Homelab-manager/agent/linux/report-tasks.sh

#!/usr/bin/env bash
# Collects cron jobs and systemd timers on this host and POSTs them to the
# Homelab Manager API. Intended to run as root via a periodic systemd timer
# (see install.sh) but can be run manually for testing:
#
#   API_URL=https://homelab.example.lan API_TOKEN=hlm_xxx ./report-tasks.sh --dry-run
#
set -euo pipefail

ENV_FILE="${ENV_FILE:-/etc/homelab-manager-agent.env}"
if [[ -f "$ENV_FILE" ]]; then
  # shellcheck disable=SC1090
  source "$ENV_FILE"
fi

API_URL="${API_URL:-}"
API_TOKEN="${API_TOKEN:-}"
DRY_RUN=0
[[ "${1:-}" == "--dry-run" ]] && DRY_RUN=1

if [[ -z "$API_URL" || -z "$API_TOKEN" ]]; then
  echo "API_URL and API_TOKEN must be set (env vars or $ENV_FILE)" >&2
  exit 1
fi

suggest_package_install() {
  local pkg="$1"
  if command -v apt-get >/dev/null 2>&1; then
    echo "  sudo apt-get update && sudo apt-get install -y $pkg"
  elif command -v dnf >/dev/null 2>&1; then
    echo "  sudo dnf install -y $pkg"
  elif command -v yum >/dev/null 2>&1; then
    echo "  sudo yum install -y $pkg"
  elif command -v apk >/dev/null 2>&1; then
    echo "  sudo apk add $pkg"
  elif command -v pacman >/dev/null 2>&1; then
    echo "  sudo pacman -S $pkg"
  elif command -v zypper >/dev/null 2>&1; then
    echo "  sudo zypper install -y $pkg"
  fi
}

for bin in curl jq; do
  if ! command -v "$bin" >/dev/null 2>&1; then
    echo "Required dependency '$bin' is not installed. Try:" >&2
    suggest_package_install "$bin" >&2
    exit 1
  fi
done

TASKS_JSON="[]"

add_task() {
  local schedule_type="$1" name="$2" command="$3" schedule_expression="$4" source="$5" enabled="$6"
  TASKS_JSON=$(jq -c \
    --arg schedule_type "$schedule_type" \
    --arg name "$name" \
    --arg command "$command" \
    --arg schedule_expression "$schedule_expression" \
    --arg source "$source" \
    --argjson enabled "$enabled" \
    '. + [{
      schedule_type: $schedule_type,
      name: $name,
      command: $command,
      schedule_expression: $schedule_expression,
      source: $source,
      enabled: $enabled
    }]' <<<"$TASKS_JSON")
}

add_task_with_next_run() {
  local schedule_type="$1" name="$2" command="$3" schedule_expression="$4" source="$5" enabled="$6" next_run_at="$7"
  TASKS_JSON=$(jq -c \
    --arg schedule_type "$schedule_type" \
    --arg name "$name" \
    --arg command "$command" \
    --arg schedule_expression "$schedule_expression" \
    --arg source "$source" \
    --argjson enabled "$enabled" \
    --arg next_run_at "$next_run_at" \
    '. + [{
      schedule_type: $schedule_type,
      name: $name,
      command: $command,
      schedule_expression: $schedule_expression,
      source: $source,
      enabled: $enabled,
      next_run_at: $next_run_at
    }]' <<<"$TASKS_JSON")
}

parse_crontab_lines() {
  # Reads 5-field-schedule + command cron lines from stdin.
  # $1 = source label, $2 = "system" (7-field, includes a user column) or "user" (6-field)
  local source="$1" mode="$2"
  while IFS= read -r line; do
    line="${line%%$'\r'}"
    [[ -z "$line" ]] && continue
    [[ "$line" =~ ^[[:space:]]*# ]] && continue
    [[ "$line" =~ ^[[:space:]]*[A-Za-z_][A-Za-z0-9_]*= ]] && continue

    if [[ "$mode" == "system" ]]; then
      # min hour dom mon dow user command...
      if [[ "$line" =~ ^[[:space:]]*([^[:space:]]+)[[:space:]]+([^[:space:]]+)[[:space:]]+([^[:space:]]+)[[:space:]]+([^[:space:]]+)[[:space:]]+([^[:space:]]+)[[:space:]]+[^[:space:]]+[[:space:]]+(.+)$ ]]; then
        local sched="${BASH_REMATCH[1]} ${BASH_REMATCH[2]} ${BASH_REMATCH[3]} ${BASH_REMATCH[4]} ${BASH_REMATCH[5]}"
        local cmd="${BASH_REMATCH[6]}"
        add_task "cron" "$cmd" "$cmd" "$sched" "$source" true
      fi
    else
      # min hour dom mon dow command...
      if [[ "$line" =~ ^[[:space:]]*([^[:space:]]+)[[:space:]]+([^[:space:]]+)[[:space:]]+([^[:space:]]+)[[:space:]]+([^[:space:]]+)[[:space:]]+([^[:space:]]+)[[:space:]]+(.+)$ ]]; then
        local sched="${BASH_REMATCH[1]} ${BASH_REMATCH[2]} ${BASH_REMATCH[3]} ${BASH_REMATCH[4]} ${BASH_REMATCH[5]}"
        local cmd="${BASH_REMATCH[6]}"
        add_task "cron" "$cmd" "$cmd" "$sched" "$source" true
      fi
    fi
  done
}

