BREAKING: Cloudflare DO/D1 Degraded in Hong Kong & Singapore 🚨 Workarounds Inside

Cloudflare Durable Objects and D1 database experiencing partial degradation in APAC region. Immediate actions for affected indie hackers.

BREAKING: Cloudflare DO/D1 Partial Outage – Hong Kong & Singapore

Status: Investigating | Severity: MINOR | Impact: Partial disruption | Updated: NOW

---

What's Down & Who's Affected

Cloudflare is currently experiencing degraded performance on:

Durable Objects (DO) – serverless stateful compute

D1 Database – SQLite edge database

Region: Hong Kong and Singapore data centers

You're affected if:

Your app uses DO for real-time state (chat, gaming, collaboration tools)

D1 is your primary database layer in APAC

Users are geographically in HK/SG or routed through these regions

You have failover logic dependent on these services

Unaffected services: Workers (global), Pages, CDN, DNS, Email Routing continue normal operation.

---

Immediate Workarounds (RIGHT NOW)

1. Enable Geographic Failover

If you're using DO or D1, immediately implement region fallback: ```javascript // Route requests away from HK/SG const getPrimaryRegion = () => { const userRegion = request.headers.get('cf-ipcountry'); if (['HK', 'SG'].includes(userRegion)) { return 'us'; // failover to US } return 'default'; }; ```

2. Switch to Alternative Database (Temporary)

Redirect D1 queries to PostgreSQL (Neon/Supabase) or MongoDB Atlas

Use environment variables to swap connection strings

Cache reads aggressively in Workers KV to minimize database calls

3. Reduce DO State Writes

Queue state updates instead of writing directly

Use localStorage/sessionStorage on client-side temporarily

Batch writes when service recovers

4. Increase Cache TTL

```javascript // Aggressive edge caching during incident response.headers.set('Cache-Control', 'public, max-age=300'); ```

---

Check If YOUR Project Is Affected

Quick diagnostic: 1. Open Cloudflare Dashboard → Status Page 2. Check your Worker logs: wrangler tail --format pretty 3. Look for errors: - "DO request timeout" - "D1 connection refused" - 503/504 responses from APAC regions 4. Test from Hong Kong/Singapore IP using curl or browser DevTools

Monitoring your app: ```bash

Watch for DO timeouts

wrangler tail --format json | grep -i "timeout\|error"

Check D1 query latency

wrangler d1 execute <db-name> "SELECT 1" --remote ```

---

Alternative Tools to Consider

| Service | Alternative | Use Case | |---------|-------------|----------| | D1 | Neon / Supabase | PostgreSQL with edge support | | DO | Redis (Upstash) | Distributed state/caching | | Workers | Vercel Functions | Edge compute backup | | KV | Upstash Redis | Session/cache layer |

Pro tip: Don't migrate fully yet. This is MINOR severity—maintain hybrid setup for 24-48 hours.

---

Monitor Recovery

Official updates:

Cloudflare Status Page: https://www.cloudflarestatus.com/

Subscribe to incident email notifications

Follow @Cloudflare on Twitter

DIY monitoring: ```bash

Test connectivity every 5 minutes

watch -n 300 'wrangler d1 execute <db-name> "SELECT 1" --remote && echo "✅ OK" || echo "❌ Failed"' ```

Recovery timeline: Cloudflare typically resolves regional degradation within 30-120 minutes. Expect full restoration before peak traffic hours.

---

Action Items (Priority Order)

1. ✅ Verify if you're affected (2 min) 2. ✅ Enable region failover if using DO/D1 (10 min) 3. ✅ Set up alerting on worker errors (5 min) 4. ⏳ Prepare DB backup connection strings (15 min) 5. ⏳ Brief your users if applicable (monitor, don't panic-post)

Stay calm. This is partial. Global infrastructure is resilient. We've handled this before.

Will update as soon as Cloudflare confirms recovery. Keep monitoring.