checkpoint: pipeline + 27k/h baseline before sitemap discovery

This commit is contained in:
qananasikq
2026-04-18 16:34:09 +03:00
parent a8c5f8aa41
commit 5999c2e42d
30 changed files with 4993 additions and 61 deletions

55
_enable_proxy.sh Normal file
View File

@@ -0,0 +1,55 @@
#!/usr/bin/env bash
set -euo pipefail
cd /root/iaai_scraper_project
echo "== patch .env =="
# удалить старые SOCKS5_*/IAAI_PROXY_* строки если есть
sed -i '/^SOCKS5_PROXY_/d; /^IAAI_PROXY_/d' .env
cat >> .env <<'EOF'
# --- SOCKS5 US residential proxy (BulletProof) ---
SOCKS5_PROXY_HOST=residential.bpproxy.at
SOCKS5_PROXY_PORT=1002
SOCKS5_PROXY_USER=bpuser-S7o3VlC0
SOCKS5_PROXY_PASS=YKmRzvjvnnvyQdEUQ8hM_country-US_session-q3TGNZZO_lifetime-30
EOF
grep -E '^(SOCKS5_|IAAI_PROXY_)' .env
echo "== restart worker =="
docker compose stop worker
docker compose up -d worker
echo "== wait 8s for worker + bridge =="
sleep 8
echo "== verify proxy bridge + external IP =="
docker compose exec -T worker sh -c 'curl -s --max-time 20 --proxy http://127.0.0.1:8899 https://api.ipify.org' || echo "proxy check FAILED"
echo
echo "== direct check (no proxy) =="
docker compose exec -T worker sh -c 'curl -s --max-time 10 https://api.ipify.org' || true
echo
echo "== clear locks/state =="
docker compose exec -T redis redis-cli DEL iaai:locks:sync_listing iaai:state:startup_sync_dispatched iaai:state:sync_listing_followup_pending
echo "== dispatch fresh task =="
TID=$(docker compose exec -T worker python -c "from iaai_scraper.worker.tasks import sync_listing; print(sync_listing.delay().id)")
echo "dispatched: $TID"
echo "== wait 90s =="
sleep 90
echo
echo "== DB =="
docker compose exec -T postgres psql -U iaai -d iaai_scraper -c "SELECT count(*) AS total, count(*) FILTER (WHERE last_seen_at > now() - interval '5 minutes') AS last_5min, max(last_seen_at) AS max FROM cars;"
echo
echo "== last 30 log lines (batch / upserted / HTTP-mode page) =="
docker compose logs --tail=500 worker 2>&1 | grep -E 'HTTP-mode batch|upserted=|HTTP-mode listing done|Segment .* done|HTTP fast-path using proxy|proxy_bridge' | tail -n 30
echo
echo "== sample of recent pagination lines =="
docker compose logs --tail=300 worker 2>&1 | grep -E 'HTTP-mode page' | tail -n 10