BREAKING: Cloudflare MINOR — workarounds inside [ckb803x3fd8p]

Cloudflare is down: Elevated errors in item PUT requests to AI Search. Immediate workarounds for indie hackers.

BREAKING: Cloudflare Experiencing Elevated Errors on AI Search PUT Requests

Status: Investigating | Severity: Minor | Impact: Partial disruption

---

🔴 What's Down & Who's Affected

Cloudflare is reporting elevated error rates specifically on item PUT requests to AI Search endpoints. This affects:

Projects using Cloudflare Workers AI Search indexing

Any indie app performing batch document updates or data sync operations

Integrations relying on vector database PUT operations through Cloudflare

Search features that depend on real-time index updates

GET requests appear unaffected. Read operations are stable. Only write/update operations to the search index are experiencing issues.

---

⚡ Immediate Workarounds (Do This Now)

1. Queue Your PUT Requests

Implement a simple retry queue in your application: ```javascript const retryQueue = []; const maxRetries = 5; const baseDelay = 2000; // 2 seconds

async function queuedPUT(endpoint, data, retries = 0) { try { return await fetch(endpoint, { method: 'PUT', body: JSON.stringify(data) }); } catch (e) { if (retries < maxRetries) { const delay = baseDelay * Math.pow(2, retries); setTimeout(() => queuedPUT(endpoint, data, retries + 1), delay); } } } ```

2. Defer Non-Critical Updates

Postpone index updates that aren't urgent. Queue them for off-peak hours (typically 2-6 AM UTC where Cloudflare incidents resolve faster).

3. Switch to Read Mode

If you're in a feature launch window, temporarily disable write operations and serve cached search results. Your users won't notice.

4. Use Regional Fallbacks

If you have data in multiple regions, route PUT requests to alternative Cloudflare zones or endpoints temporarily.

---

🔍 How to Check If You're Affected

Run this diagnostic:

```bash curl -X PUT https://api.cloudflare.com/client/v4/accounts/{account}/ai/search/indexes/{index}/documents \ -H "Authorization: Bearer {token}" \ -H "Content-Type: application/json" \ -d '{"id":"test","data":{}}' ```

If you see 429 (rate limit) or 5xx errors, you're hitting the incident. 200 or 201 = you're clear.

Alternatively, check [Cloudflare's status page](https://www.cloudflarestatus.com/) for real-time updates.

---

🛠️ Alternative Tools to Consider

If this persists beyond 2 hours, consider temporary migration:

Typesense — Open-source, self-hosted search (fast setup)

Meilisearch — Lightweight, developer-friendly alternative

Pinecone — Managed vector DB (similar to Cloudflare, different infrastructure)

Supabase pgvector — If you need vector search with PostgreSQL

Elasticsearch — Enterprise option if you're at scale

---

📊 How to Monitor Recovery

1. Watch Cloudflare Status: Updates posted every 15-30 minutes 2. Test Regularly: Run the diagnostic curl command above every 5 minutes 3. Monitor Your Logs: Set up alerts on 5xx error rates in your application 4. Join Cloudflare Discord: #status-updates channel posts real-time info 5. Check Status API: https://www.cloudflarestatus.com/api/v2/components.json

---

💡 What We Know

Outage started ~15 minutes ago

GET requests are unaffected (read-only operations work fine)

Cloudflare engineers are actively investigating

ETA for resolution: typically 30-60 minutes for "Investigating" status

Stay calm. This is a partial disruption. Your site isn't down — only search indexing is degraded.

We'll update this post every 10 minutes. Check back soon.