Initial commit: Momentry Core v0.1
- Rust-based digital asset management system - Video analysis: ASR, OCR, YOLO, Face, Pose - RAG capabilities with Qdrant vector database - Multi-database support: PostgreSQL, Redis, MongoDB - Monitoring system with launchd plists - n8n workflow automation integration
This commit is contained in:
accusys
362
monitor/control/monitor_control.sh
Executable file
362
monitor/control/monitor_control.sh
Executable file
@@ -0,0 +1,362 @@
|
||||
#!/bin/bash
|
||||
|
||||
# Momentry 監控系統控制腳本
|
||||
# 路徑: /Users/accusys/momentry_core_0.1/monitor/control/monitor_control.sh
|
||||
|
||||
set -e
|
||||
|
||||
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
|
||||
MONITOR_DIR="$(dirname "$SCRIPT_DIR")"
|
||||
CONFIG_DIR="$MONITOR_DIR/config"
|
||||
LOG_DIR="/Users/accusys/momentry/log/monitor"
|
||||
|
||||
# 顏色定義
|
||||
RED='\033[0;31m'
|
||||
GREEN='\033[0;32m'
|
||||
YELLOW='\033[1;33m'
|
||||
BLUE='\033[0;34m'
|
||||
NC='\033[0m'
|
||||
|
||||
# ============================================================
|
||||
# 幫助信息
|
||||
# ============================================================
|
||||
|
||||
show_help() {
|
||||
cat << EOF
|
||||
Momentry 監控系統控制腳本
|
||||
|
||||
用法: $0 <command> [options]
|
||||
|
||||
命令:
|
||||
status 查看監控狀態
|
||||
check <layer> 執行特定層級檢查
|
||||
layers: service, workflow, portal, database, users, storage, external, backup, node, python, all
|
||||
monitor 持續監控 (每 5 分鐘)
|
||||
init 初始化監控數據庫表
|
||||
logs <layer> [lines] 查看日誌
|
||||
clean 清理歷史數據
|
||||
help 顯示幫助
|
||||
|
||||
示例:
|
||||
$0 status 查看所有監控狀態
|
||||
$0 check service 檢查服務狀態
|
||||
$0 check backup 檢查備份狀態
|
||||
$0 check all 執行全面檢查
|
||||
$0 logs anomaly 50 查看最近 50 條異常
|
||||
$0 init 初始化數據庫表
|
||||
|
||||
EOF
|
||||
}
|
||||
|
||||
# ============================================================
|
||||
# 初始化
|
||||
# ============================================================
|
||||
|
||||
init_monitor() {
|
||||
echo -e "${BLUE}初始化監控系統...${NC}"
|
||||
|
||||
# 創建日誌目錄
|
||||
mkdir -p "$LOG_DIR"
|
||||
|
||||
# 創建數據庫表
|
||||
echo "創建監控數據庫表..."
|
||||
psql -U accusys -h localhost -d momentry -f "$MONITOR_DIR/database/schema.sql" 2>/dev/null || \
|
||||
echo "數據庫表可能已存在"
|
||||
|
||||
echo -e "${GREEN}初始化完成${NC}"
|
||||
}
|
||||
|
||||
# ============================================================
|
||||
# 狀態查看
|
||||
# ============================================================
|
||||
|
||||
show_status() {
|
||||
echo ""
|
||||
echo "========================================"
|
||||
echo -e "${BLUE}Momentry 監控系統狀態${NC}"
|
||||
echo "========================================"
|
||||
echo ""
|
||||
|
||||
# 服務狀態
|
||||
echo -e "${YELLOW}Layer 2: 服務狀態${NC}"
|
||||
local service_count=$(psql -U accusys -h localhost -d momentry -t -A -c "SELECT COUNT(*) FROM monitor_services WHERE checked_at > NOW() - INTERVAL '5 minutes' AND status = 'up';" 2>/dev/null || echo "0")
|
||||
local service_total=$(psql -U accusys -h localhost -d momentry -t -A -c "SELECT COUNT(DISTINCT service_name) FROM monitor_services;" 2>/dev/null || echo "0")
|
||||
echo " 服務: $service_count / $service_total 正常"
|
||||
echo ""
|
||||
|
||||
# Workflow 狀態
|
||||
