No widespread internet outage was detected on October 21, 2026. Major cloud, CDN and platform status pages were clear, with a few smaller incidents noted below.
Ongoing
Twilio: several regional messaging and voice issues remain open, including Brazil voice calls and Italy SMS (both recovering), Israel and Thailand voice post-dial delay (cause identified), and MMS delays or failures for GCI (US) and Australia (still investigating). Planned carrier maintenance for US MMS and T-Mobile SMS/MMS was also in progress. Source: status.twilio.comCloudflare: a minor incident affecting incorrect geolocation for some WARP users is being worked on; routine datacenter maintenance is also under way. Source: cloudflarestatus.comZoom: scheduled Zoom Phone maintenance in the AMS3 data center (EMEA) runs through the evening of October 21. Source: zoomstatus.comRecently resolved (past 72 hours)
Google Cloud: a multi-service incident in europe-west2 lasted about 8 hours and was closed on October 20. Source: status.cloud.google.comGitHub: degraded Actions, Pages and Webhooks on October 19, resolved. Source: githubstatus.comCursor: Grok Bot and Opus 5.5 errors on October 19, resolved. Source: status.cursor.comCloudflare: a Zero Trust connectivity incident on October 20 is marked resolved.Background: Akamai's September APAC origin connectivity issue, caused by upstream subsea cable maintenance, is marked resolved. Its final notice did not confirm that cable repairs were complete.
All clear at time of check: AWS (status page returned no details), Azure, OpenAI, Anthropic, Discord, Akamai, Atlassian, Slack, Vercel, DigitalOcean, Groq, ElevenLabs, Cohere and Stability AI. Heroku, Stripe, Replicate and Meta status pages did not return usable content, so their status is unconfirmed. Web searches and crowd-sourced trackers surfaced no current widespread outage or BGP incident.