Cache systemctl list-unit-files to avoid repeated calls

Add get_unit_files() and unit_file_exists() helper functions
that cache the output of systemctl list-unit-files. This
avoids running the same command multiple times when checking
for node_exporter and hwmon.timer unit files.

#6

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

proxDoc.sh

@@ -2,6 +2,37 @@
 
 VERSION="1.1.0"
 
+###################
+# Timeout Configuration
+###################
+readonly CMD_TIMEOUT=30  # Default timeout in seconds for external commands
+
+###################
+# Logging Configuration
+###################
+# Optional log file - set via environment variable PROXDOC_LOGFILE
+LOGFILE="${PROXDOC_LOGFILE:-}"
+
+###################
+# Cached Data
+###################
+# Disk list cache - populated on first use
+DISK_LIST=""
+# Unit files cache - populated on first use
+UNIT_FILES=""
+
+###################
+# Pattern Constants
+###################
+# Virtual/firewall interface patterns to skip
+readonly VIRTUAL_IFACE_PATTERN="^(veth|fwbr|fwln|fwpr|tap)"
+# Storage controller patterns for HBA detection
+readonly STORAGE_CONTROLLER_PATTERN="RAID|SAS|SATA|SCSI|Mass storage|Serial Attached|Fibre Channel|NVMe"
+# Disk device patterns
+readonly DISK_DEVICE_PATTERN="^sd|^nvme"
+# PCI devices to exclude from hardware info
+readonly EXCLUDED_PCI_PATTERN="Host bridge|PCI bridge|ISA bridge|SMBus|IOMMU|Dummy|USB controller|Audio device|Encryption controller|Multimedia controller"
+
 ###################
 # Color Definitions
 ###################
@@ -31,18 +62,42 @@ print_header() {
 }
 
 
+# Error handling flags
+ERRORS_OCCURRED=0
+WARNINGS_OCCURRED=0
+
+cleanup() {
+    # Cleanup function called on exit
+    local exit_code=$?
+    if [[ $exit_code -ne 0 ]]; then
+        echo -e "\n${RED}Script terminated with exit code: $exit_code${NC}"
+    fi
+    # Add any cleanup tasks here (temp files, etc.)
+}
+
 handle_error() {
-    echo -e "${RED}Error: $1${NC}"
+    local message="$1"
-    exit 1
+    local fatal="${2:-true}"  # Default to fatal error
+    echo -e "${RED}Error: $message${NC}" >&2
+    ERRORS_OCCURRED=$((ERRORS_OCCURRED + 1))
+    if [[ "$fatal" == "true" ]]; then
+        exit 1
+    fi
 }
 
 log_message() {
-    local level=$1
+    local level="$1"
-    local message=$2
+    local message="$2"
-    case $level in
+    case "$level" in
         info)    echo -e "${GREEN}[INFO]${NC} $message" ;;
-        warn)    echo -e "${YELLOW}[WARN]${NC} $message" ;;
+        warn)
-        error)   echo -e "${RED}[ERROR]${NC} $message" ;;
+            echo -e "${YELLOW}[WARN]${NC} $message"
+            WARNINGS_OCCURRED=$((WARNINGS_OCCURRED + 1))
+            ;;
+        error)
+            echo -e "${RED}[ERROR]${NC} $message" >&2
+            ERRORS_OCCURRED=$((ERRORS_OCCURRED + 1))
+            ;;
     esac
 }
 
@@ -67,6 +122,46 @@ checkIfOnHypervisor() {
     command -v pveversion >/dev/null 2>&1
 }
 