echo -e "${YELLOW}Layer 3: Workflow 狀態${NC}"
|
||||
local active_wf=$(psql -U accusys -h localhost -d momentry -t -A -c "SELECT COUNT(*) FROM monitor_workflows WHERE is_active = true;" 2>/dev/null || echo "0")
|
||||
local idle_wf=$(psql -U accusys -h localhost -d momentry -t -A -c "SELECT COUNT(*) FROM monitor_workflows WHERE idle_days > 30;" 2>/dev/null || echo "0")
|
||||
echo " 啟用 Workflow: $active_wf"
|
||||
echo " 閒置 Workflow (>30天): $idle_wf"
|
||||
echo ""
|
||||
|
||||
# Database 狀態
|
||||
echo -e "${YELLOW}Layer 5: Database 狀態${NC}"
|
||||
local db_count=$(psql -U accusys -h localhost -d momentry -t -A -c "SELECT COUNT(DISTINCT db_type) FROM monitor_databases WHERE checked_at > NOW() - INTERVAL '5 minutes';" 2>/dev/null || echo "0")
|
||||
echo " 監控資料庫: $db_count"
|
||||
echo ""
|
||||
|
||||
# 異常狀態
|
||||
echo -e "${YELLOW}Layer 6: 異常狀態${NC}"
|
||||
local critical=$(psql -U accusys -h localhost -d momentry -t -A -c "SELECT COUNT(*) FROM monitor_anomalies WHERE severity = 'critical' AND detected_at > NOW() - INTERVAL '24 hours';" 2>/dev/null || echo "0")
|
||||
local warning=$(psql -U accusys -h localhost -d momentry -t -A -c "SELECT COUNT(*) FROM monitor_anomalies WHERE severity IN ('medium', 'high') AND detected_at > NOW() - INTERVAL '24 hours';" 2>/dev/null || echo "0")
|
||||
echo " Critical: $critical"
|
||||
echo " Warning: $warning"
|
||||
echo ""
|
||||
|
||||
# Node.js 狀態
|
||||
echo -e "${YELLOW}Node.js 運行環境${NC}"
|
||||
local node_compliant=$(psql -U accusys -h localhost -d momentry -t -A -c "SELECT COUNT(*) FROM node_version_baseline WHERE is_compliant = true;" 2>/dev/null || echo "0")
|
||||
local node_total=$(psql -U accusys -h localhost -d momentry -t -A -c "SELECT COUNT(*) FROM node_version_baseline;" 2>/dev/null || echo "0")
|
||||
echo " 版本合規: $node_compliant / $node_total"
|
||||
echo ""
|
||||
|
||||
# Python 狀態
|
||||
echo -e "${YELLOW}Python 運行環境${NC}"
|
||||
local python_compliant=$(psql -U accusys -h localhost -d momentry -t -A -c "SELECT COUNT(*) FROM python_version_baseline WHERE is_compliant = true;" 2>/dev/null || echo "0")
|
||||
local python_total=$(psql -U accusys -h localhost -d momentry -t -A -c "SELECT COUNT(*) FROM python_version_baseline;" 2>/dev/null || echo "0")
|
||||
echo " 版本合規: $python_compliant / $python_total"
|
||||
echo ""
|
||||
|
||||
# 備份狀態
|
||||
echo -e "${YELLOW}Layer 7: 備份狀態${NC}"
|
||||
local total_backups=$(psql -U accusys -h localhost -d momentry -t -A -c "SELECT COUNT(*) FROM backup_registry WHERE created_at > NOW() - INTERVAL '7 days';" 2>/dev/null || echo "0")
|
||||
local failed_backups=$(psql -U accusys -h localhost -d momentry -t -A -c "SELECT COUNT(*) FROM backup_registry WHERE status = 'failed' AND created_at > NOW() - INTERVAL '7 days';" 2>/dev/null || echo "0")
|
||||
echo " 本週備份: $total_backups"
|
||||
echo " 失敗: $failed_backups"
|
||||
echo ""
|
||||
|
||||
echo "========================================"
|
||||
echo "使用 '$0 check <layer>' 執行檢查"
|
||||
echo ""
|
||||
}
|
||||
|
||||
# ============================================================
|
||||
# 執行檢查
|
||||
# ============================================================
|
||||
|
||||
check_layer() {
|
||||
local layer=$1
