fix: CRITICAL - Prevent infinite DROP DATABASE loop

ROOT CAUSE: Monitor was doing DROP DATABASE when NODE2 agents were missing,
but the backup didn't have NODE2 agents, causing an infinite loop.

FIX:
- FULL RECOVERY (DROP DATABASE) only when MicroDAOs < 5 (critical data loss)
- SOFT RECOVERY (just sync agents) when MicroDAOs exist but agents missing
- Prefer backup with NODE2 agents (full_backup_with_node2*.sql)
- Never DROP DATABASE if MicroDAOs exist

This prevents the daily data loss issue.

scripts/monitor-db-stability.sh

@@ -1,6 +1,10 @@
 #!/bin/bash
 # Monitor database stability and auto-recover if needed
 # Run this periodically (e.g., every 5 minutes via cron)
+#
+# IMPORTANT: This script has two recovery modes:
+# 1. FULL RECOVERY (DROP DATABASE) - only when MicroDAOs < 5 (critical data loss)
+# 2. SOFT RECOVERY (add NODE2 agents) - when MicroDAOs exist but agents missing
 
 set -e
 
@@ -42,17 +46,22 @@ NODE2_AGENT_COUNT=${NODE2_AGENT_COUNT:-0}
 
 log "📊 Data check: MicroDAOs=$MICRODAO_COUNT, Agents=$AGENT_COUNT, NODE2=$NODE2_AGENT_COUNT"
 
-# If data is missing, try to restore
-# Need at least 5 MicroDAOs, 50 total agents, and 45+ NODE2 agents
-if [ "$MICRODAO_COUNT" -lt 5 ] || [ "$AGENT_COUNT" -lt 50 ] || [ "$NODE2_AGENT_COUNT" -lt 45 ]; then
-  log "⚠️  Data loss detected! Attempting recovery..."
+# ============================================================================
+# CASE 1: CRITICAL DATA LOSS - MicroDAOs missing (< 5)
+# Only in this case we do FULL RECOVERY with DROP DATABASE
+# ============================================================================
+if [ "$MICRODAO_COUNT" -lt 5 ]; then
+  log "🔴 CRITICAL: MicroDAOs missing ($MICRODAO_COUNT < 5). Starting FULL RECOVERY..."
   
-  # Check for recent backup
-  LATEST_BACKUP=$(ls -t /opt/microdao-daarion/db_backups/*.sql 2>/dev/null | head -1)
+  # Check for recent backup (prefer backup with NODE2 agents)
+  LATEST_BACKUP=$(ls -t /opt/microdao-daarion/db_backups/full_backup_with_node2*.sql 2>/dev/null | head -1)
+  if [ -z "$LATEST_BACKUP" ]; then
+    LATEST_BACKUP=$(ls -t /opt/microdao-daarion/db_backups/*.sql 2>/dev/null | head -1)
+  fi
   
   if [ -n "$LATEST_BACKUP" ]; then
     log "📦 Found backup: $LATEST_BACKUP"
-    log "🔄 Restoring from backup..."
+    log "🔄 Starting full restoration..."
     
     # Terminate all connections to the database first
     log "🔒 Terminating active connections..."
@@ -72,31 +81,60 @@ if [ "$MICRODAO_COUNT" -lt 5 ] || [ "$AGENT_COUNT" -lt 50 ] || [ "$NODE2_AGENT_C
       docker exec -i daarion-postgres psql -U postgres -d daarion < "$f" 2>&1 | grep -v "already exists\|does not exist" || true
     done
     
-    # Sync NODE2 agents (force sync even if they exist)
+    # Sync NODE2 agents
     log "🤖 Syncing NODE2 agents..."
     python3 scripts/sync-node2-dagi-agents.py 2>&1 | tail -10 || true
     