collect_cron() {
  [[ -r /etc/crontab ]] && parse_crontab_lines "/etc/crontab" "system" < /etc/crontab

  if [[ -d /etc/cron.d ]]; then
    for f in /etc/cron.d/*; do
      [[ -f "$f" && -r "$f" ]] || continue
      parse_crontab_lines "$f" "system" < "$f"
    done
  fi

  if command -v crontab >/dev/null 2>&1 && [[ -r /etc/passwd ]]; then
    while IFS=: read -r user _ uid _ _ _ shell; do
      case "$shell" in
        */nologin|*/false|"") continue ;;
      esac
      [[ "$uid" -lt 1000 && "$uid" != "0" ]] && continue
      local_crontab=$(crontab -l -u "$user" 2>/dev/null || true)
      [[ -z "$local_crontab" ]] && continue
      parse_crontab_lines "crontab:$user" "user" <<<"$local_crontab"
    done < /etc/passwd
  fi
}

collect_systemd_timers() {
  command -v systemctl >/dev/null 2>&1 || return 0

  local timers_json
  timers_json=$(systemctl list-timers --all --output=json 2>/dev/null || echo "[]")

  while IFS= read -r entry; do
    local unit activates next_usec enabled_state schedule_expr next_run_at
    unit=$(jq -r '.unit' <<<"$entry")
    activates=$(jq -r '.activates // ""' <<<"$entry")
    next_usec=$(jq -r '.next // 0' <<<"$entry")

    enabled_state=$(systemctl is-enabled "$unit" 2>/dev/null || echo "unknown")
    local enabled_bool="false"
    [[ "$enabled_state" == "enabled" || "$enabled_state" == "static" ]] && enabled_bool="true"

    schedule_expr=$(systemctl show "$unit" -p TimersCalendar --value 2>/dev/null | sed -n 's/.*OnCalendar=\([^;]*\);.*/\1/p' | sed 's/[[:space:]]*$//')
    [[ -z "$schedule_expr" ]] && schedule_expr="(see: systemctl status $unit)"

    next_run_at=""
    if [[ "$next_usec" =~ ^[0-9]+$ && "$next_usec" -gt 0 ]]; then
      next_run_at=$(date -u -d "@$((next_usec / 1000000))" +"%Y-%m-%dT%H:%M:%SZ" 2>/dev/null || echo "")
    fi

    if [[ -n "$next_run_at" ]]; then
      add_task_with_next_run "systemd_timer" "$unit" "$activates" "$schedule_expr" "$unit" "$enabled_bool" "$next_run_at"
    else
      add_task "systemd_timer" "$unit" "$activates" "$schedule_expr" "$unit" "$enabled_bool"
    fi
  done < <(jq -c '.[]' <<<"$timers_json")
}

collect_cron
collect_systemd_timers

HOSTNAME_VALUE=$(hostname -f 2>/dev/null || hostname)
PAYLOAD=$(jq -n \
  --arg hostname "$HOSTNAME_VALUE" \
  --arg os_type "linux" \
  --arg reported_at "$(date -u +"%Y-%m-%dT%H:%M:%SZ")" \
  --argjson tasks "$TASKS_JSON" \
  '{hostname: $hostname, os_type: $os_type, reported_at: $reported_at, tasks: $tasks}')

if [[ "$DRY_RUN" == "1" ]]; then
  echo "$PAYLOAD" | jq .
  exit 0
fi

response=$(curl -sS -o /tmp/hlm-agent-response.json -w "%{http_code}" \
  -X POST "$API_URL/api/agent/report" \
  -H "Authorization: Bearer $API_TOKEN" \
  -H "Content-Type: application/json" \
  -d "$PAYLOAD")

if [[ "$response" -lt 200 || "$response" -ge 300 ]]; then
  echo "Report failed with HTTP $response:" >&2
  cat /tmp/hlm-agent-response.json >&2
  exit 1
fi

echo "Reported $(jq 'length' <<<"$TASKS_JSON") task(s) successfully."