|
||||
|
||||
case $layer in
|
||||
service)
|
||||
echo -e "${BLUE}執行 Layer 2: 服務監控...${NC}"
|
||||
bash "$MONITOR_DIR/service/health_check.sh"
|
||||
;;
|
||||
workflow)
|
||||
echo -e "${BLUE}執行 Layer 3: Workflow 監控...${NC}"
|
||||
bash "$MONITOR_DIR/workflow/n8n_workflow_monitor.sh"
|
||||
;;
|
||||
portal)
|
||||
echo -e "${BLUE}執行 Layer 4: Portal 監控...${NC}"
|
||||
bash "$MONITOR_DIR/portal/page_monitor.sh"
|
||||
;;
|
||||
database)
|
||||
echo -e "${BLUE}執行 Layer 5: Database 監控...${NC}"
|
||||
bash "$MONITOR_DIR/database/postgres_monitor.sh"
|
||||
bash "$MONITOR_DIR/database/redis_monitor.sh"
|
||||
bash "$MONITOR_DIR/database/qdrant_monitor.sh"
|
||||
;;
|
||||
users)
|
||||
echo -e "${BLUE}執行 Layer 6: 使用者監控...${NC}"
|
||||
bash "$MONITOR_DIR/users/session_tracker.sh"
|
||||
;;
|
||||
storage)
|
||||
echo -e "${BLUE}執行 Layer 7: Storage 監控...${NC}"
|
||||
bash "$MONITOR_DIR/storage/storage_manager.sh" status
|
||||
;;
|
||||
backup)
|
||||
echo -e "${BLUE}執行 Layer 7: 備份監控...${NC}"
|
||||
bash "$MONITOR_DIR/storage/backup_monitor.sh" status
|
||||
;;
|
||||
external)
|
||||
echo -e "${BLUE}執行 Layer 1: External 監控...${NC}"
|
||||
bash "$MONITOR_DIR/service/external_monitor.sh"
|
||||
;;
|
||||
node)
|
||||
echo -e "${BLUE}執行 Node.js 版本監控...${NC}"
|
||||
bash "$MONITOR_DIR/service/node_monitor.sh"
|
||||
;;
|
||||
python)
|
||||
echo -e "${BLUE}執行 Python 版本監控...${NC}"
|
||||
bash "$MONITOR_DIR/service/python_monitor.sh"
|
||||
;;
|
||||
all)
|
||||
echo -e "${BLUE}執行全面監控檢查...${NC}"
|
||||
check_layer external
|
||||
check_layer service
|
||||
check_layer node
|
||||
check_layer python
|
||||
check_layer workflow
|
||||
check_layer portal
|
||||
check_layer database
|
||||
check_layer users
|
||||
echo -e "${GREEN}全面檢查完成${NC}"
|
||||
;;
|
||||
*)
|
||||
echo -e "${RED}未知層級: $layer${NC}"
|
||||
show_help
|
||||
exit 1
|
||||
;;
|
||||
esac
|
||||
}
|
||||
|
||||
# ============================================================
|
||||
# 持續監控
|
||||
# ============================================================
|
||||
|
||||
run_monitor() {
|
||||
echo -e "${BLUE}開始持續監控 (每 5 分鐘)${NC}"
|
||||
echo "按 Ctrl+C 停止"
|
||||
echo ""
|
||||
|
||||
while true; do
|
||||
local start_time=$(date +%s)
|
||||
|
||||
check_layer all
|
||||
|
||||
local end_time=$(date +%s)
|
||||
local elapsed=$((end_time - start_time))
|
||||
|
||||
if [ $elapsed -lt 300 ]; then
|
||||
sleep $((300 - elapsed))
|
||||
fi
|
||||
done
|
||||
}
|
||||
|
||||
# ============================================================
|
||||
# 查看日誌
|
||||
# ============================================================
|
||||
|
||||
show_logs() {
|
||||
local layer=${1:-anomaly}
|
||||
local lines=${2:-20}
|
||||
|
||||
case $layer in
|
||||
anomaly)
|
||||
echo -e "${BLUE}最近異常記錄:${NC}"
|
||||
psql -U accusys -h localhost -d momentry -c "
|
||||
SELECT
|
||||
TO_CHAR(detected_at, 'YYYY-MM-DD HH24:MI:SS') as time,
|
||||
severity,
|
||||
anomaly_type,
|
||||
username,
|
||||
LEFT(description, 50) as desc
|
||||
FROM monitor_anomalies
|
||||
ORDER BY detected_at DESC
|
||||
LIMIT $lines;
|
||||
" 2>/dev/null || echo "無法連接資料庫"
|
||||
;;
|
||||
service)
|
||||
echo -e "${BLUE}最近服務狀態:${NC}"
|
