AI Service Outages Sep–Oct 2026: OpenAI and Claude Incident Stats
110 incidents from seven official status pages, 2026-09-17 to 10-17, broken down by vendor, severity and duration, with retry, fallback and failover tactics.
Blog
Networks and models, explained.
Troubleshooting guides, IP and privacy explainers, AI service updates and model watch — each backed by tools and data you can check yourself.
AI API Latency Breakdown: Connect Time, Gateway RTT, and TTFT
Why is your AI API slow? Split a call into connect time, gateway round trip, and time to first token (TTFT), with curl commands to measure each segment.
Browser Fingerprint: What Timezone, Language and IP Mismatches Reveal
Sites read your timezone, language and fonts, which often clash with your proxy exit IP. See what each mismatch reveals and how to check it with cdnping /ip.
Claude Code Connection Timeout and Errors: 4-Layer Checklist
Fix Claude Code connection timeouts and errors layer by layer: service status, local network, exit IP and local config. Includes a curl timing command, an error-code table and cdnping tools.
Cloudflare for AI Apps: Caching, Rate Limiting and Origin Protection
What to cache, how to rate limit AI APIs, lock your origin to Cloudflare, and fix SSE buffering and 524 errors. Includes Workers code and a launch checklist.
Codex Usage Limit Resets: Timeline and How to Track the Next
A sourced timeline of the two 2026 Codex usage limit resets, how short-window limits differ from weekly limits, and how to track the next reset with cdnping.
Custom API Endpoint for Claude Code, Codex and OpenCode
Where Claude Code, Codex and OpenCode read the endpoint URL and API key, with copy-paste env vars and config files, a 401/404/timeout fix table and network self-checks.
Datacenter IP vs Residential IP: 5 Signals Sites Use to Tell Them Apart
How datacenter and residential IPs differ, and the five signals sites use for IDC IP detection (ASN, RDAP netname, reverse DNS, reputation, latency), with self-check commands.
Developer Network Self-Check (2026): Exit IP to AI API Latency
A nine-point developer network self-check: exit IP, split routing, DNS, WebRTC, IPv6, AI API latency, CDNs and status feeds, with CLI checks and cdnping tools.
DNS over HTTPS Setup Guide for Browsers, Windows and macOS
What DNS over HTTPS does and doesn't protect, how to enable DoH in Chrome, Edge, Firefox, Windows 11 and macOS, and how to confirm it works with the cdnping /ip DNS leak test.
How to Fix a WebRTC Leak in Chrome, Firefox and Safari
A WebRTC leak can expose your real IP. Per-browser fixes: Firefox about:config, Chrome's extension and WebRtcIPHandling policy, Safari settings, side effects, and a retest on /ip.
How Chinese Sites See Your Real IP: Split-Tunnel Detection Explained
How split-tunnel detection works: Chinese CDNs return your client IP, CORS decides if a page can read it, and what the three results mean. Test it on cdnping.
How to Read OpenAI and Claude Status Pages: Levels & Components
What do none, minor, major and critical mean on OpenAI and Claude status pages? Components, incident stages, lag, and how to cross-check with API probes.
IP Purity Score Explained: 14 Reputation Sources and Weights
cdnping's IP purity score is 100 minus a risk value. See what each of the 14 reputation sources checks, its weight, why false positives happen, and why checkers disagree.
IPv6 Proxy Not Working on Dual-Stack: Why IPv4 and IPv6 Exits Differ
IPv6 proxy not working? Often the proxy captures only IPv4 while IPv6 goes direct. Learn the causes, how to confirm with /ip and curl -4/-6, and four fixes.
Is ChatGPT Down or Is It Your Network? A 3-Step Check
When ChatGPT won't load, check official status and user reports, then API probes, then your own exit IP and DNS. Includes commands and an error table.
Is the Model Dumber or the Provider Different? LLM Quantization (fp8, fp4)
The same model can swing between good and bad because inference providers quantize differently. We count fp8, fp4 and int4 across 137 OpenRouter endpoints and show how to pin precision.
Multi-Model Routing: Cut LLM Costs by Task Without Losing Quality
How to route LLM requests by task tier, estimate token costs from OpenRouter list prices, compare DIY routing with aggregation gateways, and verify quality.
OpenRouter Provider Uptime: How Much Does It Vary by Model?
Compare 1-day uptime and quantization for the same model across OpenRouter providers using a 2026-10-17 cdnping snapshot, learn how to read nulls, and see how to set provider preferences.
Track LLM API Price Changes: Changelogs, Model Catalog, Billing
How to track LLM API price changes, new models and deprecations: three sources to watch, cdnping's /changes RSS and /models, and team alerts.