+# Get disk list with caching to avoid multiple lsblk calls
+get_disk_list() {
+    if [[ -z "$DISK_LIST" ]]; then
+        DISK_LIST=$(lsblk -d -o name 2>/dev/null | grep -E "$DISK_DEVICE_PATTERN")
+    fi
+    echo "$DISK_LIST"
+}
+
+# Get systemctl unit files with caching
+get_unit_files() {
+    if [[ -z "$UNIT_FILES" ]]; then
+        UNIT_FILES=$(systemctl list-unit-files 2>/dev/null)
+    fi
+    echo "$UNIT_FILES"
+}
+
+# Check if a unit file exists (uses cached data)
+unit_file_exists() {
+    local unit_name="$1"
+    get_unit_files | grep -q "$unit_name"
+}
+
+# Efficient process wait with optional spinner
+# Usage: wait_for_process $pid [delay]
+# Uses kill -0 instead of ps -p for efficiency
+wait_for_process() {
+    local pid="$1"
+    local delay="${2:-0.1}"
+    local spinner='|/-\'
+    local i=0
+
+    while kill -0 "$pid" 2>/dev/null; do
+        printf "\r%c " "${spinner:i++%${#spinner}:1}"
+        sleep "$delay"
+    done
+    printf "\r  \r"  # Clear spinner
+    wait "$pid"
+    return $?
+}
+
 ###################
 # System Information Functions
 ###################
@@ -102,35 +197,40 @@ get_temp_info() {
 get_disk_health() {
     echo -e "\n${GREEN}=== Disk Health Status ===${NC}"
     if command -v smartctl >/dev/null 2>&1; then
-        for disk in $(lsblk -d -o name | grep -E '^sd|^nvme'); do
+        while IFS= read -r disk; do
+            [[ -z "$disk" ]] && continue
             echo -e "\nChecking /dev/$disk:"
-            smartctl -H /dev/$disk
+            if ! timeout $CMD_TIMEOUT smartctl -H "/dev/$disk"; then
-        done
+                log_message warn "smartctl timed out or failed for /dev/$disk"
+            fi
+        done <<< "$(get_disk_list)"
     else
         log_message warn "smartctl not found. Install smartmontools for disk health monitoring"
     fi
 }
 
 get_cpu_info() {
-    cpu_info=$(grep -m 1 -w 'model name' /proc/cpuinfo | awk -F: '{print $2}' | xargs) || {
+    local cpu_info cpu_cores cpu_mhz
-        echo -e "${RED}Failed to retrieve CPU model information.${NC}"
-    }
-    cpu_cores=$(lscpu | grep '^CPU(s):' | awk '{print $2}')
-    cpu_mhz=$(lscpu | grep 'MHz' | awk '{print $4}')
 
-    echo -e "${GREEN}CPU Model:${NC} $cpu_info"
+    cpu_info=$(grep -m 1 -w 'model name' /proc/cpuinfo 2>/dev/null | awk -F: '{print $2}' | xargs)
-    echo -e "${GREEN}CPU Cores:${NC} $cpu_cores"
+    cpu_cores=$(lscpu 2>/dev/null | grep '^CPU(s):' | awk '{print $2}')
-    echo -e "${GREEN}CPU MHz:${NC} $cpu_mhz"
+    cpu_mhz=$(lscpu 2>/dev/null | grep 'MHz' | awk '{print $4}')
+
+    echo -e "${GREEN}CPU Model:${NC} ${cpu_info:-Unknown}"
+    echo -e "${GREEN}CPU Cores:${NC} ${cpu_cores:-Unknown}"
+    echo -e "${GREEN}CPU MHz:${NC} ${cpu_mhz:-Unknown}"
 }
 
 get_ram_info() {
-    ram_total=$(free -h | grep 'Mem:' | awk '{print $2}')
+    local ram_total ram_used ram_free
-    ram_used=$(free -h | grep 'Mem:' | awk '{print $3}')
-    ram_free=$(free -h | grep 'Mem:' | awk '{print $4}')
 
