ShadowGraph
Start free trial
Internet Outage Radar · live

All monitored providers operational

What is down, which provider is behind it, and who is at risk next.

—services probed
—reachable
—with errors
—hosting networks
—median latency
—last sweep · every 5 min
cloud regionserver location (city or country)errorsincident · blast radiusmeasurement pointCircle size = services hosted there. Drag to rotate, pinch or scroll to zoom, tap a node for details.
Measured from six continents

Is it down everywhere, or only here?

Our main radar probes all 900 services every 5 minutes from Frankfurt. Every 30 minutes, key infrastructure and every service that Frankfurt sees failing are checked again from North and South America, Europe, Asia, Oceania and Africa. That separates a global outage from a regional or measurement-path problem.

Dependency graph

Follow every path

Click a provider to see every service that depends on it and what else those services rely on. Click a service to see its full chain: CDN, hosting, region, DNS and email.

Provider incidents

None right now. An incident is declared only when several independent services on the same provider, region, CDN or DNS fail in two consecutive sweeps.

Live event feed

    Error rate · 24h

    By category

    How concentrated is the internet?

    share of monitored services per provider

    Hosting

    CDN / edge

    DNS

    Email

    Providers

    Data-center regions

    ProviderRegionServicesHealth
    Stack alerts · 14-day free trial

    Know before your vendor's status page does

    Pick the services you depend on and add your own endpoints. When one of them, or the provider underneath it, starts failing, you get a Slack, Teams, webhook or email alert with the blast radius. $29 a month after the trial.

    Start free trialCheck your stack first

    All monitored services (0)

    ServiceStatusLatency24h uptimeHostingCDNDNS

    Full sweeps run every 5 minutes from our main measurement node in Frankfurt; a service counts as down after two consecutive failed probes, and targets our node can never reach are removed automatically. Every 30 minutes, infrastructure anchors and every service Frankfurt sees failing are re-checked from six continents through the Globalping probe network. Any HTTP answer below 500 counts as reachable. Infrastructure is inferred from public DNS, network ownership (ASN) and the AWS / Google Cloud published IP ranges; city locations of directly visible servers come from IP Geolocation by DB-IP (IP-to-City Lite, CC BY 4.0, approximate). Worldwide checks: Globalping. Map: Natural Earth.