-    # Verify agent count after sync
-    AGENT_COUNT_AFTER=$(docker exec daarion-postgres psql -U postgres -d daarion -t -c "SELECT COUNT(*) FROM agents;" 2>/dev/null | tr -d ' \n' || echo "0")
-    log "📊 Agents after sync: $AGENT_COUNT_AFTER"
-    
     # Remove test agents
     bash scripts/remove-test-agents.sh 2>&1 | tail -3 || true
     
-    # Fix asset URLs (logos and banners)
+    # Fix asset URLs
     log "🖼️  Fixing asset URLs..."
     bash scripts/fix-asset-urls.sh 2>&1 | tail -5 || true
     
-    log "✅ Recovery complete"
+    # Verify
+    AGENT_COUNT_AFTER=$(docker exec daarion-postgres psql -U postgres -d daarion -t -c "SELECT COUNT(*) FROM agents;" 2>/dev/null | tr -d ' \n' || echo "0")
+    log "📊 Agents after full recovery: $AGENT_COUNT_AFTER"
+    
+    log "✅ Full recovery complete"
   else
-    log "❌ No backup found for recovery"
+    log "❌ No backup found for full recovery"
   fi
+
+# ============================================================================
+# CASE 2: SOFT RECOVERY - MicroDAOs exist but NODE2 agents missing
+# DO NOT DROP DATABASE - just sync NODE2 agents
+# ============================================================================
+elif [ "$NODE2_AGENT_COUNT" -lt 45 ]; then
+  log "🟡 WARNING: NODE2 agents missing ($NODE2_AGENT_COUNT < 45). Starting SOFT RECOVERY (no DROP)..."
+  
+  cd /opt/microdao-daarion
+  
+  # Just sync NODE2 agents - NO DROP DATABASE!
+  log "🤖 Syncing NODE2 agents..."
+  python3 scripts/sync-node2-dagi-agents.py 2>&1 | tail -10 || true
+  
+  # Remove test agents if any
+  bash scripts/remove-test-agents.sh 2>&1 | tail -3 || true
+  
+  # Fix asset URLs
+  log "🖼️  Fixing asset URLs..."
+  bash scripts/fix-asset-urls.sh 2>&1 | tail -5 || true
+  
+  # Verify
+  NODE2_AFTER=$(docker exec daarion-postgres psql -U postgres -d daarion -t -c "SELECT COUNT(*) FROM agents WHERE node_id = 'node-2-macbook-m4max';" 2>/dev/null | tr -d ' \n' || echo "0")
+  AGENT_COUNT_AFTER=$(docker exec daarion-postgres psql -U postgres -d daarion -t -c "SELECT COUNT(*) FROM agents;" 2>/dev/null | tr -d ' \n' || echo "0")
+  log "📊 After soft recovery: Agents=$AGENT_COUNT_AFTER, NODE2=$NODE2_AFTER"
+  
+  log "✅ Soft recovery complete"
+else
+  log "✅ Data integrity OK"
 fi
 
-# Check PostgreSQL logs for errors
-ERROR_COUNT=$(docker logs daarion-postgres --since 5m 2>&1 | grep -i "fatal\|error\|panic" | wc -l)
-if [ "$ERROR_COUNT" -gt 0 ]; then
-  log "⚠️  Found $ERROR_COUNT errors in PostgreSQL logs in last 5 minutes"
+# Check PostgreSQL logs for errors (informational only)
+ERROR_COUNT=$(docker logs daarion-postgres --since 5m 2>&1 | grep -i "fatal\|error\|panic" | grep -v "already exists\|does not exist\|column.*does not exist" | wc -l)
+if [ "$ERROR_COUNT" -gt 5 ]; then
+  log "⚠️  Found $ERROR_COUNT significant errors in PostgreSQL logs in last 5 minutes"
 fi
 
 # Check container restart count
@@ -106,4 +144,3 @@ if [ "$RESTART_COUNT" -gt 10 ]; then
 fi
 
 log "✅ Stability check complete"
-