-    echo -e "${GREEN}Total RAM:${NC} $ram_total"
+    ram_total=$(free -h 2>/dev/null | grep 'Mem:' | awk '{print $2}')
-    echo -e "${GREEN}Used RAM:${NC} $ram_used"
+    ram_used=$(free -h 2>/dev/null | grep 'Mem:' | awk '{print $3}')
-    echo -e "${GREEN}Free RAM:${NC} $ram_free"
+    ram_free=$(free -h 2>/dev/null | grep 'Mem:' | awk '{print $4}')
+
+    echo -e "${GREEN}Total RAM:${NC} ${ram_total:-Unknown}"
+    echo -e "${GREEN}Used RAM:${NC} ${ram_used:-Unknown}"
+    echo -e "${GREEN}Free RAM:${NC} ${ram_free:-Unknown}"
 }
 
 get_storage_info() {
@@ -144,10 +244,13 @@ get_storage_info() {
 }
 
 get_network_info() {
-    default_gateway=$(ip route | grep default | awk '{print $3}')
+    local default_gateway ip_addresses
-    ip_addresses=$(hostname -I | xargs)
+
-    echo -e "${GREEN}Default Gateway:${NC} $default_gateway"
+    default_gateway=$(ip route 2>/dev/null | grep default | awk '{print $3}')
-    echo -e "${GREEN}IP Addresses:${NC} $ip_addresses"
+    ip_addresses=$(hostname -I 2>/dev/null | xargs)
+
+    echo -e "${GREEN}Default Gateway:${NC} ${default_gateway:-Not configured}"
+    echo -e "${GREEN}IP Addresses:${NC} ${ip_addresses:-None detected}"
 }
 
 get_detailed_network() {
@@ -173,7 +276,7 @@ get_hardware_info() {
     echo -e "${GREEN}BIOS Version:${NC} $(dmidecode -s bios-version)"
     echo -e "\n${GREEN}=== PCI Devices ===${NC}"
     # Show interesting devices, exclude bridges, infrastructure, and integrated motherboard devices
-    lspci | grep -v -E "Host bridge|PCI bridge|ISA bridge|SMBus|IOMMU|Dummy|USB controller|Audio device|Encryption controller|Multimedia controller"
+    lspci | grep -v -E "$EXCLUDED_PCI_PATTERN"
 }
 
 get_motherboard_info() {
@@ -350,7 +453,7 @@ get_physical_interfaces() {
         [[ "$iface" == "lo" ]] && continue
         
         # Skip virtual/firewall interfaces
-        [[ "$iface" =~ ^(veth|fwbr|fwln|fwpr|tap) ]] && continue
+        [[ "$iface" =~ $VIRTUAL_IFACE_PATTERN ]] && continue
         
         # This is a physical interface
         echo "$iface"
@@ -361,9 +464,9 @@ get_hba_info() {
     echo -e "\n${GREEN}=== HBA/Storage Controller Information ===${NC}"
 
     # Find RAID, SAS, SATA, SCSI, and storage controllers
-    lspci -vmm 2>/dev/null | awk '
+    lspci -vmm 2>/dev/null | awk -v pattern="$STORAGE_CONTROLLER_PATTERN" '
         BEGIN { RS=""; FS="\n" }
-        /RAID|SAS|SATA|SCSI|Mass storage|Serial Attached|Fibre Channel|NVMe/ {
+        $0 ~ pattern {
             for (i=1; i<=NF; i++) {
                 if ($i ~ /^Slot:/) slot = substr($i, 7)
                 if ($i ~ /^Class:/) class = substr($i, 8)
@@ -382,7 +485,7 @@ get_hba_info() {
 
