Is it us or the internet? Routing and outage signals

When a site is unreachable for some people, the question is often whether the problem is your server, your DNS, your hosting network or the internet in a region. Signalhouse chains four tools to answer it.

The chain

Open the template Is it us or the internet?. Its suite is DNS & certificate, BGP routing, Network (AS) and Outage signals. The checks depend on each other:

1. DNS & certificate finds the site's addresses.
2. BGP routing takes the first address and looks up the announced prefix: origin AS, RPKI validity and how many route collectors see it.
3. Network (AS) takes that origin AS: holder, registry country, upstream and downstream neighbours, announced prefixes.
4. Outage signals correlates public signals (IODA, RIPEstat BGP visibility, a live probe) for the domain in the look-back window.

Signalhouse orders the jobs automatically. If you run Network (AS) alone, it adds the routing and DNS checks first when no earlier results exist.

Reading it

On the Target view, the card Path from the domain to its network shows the chain at a glance: domain → address → BGP prefix → origin network → network → outage signals. Each check below has the details; an RPKI invalid origin or a prefix that is not announced turns the routing check red.

Watch countries and networks

In the Lab, Outage watch keeps a list of country codes (DE) and AS numbers (AS3320). Add the origin networks of my targets adds every origin AS found by the routing checks. Check the watch list runs the outage tool for each entry through the same paced queue.

The tool decides whether signals count as a disruption. Short BGP bursts below its threshold are listed as signals below the outage threshold, not as incidents.

Limits

All live checks run from one server location. A site can be reachable from there and still fail elsewhere; the outage signals are public aggregate data, not measurements from your users. The look-back window (default 48 hours) is in Project settings.