Internet circuit / WAN failure
Office is offline. Diagnose: ISP vs onsite hardware vs config — and prove it before paying for an emergency engineer who finds 'no fault'.
Indicators
- All services dependent on cloud are down
- WAN port LED off, or up but no traffic
- Ping to ISP gateway fails
- Failover circuit didn't take over (if configured)
Likely causes
- ISP outage (regional, planned works, or unannounced)
- ONT / NTU device failed (FTTC/FTTP)
- Router WAN config corrupt or interface administratively down
- Cable damaged, power to ONT lost
- DDoS or saturating local-loop traffic
Diagnostic steps
-
Check ONT/NTU status lights vs ISP-published normal state
-
Console into router: show interface WAN — physical up/down, errors, drops
-
Test ISP gateway reachability from inside the router (ping, traceroute)
-
Check ISP status page; call ISP support — establish ticket, capture circuit reference
-
Failover: confirm secondary WAN took over (or didn't). For SD-WAN: check link-quality probes
-
Provide a tether / 4G fallback for critical staff while ISP works
Resolution path
- Confirm ISP fault vs onsite
- Engage ISP with circuit reference
- Activate failover or 4G if available
- Restore primary, validate failback works
Prevention
- Diverse second circuit (different ISP / technology) for critical sites
- SD-WAN with monitored link-quality probes
- 4G/5G fallback router on critical retainer sites
- ISP escalation contact pre-agreed
Tools
- Router CLI / dashboard
- ISP status page (use 4G to access it)
- 4G/5G dongle or hotspot — tether-test bypasses local equipment
- downdetector.co.uk for cross-customer correlation
- BGP looking glass (for businesses with ASN)
References
- Ofcom — UK ISP service quality reporting
- Vendor SD-WAN failover documentation