     # Show detailed info for storage controllers
     echo -e "\n${GREEN}=== Storage Controller Details ===${NC}"
-    for ctrl in $(lspci | grep -iE "RAID|SAS|SATA|SCSI|Mass storage|NVMe" | awk '{print $1}'); do
+    for ctrl in $(lspci | grep -iE "$STORAGE_CONTROLLER_PATTERN" | awk '{print $1}'); do
         echo -e "\n${GREEN}Controller $ctrl:${NC}"
         lspci -vvs "$ctrl" 2>/dev/null | grep -E "^\s+(Subsystem|LnkSta|Kernel driver)" | head -5
     done
@@ -418,16 +521,16 @@ get_ceph_health() {
     echo -e "\n${GREEN}=== Ceph Cluster Health ===${NC}"
     if command -v ceph >/dev/null 2>&1; then
         echo -e "${GREEN}Health Status:${NC}"
-        ceph health detail 2>/dev/null || log_message warn "Cannot connect to Ceph cluster"
+        timeout $CMD_TIMEOUT ceph health detail 2>/dev/null || log_message warn "Cannot connect to Ceph cluster or timed out"
 
         echo -e "\n${GREEN}=== Ceph OSD Tree ===${NC}"
-        ceph osd tree 2>/dev/null || true
+        timeout $CMD_TIMEOUT ceph osd tree 2>/dev/null || log_message warn "Ceph OSD tree timed out"
 
         echo -e "\n${GREEN}=== Ceph Pool Usage ===${NC}"
-        ceph df 2>/dev/null || true
+        timeout $CMD_TIMEOUT ceph df 2>/dev/null || log_message warn "Ceph df timed out"
 
         echo -e "\n${GREEN}=== Ceph OSD Usage ===${NC}"
-        ceph osd df 2>/dev/null || true
+        timeout $CMD_TIMEOUT ceph osd df 2>/dev/null || log_message warn "Ceph OSD df timed out"
     else
         log_message info "Ceph tools not installed on this node"
     fi
@@ -444,7 +547,7 @@ get_node_exporter_status() {
         else
             log_message warn "Port 9100 not listening"
         fi
-    elif systemctl list-unit-files 2>/dev/null | grep -q node_exporter; then
+    elif unit_file_exists node_exporter; then
         log_message warn "Node Exporter is installed but not running"
         echo -e "Start with: systemctl start node_exporter"
     else
@@ -459,7 +562,7 @@ get_hwmon_status() {
         systemctl list-timers hwmon.timer --no-pager 2>/dev/null
         echo -e "\n${GREEN}Last Run:${NC}"
         journalctl -u hwmon.service -n 3 --no-pager 2>/dev/null || true
-    elif systemctl list-unit-files 2>/dev/null | grep -q hwmon.timer; then
+    elif unit_file_exists hwmon.timer; then
         log_message warn "hwmon timer is installed but not active"
         echo -e "Enable with: systemctl enable --now hwmon.timer"
     else
@@ -467,6 +570,51 @@ get_hwmon_status() {
     fi
 }
 
