web-api: add googlevideo CDN collection and API

router/collect-googlevideo.sh

+#!/bin/bash
+# Collect googlevideo CDN domains from BIND query log and send to route-web-api.
+# Runs as a systemd service on dhcp (DNS server).
+# Tails the query log, extracts new rr*---sn-*.googlevideo.com domains,
+# normalizes them and sends via API to igw.
+
+API_URL="http://igw.etersoft.ru/api/googlevideo"
+QUERY_LOG="/var/lib/bind/var/log/named/queries"
+
+# Local cache of already-sent patterns (survives restarts)
+SEEN_FILE="/var/lib/collect-googlevideo/seen-patterns"
+
+mkdir -p "$(dirname "$SEEN_FILE")"
+touch "$SEEN_FILE"
+
+log() {
+    echo "$(date '+%Y-%m-%d %H:%M:%S') $*"
+}
+
+# Load known patterns into memory
+declare -A SEEN
+while IFS= read -r line; do
+    [ -n "$line" ] && SEEN["$line"]=1
+done < "$SEEN_FILE"
+
+log "Started, ${#SEEN[@]} known patterns"
+
+tail -F "$QUERY_LOG" 2>/dev/null | grep --line-buffered 'googlevideo\.com' | while IFS= read -r line; do
+    # Extract domain: "query: rr4---sn-ajixh5-55.googlevideo.com IN A"
+    domain=$(echo "$line" | grep -oP 'query: \Krr\d+---sn-[a-z0-9-]+\.googlevideo\.com')
+    [ -z "$domain" ] && continue
+
+    # Normalize: rr4---sn-xxx -> sn-xxx
+    suffix="${domain#rr*---}"
+    pattern="rr[1-8]---${suffix}"
+
+    # Skip if already seen
+    [ -n "${SEEN[$pattern]}" ] && continue
+    SEEN["$pattern"]=1
+
+    # Send to API
+    response=$(curl -s -o /dev/null -w '%{http_code}' \
+        -X POST "$API_URL" \
+        -H 'Content-Type: application/json' \
+        -d "{\"domain\":\"$domain\"}" \
+        --connect-timeout 5 --max-time 10)
+
+    if [ "$response" = "200" ]; then
+        echo "$pattern" >> "$SEEN_FILE"
+        log "NEW $pattern (from $domain) -> API $response"
+    else
+        # Don't cache on failure — retry next time
+        unset 'SEEN[$pattern]'
+        log "FAIL $pattern (from $domain) -> API $response"
+    fi
+done