dev/null 2>&1; then
NET_ERR_HUNT=$(
journalctl \
-k \
-b 0 \
--no-pager \
2>/dev/null | \
grep -Ei \
'link is down|link down|lost carrier|carrier lost|NETDEV WATCHDOG|transmit queue timed out|tx timeout|reset adapter|firmware failed' | \
tail -15
)
else
NET_ERR_HUNT=$(
dmesg 2>/dev/null | \
grep -Ei \
'link is down|link down|lost carrier|carrier lost|NETDEV WATCHDOG|transmit queue timed out|tx timeout|reset adapter|firmware failed' | \
tail -15
)
fi
if [ -n "$NET_ERR_HUNT" ]; then
echo "WARNING: Network link or adapter errors detected:"
printf "%s\n" "$NET_ERR_HUNT"
[ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
else
echo "OK: No network link or adapter errors detected in the current boot"
fi
printf "\n%.0s" {1..3}
# --- 10.9 SMART HEALTH ---
echo "10.9 SMART HEALTH"
SMART_ERR_FLAG=0
SMART_UNKNOWN_FLAG=0
SMART_DEVICE_COUNT=0
if command -v smartctl >/dev/null 2>&1; then
while IFS= read -r DEV_ID; do
[ -z "$DEV_ID" ] && continue
SMART_DEVICE_COUNT=$((SMART_DEVICE_COUNT + 1))
SMART_OUTPUT=$(
smartctl -H "$DEV_ID" 2>/dev/null
)
if printf "%s\n" "$SMART_OUTPUT" | \
grep -Eqi 'PASSED|SMART Health Status: OK|overall-health.*PASSED'
then
echo "OK: SMART health passed on $DEV_ID"
elif printf "%s\n" "$SMART_OUTPUT" | \
grep -Eqi 'FAILED|BAD|CRITICAL'
then
echo "CRITICAL: SMART health failure reported on $DEV_ID"
printf "%s\n" "$SMART_OUTPUT"
SMART_ERR_FLAG=1
OVERALL_STATUS="CRITICAL"
else
echo "INFO: SMART health could not be determined for $DEV_ID"
SMART_UNKNOWN_FLAG=1
fi
done < <(
smartctl --scan 2>/dev/null | \
awk '{print $1}'
)
if [ "$SMART_DEVICE_COUNT" -eq 0 ]; then
echo "INFO: No SMART-capable devices discovered by smartctl --scan"
elif [ "$SMART_ERR_FLAG" -eq 0 ] && [ "$SMART_UNKNOWN_FLAG" -eq 0 ]; then
echo "OK: All discovered SMART devices reported healthy"
fi
else
echo "INFO: smartctl is not installed"
fi
printf "\n%.0s" {1..3}
# --- 10.10 ZFS HEALTH ---
echo "10.10 ZFS HEALTH"
ZFS_HEALTH_STATUS="NOT INSTALLED"
if command -v zpool >/dev/null 2>&1; then
ZPOOL_FAULT_MAP=$(
zpool status -x 2>/dev/null
)
if [ -z "$ZPOOL_FAULT_MAP" ]; then
echo "INFO: No ZFS pools detected"
ZFS_HEALTH_STATUS="NO POOLS"
elif printf "%s\n" "$ZPOOL_FAULT_MAP" | \
grep -qi 'all pools are healthy'
then
echo "OK: All ZFS pools are healthy"
ZFS_HEALTH_STATUS="OK"
else
echo "CRITICAL: ZFS pool issue detected:"
printf "%s\n" "$ZPOOL_FAULT_MAP"
ZFS_HEALTH_STATUS="CRITICAL"
OVERALL_STATUS="CRITICAL"
fi
else
echo "INFO: ZFS tools are not installed"
fi
printf "\n%.0s" {1..3}
# --- 10.11 MDADM RAID HEALTH ---
echo "10.11 MDADM RAID HEALTH"
MDADM_HEALTH_STATUS="NOT CONFIGURED"
if [ -r /proc/mdstat ] && grep -q '^md[0-9]' /proc/mdstat; then
MDADM_HEALTH_STATUS="OK"
MDADM_PROBLEMS=$(
grep -E \
'inactive|recovery|resync|reshape|check|\[[U_]*_[U_]*\]' \
/proc/mdstat \
2>/dev/null
)
if [ -n "$MDADM_PROBLEMS" ]; then
echo "WARNING: mdadm array is degraded, inactive, or running maintenance:"
cat /proc/mdstat
MDADM_HEALTH_STATUS="WARNING"
[ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
else
echo "OK: Active mdadm arrays appear healthy"
fi
else
echo "INFO: No active mdadm arrays detected"
fi
printf "\n%.0s" {1..3}
# --- 10.12 TIME SYNCHRONIZATION ---
echo "10.12 TIME SYNCHRONIZATION"
TIME_SYNC_STATUS="UNKNOWN"
if command -v timedatectl >/dev/null 2>&1; then
TIME_SYNC_STATUS=$(
timedatectl show \
-p NTPSynchronized \
--value \
2>/dev/null
)
if [ "$TIME_SYNC_STATUS" = "yes" ]; then
echo "OK: System clock is synchronized"
TIME_SYNC_STATUS="OK"
else
echo "WARNING: System clock is not synchronized"
timedatectl status 2>/dev/null
TIME_SYNC_STATUS="WARNING"
[ "$OVERALL_STATUS" = "OK" ] && OVERALL_STATUS="WARNING"
fi
else
echo "INFO: timedatectl is not available"
fi
printf "\n%.0s" {1..3}
# --- 10.13 SUMMARY ---
echo "10.13 SUMMARY"
echo "--- Core Global Infrastructure Health Matrix ---"
printf "Platform Node Instance...%s\n" \
"$(hostname 2>/dev/null || echo unresolved)"
if [ "${FAILED_SERVICES_COUNT:-0}" -eq 0 ]; then
echo "Systemd Services..........OK"
else
echo "Systemd Services..........CRITICAL ($FAILED_SERVICES_COUNT failed)"
fi
if [ -z "$FS_OVER_LIMIT" ]; then
echo "Filesystem Capacity.......OK"
else
echo "Filesystem Capacity.......WARNING"
fi
if [ -z "$INODE_OVER_LIMIT" ]; then
echo "Inode Capacity............OK"
else
echo "Inode Capacity............WARNING"
fi
if [ -z "$DISK_ERR_HUNT" ]; then
echo "Kernel Disk Health........OK"
else
echo "Kernel Disk Health........CRITICAL"
fi
if [ -z "$NET_ERR_HUNT" ]; then
echo "Network Link Health.......OK"
else
echo "Network Link Health.......WARNING"
fi
echo "ZFS Health................$ZFS_HEALTH_STATUS"
echo "mdadm Health..............$MDADM_HEALTH_STATUS"
echo "Time Synchronization......$TIME_SYNC_STATUS"
echo "Normalized Load/Core......$LOAD_PER_CORE"
echo "Overall Health............$OVERALL_STATUS"