+# Valid check names for selective mode
+readonly VALID_CHECKS="cpu ram memory storage disk network hardware temps services ceph vms containers"
+
+run_selective_checks() {
+    local checks="$1"
+    if [[ -z "$checks" ]]; then
+        log_message error "No checks specified. Use --checks=cpu,ram,disk"
+        echo "Valid checks: $VALID_CHECKS"
+        exit 1
+    fi
+
+    # Validate check names
+    IFS=',' read -ra check_array <<< "$checks"
+    for check in "${check_array[@]}"; do
+        if [[ ! " $VALID_CHECKS " =~ " $check " ]]; then
+            log_message error "Unknown check: $check"
+            echo "Valid checks: $VALID_CHECKS"
+            exit 1
+        fi
+    done
+
+    log_message info "Running selective checks: $checks"
+    echo ""
+
+    for check in "${check_array[@]}"; do
+        case "$check" in
+            cpu)        log_message info "Checking CPU..."; get_cpu_info ;;
+            ram)        log_message info "Checking RAM..."; get_ram_info ;;
+            memory)     log_message info "Checking memory details..."; get_memory_details ;;
+            storage)    log_message info "Checking storage..."; get_storage_info ;;
+            disk)       log_message info "Checking disk health..."; get_disk_health ;;
+            network)    log_message info "Checking network..."; get_network_info; get_detailed_network; get_nic_details ;;
+            hardware)   log_message info "Checking hardware..."; get_hardware_info; get_motherboard_info; get_hba_info ;;
+            temps)      log_message info "Checking temperatures..."; get_temp_info ;;
+            services)   log_message info "Checking services..."; check_services ;;
+            ceph)       log_message info "Checking Ceph..."; get_ceph_health ;;
+            vms)        log_message info "Checking VMs..."; list_vms ;;
+            containers) log_message info "Checking containers..."; list_containers ;;
+        esac
+    done
+
+    echo ""
+    log_message info "Selective checks complete"
+}
+
 quick_health_check() {
     echo -e "\n${GREEN}=== Quick Health Check ===${NC}"
     echo -e "Running quick health assessment...\n"
@@ -480,12 +628,15 @@ quick_health_check() {
     # Disk health (quick)
     echo -e "\n${GREEN}=== Disk Health Summary ===${NC}"
     if command -v smartctl >/dev/null 2>&1; then
-        for disk in $(lsblk -d -o name | grep -E '^sd|^nvme'); do
+        while IFS= read -r disk; do
-            health=$(smartctl -H /dev/$disk 2>/dev/null | grep -i "health" | awk -F: '{print $2}' | xargs)
+            [[ -z "$disk" ]] && continue
+            health=$(timeout $CMD_TIMEOUT smartctl -H "/dev/$disk" 2>/dev/null | grep -i "health" | awk -F: '{print $2}' | xargs)
             if [[ -n "$health" ]]; then
                 echo -e "/dev/$disk: $health"
+            else
+                echo -e "/dev/$disk: ${YELLOW}check timed out or unavailable${NC}"
             fi
-        done
+        done <<< "$(get_disk_list)"
     fi
 
     # Node Exporter
@@ -574,6 +725,31 @@ help() {
     echo "  --vm-list        Check VM vitals"
     echo "  --ct-list        Check container vitals"
     echo "  --backup         Review backup health"
+    echo "  --checks=LIST    Run only specific checks (comma-separated)"
+    echo ""
+    echo "Valid checks for --checks option:"
+    echo "  cpu, ram, memory, storage, disk, network, hardware, temps,"
+    echo "  services, ceph, vms, containers"
+    echo ""
+    echo "Examples:"
+    echo "  Run full diagnostics:"
+    echo "    curl -sL \"http://10.10.10.63:3000/LotusGuild/proxDoc/raw/branch/main/proxDoc.sh\" | bash -s -- --diags"
+    echo ""
+    echo "  Quick health check:"
+    echo "    curl -sL \"http://10.10.10.63:3000/LotusGuild/proxDoc/raw/branch/main/proxDoc.sh\" | bash -s -- --quick"
+    echo ""
+    echo "  Check only services and VMs:"
+    echo "    curl -sL \"http://10.10.10.63:3000/LotusGuild/proxDoc/raw/branch/main/proxDoc.sh\" | bash -s -- --services"
+    echo "    curl -sL \"http://10.10.10.63:3000/LotusGuild/proxDoc/raw/branch/main/proxDoc.sh\" | bash -s -- --vm-list"
+    echo ""
+    echo "  View drive bay mapping:"
+    echo "    curl -sL \"http://10.10.10.63:3000/LotusGuild/proxDoc/raw/branch/main/proxDoc.sh\" | bash -s -- --drives"
+    echo ""
+    echo "  Check Ceph cluster health:"
+    echo "    curl -sL \"http://10.10.10.63:3000/LotusGuild/proxDoc/raw/branch/main/proxDoc.sh\" | bash -s -- --ceph"
+    echo ""
+    echo "  Run only CPU and RAM checks:"
+    echo "    curl -sL \"http://10.10.10.63:3000/LotusGuild/proxDoc/raw/branch/main/proxDoc.sh\" | bash -s -- --checks=cpu,ram"
     exit 0
 }
 