||||
psql -U accusys -h localhost -d momentry -c "
|
||||
SELECT
|
||||
service_name,
|
||||
status,
|
||||
response_time_ms,
|
||||
TO_CHAR(checked_at, 'YYYY-MM-DD HH24:MI:SS') as time
|
||||
FROM monitor_services
|
||||
ORDER BY checked_at DESC
|
||||
LIMIT $lines;
|
||||
" 2>/dev/null || echo "無法連接資料庫"
|
||||
;;
|
||||
workflow)
|
||||
echo -e "${BLUE}最近 Workflow 狀態:${NC}"
|
||||
psql -U accusys -h localhost -d momentry -c "
|
||||
SELECT
|
||||
workflow_name,
|
||||
is_active,
|
||||
idle_days,
|
||||
suggestion,
|
||||
TO_CHAR(checked_at, 'YYYY-MM-DD HH24:MI:SS') as time
|
||||
FROM monitor_workflows
|
||||
ORDER BY checked_at DESC
|
||||
LIMIT $lines;
|
||||
" 2>/dev/null || echo "無法連接資料庫"
|
||||
;;
|
||||
*)
|
||||
echo -e "${RED}未知日誌類型: $layer${NC}"
|
||||
;;
|
||||
esac
|
||||
}
|
||||
|
||||
# ============================================================
|
||||
# 清理歷史數據
|
||||
# ============================================================
|
||||
|
||||
clean_history() {
|
||||
echo -e "${YELLOW}清理歷史數據...${NC}"
|
||||
|
||||
psql -U accusys -h localhost -d momentry << EOF 2>/dev/null || true
|
||||
-- 保留 30 天
|
||||
DELETE FROM monitor_services WHERE checked_at < NOW() - INTERVAL '30 days';
|
||||
DELETE FROM monitor_workflows WHERE checked_at < NOW() - INTERVAL '30 days';
|
||||
DELETE FROM monitor_databases WHERE checked_at < NOW() - INTERVAL '30 days';
|
||||
DELETE FROM monitor_external WHERE checked_at < NOW() - INTERVAL '30 days';
|
||||
DELETE FROM monitor_portal_pages WHERE checked_at < NOW() - INTERVAL '30 days';
|
||||
|
||||
-- 保留 30 天版本基線
|
||||
DELETE FROM node_version_baseline WHERE checked_at < NOW() - INTERVAL '30 days';
|
||||
DELETE FROM node_process_tracking WHERE checked_at < NOW() - INTERVAL '30 days';
|
||||
DELETE FROM python_version_baseline WHERE checked_at < NOW() - INTERVAL '30 days';
|
||||
DELETE FROM python_script_tracking WHERE checked_at < NOW() - INTERVAL '30 days';
|
||||
|
||||
-- 保留 7 天會話
|
||||
DELETE FROM monitor_sessions WHERE connected_at < NOW() - INTERVAL '7 days';
|
||||
|
||||
-- 保留 30 天登入
|
||||
DELETE FROM monitor_logins WHERE login_at < NOW() - INTERVAL '30 days';
|
||||
|
||||
-- 保留 90 天異常
|
||||
DELETE FROM monitor_anomalies WHERE detected_at < NOW() - INTERVAL '90 days';
|
||||
|
||||
-- 清理空閒空間
|
||||
VACUUM ANALYZE;
|
||||
EOF
|
||||
|
||||
echo -e "${GREEN}清理完成${NC}"
|
||||
}
|
||||
|
||||
# ============================================================
|
||||
# 主程序
|
||||
# ============================================================
|
||||
|
||||
main() {
|
||||
# 確保日誌目錄存在
|
||||
mkdir -p "$LOG_DIR"
|
||||
|
||||
local command=${1:-status}
|
||||
|
||||
case $command in
|
||||
status)
|
||||
show_status
|
||||
;;
|
||||
check)
|
||||
check_layer ${2:-all}
|
||||
;;
|
||||
monitor)
|
||||
run_monitor
|
||||
;;
|
||||
init)
|
||||
init_monitor
|
||||
;;
|
||||
logs)
|
||||
show_logs ${2:-anomaly} ${3:-20}
|
||||
;;
|
||||
clean)
|
||||
clean_history
|
||||
;;
|
||||
help|--help|-h)
|
||||
show_help
|
||||
;;
|
||||
*)
|
||||
echo -e "${RED}未知命令: $command${NC}"
|
||||
show_help
|
||||
exit 1
|
||||
;;
|
||||
esac
|
||||
}
|
||||
|
||||
main "$@"
|
||||
Reference in New Issue
Block a user