Internet spent Q4 '25 losing fights with cables, power, and itself
- Reference: 1769442978
- News link: https://www.theregister.co.uk/2026/01/26/the_internet_spent_q4_2025/
- Source link:
Cloudflare's calendar Q4 snapshot of global connectivity reads like a reminder of how often the internet still gets cut off for very ordinary reasons. Across 2025 as a whole, the company tracked more than 180 significant disruptions, with the final quarter dominated by cable damage, power problems, and routine operational failures.
There was just one confirmed government-directed shutdown during the period. Tanzania saw a sharp drop in internet traffic on October 29 as violent protests broke out during the country's presidential election, with traffic falling by more than 90 percent. Traffic returned briefly before declining again, and routing data pointed to throttling rather than a clean shutdown.
[1]
Most of the quarter's disruption, however, had nothing to do with politics. Cable cuts continued to do what they always do. In Haiti, Digicel was hit twice after international fiber was severed, sending traffic on its network close to zero before repairs were completed. Damage to the PEACE cable disrupted international connectivity for Pakistan, while faults on the West Africa Cable System caused repeated outages across Cameroon and nearby countries.
[2]
[3]
Electricity failures proved just as effective at knocking the internet sideways. The Dominican Republic lost a big chunk of its internet traffic after a transmission line outage spiralled into a nationwide blackout. Kenya ran into its own problems after a fault on the regional grid link with Uganda, leaving connectivity sluggish for hours even as the lights came back on.
Armed conflict also left a visible mark on connectivity. Russian drone strikes damaged energy infrastructure in Ukraine's Odesa region in mid-December, cutting power and pushing local internet traffic down by as much as 57 percent. Connectivity only crept back over the following days.
[4]
Internet traffic in Jamaica dropped sharply after Hurricane Melissa made landfall in late October and remained below normal for several days. Cyclone Senyar caused flooding and landslides in Sri Lanka and Indonesia that disrupted power and telecommunications across multiple regions.
[5]Microsoft 365 outage drags on for nearly 10 hours during bad night for North American infra
[6]Akamai CEO wants help to defeat piracy, reckons he can handle edge AI alone
[7]Snowflake buys Observe to make 'Days Since Last Outage' counters obsolete
[8]When the lights went out, and the shooting started, Y2K started to feel all too real
Some of the quarter's outages had far less dramatic causes. [9]Vodafone UK managed to knock itself offline in October , with traffic across two autonomous systems falling to zero and no public explanation from the company. Fastweb in Italy ran into a DNS problem the same month that cut traffic by more than 75 percent for several hours. Providers in Israel and Indonesia saw similar DNS failures, underlining how a broken resolver can look, to users, exactly like the internet disappearing.
Cloudflare added its own entries to the outage list, [10]with two internal incidents during the quarter that made sites and apps unreachable . AWS and Microsoft Azure had regional problems too, causing knock-on performance issues for customers who've built on their platforms.
The Q4 data shows little has changed about how the internet fails. In the end, Q4 offered few surprises. The internet kept falling over in the same places it usually does, and when it did, the damage tended to travel. ®
Get our [11]Tech Resources
[1] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/networks&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2aXedxVep7AKPD7pP5gdJCwAAAA8&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[2] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/networks&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aXedxVep7AKPD7pP5gdJCwAAAA8&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/networks&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aXedxVep7AKPD7pP5gdJCwAAAA8&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/networks&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aXedxVep7AKPD7pP5gdJCwAAAA8&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[5] https://www.theregister.com/2026/01/23/microsoft_365_outage/
[6] https://www.theregister.com/2026/01/20/akamai_ceo_tom_leighton_piracy_interview/
[7] https://www.theregister.com/2026/01/09/snowflaketo_acquire_observe/
[8] https://www.theregister.com/2025/12/29/on_call/
[9] https://www.theregister.com/2025/10/13/vodafone_outage/
[10] http://google.com/search?q=cloudflare+outage+theregister&rlz=1C5CHFA_enGB905GB905&oq=cloudflare+outage+theregister&gs_lcrp=EgZjaHJvbWUyBggAEEUYOTIGCAEQIxgnMgwIAhAjGCcYgAQYigUyFQgDEC4YFBjHARiHAhixAxjRAxiABDIKCAQQABixAxiABDIKCAUQABixAxiABDIPCAYQABgUGIcCGLEDGIAEMgoIBxAAGLEDGIAEMgoICBAAGLEDGIAEMgoICRAAGLEDGIAE0gEIMzQwN2owajeoAgCwAgA&sourceid=chrome&ie=UTF-8
[11] https://whitepapers.theregister.com/