@@ -583,7 +759,28 @@ help() {
 ###################
 # Main Functions
 ###################
+
+# Setup logging if LOGFILE is specified
+setup_logging() {
+    if [[ -n "$LOGFILE" ]]; then
+        # Create log directory if needed
+        local log_dir
+        log_dir=$(dirname "$LOGFILE")
+        if [[ ! -d "$log_dir" ]]; then
+            mkdir -p "$log_dir" 2>/dev/null || {
+                log_message warn "Cannot create log directory: $log_dir"
+                LOGFILE=""
+                return
+            }
+        fi
+        log_message info "Logging output to: $LOGFILE"
+        # Redirect stdout and stderr to tee (no subshell overhead)
+        exec > >(tee -a "$LOGFILE") 2>&1
+    fi
+}
+
 runDiags() {
+    setup_logging
     log_message info "Beginning system examination..."
     
     # Check if running on Proxmox
@@ -649,10 +846,49 @@ runDiags() {
     
     echo ""
     log_message info "Examination complete"
+
+    # Print summary if there were issues
+    if [[ $WARNINGS_OCCURRED -gt 0 || $ERRORS_OCCURRED -gt 0 ]]; then
+        echo -e "\n${YELLOW}=== Summary ===${NC}"
+        [[ $WARNINGS_OCCURRED -gt 0 ]] && echo -e "Warnings: $WARNINGS_OCCURRED"
+        [[ $ERRORS_OCCURRED -gt 0 ]] && echo -e "Errors: $ERRORS_OCCURRED"
+    fi
+}
+
+# Whitelist of valid command options
+readonly VALID_OPTIONS="--help --diags --quick --drives --ceph --node-exporter --hwmon --services --vm-list --ct-list --backup --checks"
+
+validate_input() {
+    local input="$1"
+    # Check if input matches valid option pattern (starts with -- and contains only alphanumeric, hyphens, equals, commas)
+    if [[ ! "$input" =~ ^--[a-z][-a-z=,]*$ ]]; then
+        return 1
+    fi
+    # Extract the option name (before any = sign)
+    local opt_name="${input%%=*}"
+    # Check against whitelist
+    if [[ ! " $VALID_OPTIONS " =~ " $opt_name " ]]; then
+        return 1
+    fi
+    return 0
 }
 
 checkForInput() {
-    case $1 in
+    local input="$1"
+
+    # Validate input against whitelist
+    if ! validate_input "$input"; then
+        echo -e "${RED}Invalid option: $input${NC}"
+        echo -e "Use --help to see available options."
+        exit 1
+    fi
+
+    # Extract option name and value for --checks=X pattern
+    local opt_name="${input%%=*}"
+    local opt_value="${input#*=}"
+    [[ "$opt_name" == "$opt_value" ]] && opt_value=""
+
+    case "$opt_name" in
         --help)          help ;;
         --diags)         check_requirements; runDiags ;;
         --quick)         quick_health_check ;;
@@ -664,7 +900,7 @@ checkForInput() {
         --vm-list)       list_vms ;;
         --ct-list)       list_containers ;;
         --backup)        echo -e "${GREEN}Backup Status:${NC}"; pvesm status 2>/dev/null || log_message warn "pvesm not available" ;;
-        *)               echo -e "${RED}Invalid option: $1${NC}"; help ;;
+        --checks)        run_selective_checks "$opt_value" ;;
     esac
 }
 
@@ -682,8 +918,10 @@ if [[ $EUID -ne 0 ]]; then
     handle_error "This script must be run as root"
 fi
 
-# Set trap for interrupts
+# Set trap for cleanup and interrupts
-trap 'echo -e "${RED}Script interrupted.${NC}"; exit 1' INT TERM
+trap cleanup EXIT
+trap 'echo -e "\n${RED}Script interrupted by user.${NC}"; exit 130' INT
+trap 'echo -e "\n${RED}Script terminated.${NC}"; exit 143' TERM
 
 if [[ -n $argOne ]]; then
     checkForInput "$argOne"