2026-09-27 13:08 UTCIncidents Detected

APAC Latency Issues Linger

No widespread internet outage is underway this Sunday. Every major cloud platform reports normal operations: Google Cloud, Microsoft Azure, Azure DevOps, GitHub, OpenAI, Anthropic (Claude) and Discord all show no active incidents. AWS's public health dashboard lists no new events; the only open items are the long-running Middle East region (UAE and Bahrain) notices dating back to early March.

The ongoing issues are regional and minor, and most of them are in Asia-Pacific:

Akamai (APAC): Akamai is still monitoring origin-connectivity problems that it put down to "upstream congestion related to unexpected maintenance on subsea cables". A separate advisory warns of higher peak-hour latency on routes through Singapore, Hong Kong and Japan while upstream provider maintenance runs through October 3. Akamai says core services are still operational. https://www.akamaistatus.com

Cloudflare (APAC): Since September 23, Cloudflare has been mitigating intermittent connectivity problems that affect a small subset of Asia-Pacific customers. Two older, smaller issues are also still open: incorrect challenges for Cloudflare One clients (since Sept 22) and incorrect geolocation for WARP users (since Aug 27). Cloudflare has scheduled extensive datacenter maintenance for September 28-29, which may briefly reroute traffic or add latency in some regions. https://www.cloudflarestatus.com

Zoom Phone: Since September 25, some customers using Polycom desk phones have seen their devices go offline. Zoom has identified the cause. https://www.zoomstatus.com

Twilio: Several carrier-specific messaging delays are open, including MMS to GCI (US), SMS to Datora (Brazil) and MMS in Australia. None of them is a platform-wide outage. https://status.twilio.com

Also checked and operational: Atlassian, Slack, Vercel, Heroku, DigitalOcean, Cursor, Groq, ElevenLabs, Cohere and Stability AI. We could not get a reliable live reading from Stripe's or Meta's status pages this run, and we have seen no reports of problems with either.

Past 72 hours: GitHub had a multi-service degradation on September 23-24, which hit Projects, API requests and organization creation and was traced to database replicas, plus a billing-update issue on September 24. Both are resolved. On September 26, crowd-sourced reports showed short-lived spikes for Peacock, AccuWeather and Canara Bank.

One third-party tracker flagged a connectivity drop in El Salvador today. IODA's own alert feed shows no alert for the country in the past 48 hours, so we are treating that report as unconfirmed.

Bottom line: the internet is up. Expect somewhat higher latency on some Asia-Pacific routes until the subsea cable maintenance ends.

← Home View all history

APAC Latency Issues Linger

Last 30 Days