Vodafone keels over, cutting off millions of mobile and broadband customers
- Reference: 1760374176
- News link: https://www.theregister.co.uk/2025/10/13/vodafone_outage/
- Source link:
The outage began on Monday at 14.25 BST, and 30 minutes later it peaked when monitoring website Downdetector.co.uk reported that almost 140,000 customers were unable to use the service. One Register reader, Steve Maxted, noted that "Vodafone is down. Hard! Everything. Landline internet, mobile internet, website... It's not just DNS, as ping also fails."
Ah, yes, that old standby – it isn't DNS – it can't be DNS – until it is. However, something more serious appears to have affected the telco. The Register contacted Vodafone for more details, but the company has yet to respond.
[1]
Another reader told us: "One of our multi-network roaming SIM providers just warned us that 'we are currently aware of an ongoing issue with the Vodafone UK Network. This seems to be affecting a large number of consumer devices across the country.'"
[2]
[3]
Our reader's phone registered a strong signal, but data appeared to be broken, and while an inbound call worked, "trying an outbound call caused my Pixel 7 to lock up completely and do a very slow reboot – first time I've seen that."
[4]Microsoft 364 trips over its own network settings in North America
[5]Kubernetes kicks down Azure Front Door
[6]Subpoena tracking platform blames outage on AWS social engineering attack
[7]Texas man accidentally shoots cable, brings internet down
Less than ideal. Readers also reported that broadband was affected by the outage, which is odd since we would have expected cellular and internet connectivity to be largely separate. Hopefully, there are no single points of failure lurking within Vodafone UK's infrastructure.
Vodafone and Three recently announced a deal whereby customers of one could use the other's network. At the time of writing, Three does not appear to have any issues, so it would have been a good time for a network switcheroo. However, as one reader observed, the problems did not seem to be with the signal strength but rather with something else within the system.
A spokesperson at Vodafone told us:
[8]
"This afternoon, for a short time, the Vodafone network had an issue affecting broadband, 4G and 5G services. 2G voice calls and SMS messaging were unaffected and the network is now recovering. We apologise for any inconvenience this caused our customers."
Happy Monday. ®
Get our [9]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=2aO12d_I6YXjCHBB7pfQx0AAAAFg&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=44aO12d_I6YXjCHBB7pfQx0AAAAFg&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=33aO12d_I6YXjCHBB7pfQx0AAAAFg&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[4] https://www.theregister.com/2025/10/10/microsoft_365_na_outage/
[5] https://www.theregister.com/2025/10/09/kubernetes_azure_outage/
[6] https://www.theregister.com/2025/10/02/subpoena_tracking_platform_outage_blamed/
[7] https://www.theregister.com/2025/10/01/texas_internet_outage_gunshot/
[8] 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=44aO12d_I6YXjCHBB7pfQx0AAAAFg&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[9] https://whitepapers.theregister.com/
Seems to be back now. They've obviously found 50p for the meter.
Oh my days
My first bedsit when I moved to London had a 50p meter!
Stick a coin in and turn a crank handle.
Forgotten about that.
Re: Oh my days
How many condoms did it dispense?
Some students I knew…
Worked out that they could make coin sized ice cubes that would work just as well.
Re: Oh my days
Up until earlier this year the communal washing machine and tumble drier in this block of flats ran off 20 rappen coins in a meter like that. It was 100% reliable.
Our landlord replaced it with a card-based system that seems to die often and often won't read the cards...
50Pence!! twas only a shilling back in the day!
t'were thruppence or sixpence in my day :( (feeling ancient)
We had to wash our own clothes in my day, in cold water too with no soap... or water.
Three is fine...
... on both my phones.
Businesses may have merged but I think the infrastructure is still mostly separate.
Where's the redundancy?
OK, both their mobile and broadband networks went down at the same time suggesting shared infrastructure, OK their status page went down as it's probably connected into the back end, OK their web site went down later probably due to load, OK their phone lines went down as they're probably using VoIP by now, OK their social media team went down as they're probably on the same network and no one is allowed to use another device or whatever... OK it seems to have affected other countries and networks as they share infrastructure, OK people that pay for a backup line got nothing as it's the same back end, OK their press comms was extremely basic and uninformative as the key people were probably too busy to send out clear messages.. but... why do they seem to have single point of failure affecting the whole country?
My broadband ISP Onestream also fell over as they also use Vodafone for their infrastructure. DHCP was still providing my router with an IP address which would change after a few reboots but traceroute wouldn't get any further that the gateway, and pings to anything out on the internet would result in 100% packet loss.
It back online now, DNS was a bit flakey at first but seems okay now.
They unannounced most/all of their BGP routes.
[1]Cloudflare announced IP address space
By the look of it, anyway, and that also included their own nameservers.
[1] https://radar.cloudflare.com/routing/as5378#announced-ip-address-space
Re: They unannounced most/all of their BGP routes.
Thank you - I knew I could rely upon a learned Register reader to rapidly identify the cause of this afternoons fun and games.
I bet Vodafone don’t fess up …
SMS wasn’t working for me. Voice calls for some reason took around 15 seconds before it started ringing but they did work. Absolutely nothing else worked. Trace routes couldn’t see anything behind my home router, or outside of my mobile on 4g.
Vodafone mobile and broadband customer here. Interesting that my landline phone (now over broadband) still had dial tone. Also that mobile voice still worked. But everything else was well broken.
What a s***show
Fiber optic AND cellular went down. I just had time to stroll to the Vodafone store where the staff told everyone it would only take 1 hours and was due to the engineers "merging the 3 and Vodafone infrastructure". 5 minutes later the team turned the lights off, closed the doors and went home.
By the time I set up my backup cellular it was back on.
Still, is a 50% monthly discount for an all Vodafone package still worth it?
BGP
According to Cloudflare Radar, AS5378 (Vodafone UK) stopped advertising all routes for a period this afternoon. That would explain it.
Question is, was it a mistake, malfunction or malicious?
Re: BGP
We've all had an "oh sh*t" moment, you haven't done IT engineering if you haven't, just not all of us take a major ISP offline.... i did take an entire National Insurance's website offline for a period once though... something related to 1066. It wasn't because of BGP however!
Annon for obvious reasons
First thought - CA cert expiry
Second thought - advertising routes snafu
Well done Reg slooths - looks like Vodaphone disappeared themselves, look forward to the root-cause analysis!
Hopefully, there are no single points of failure...
Of course there is.
It's called management.
My connection was hit.
It connected to the net but no data was forthcoming, the system reporting a DNS resolution failure. As it wasn't my end, I did the sensible thing and made a cup of tea.
Tech is less resilient because it is complicated. A single component failure/unwise tweak can take anything down. This stuff will happen.
traceroute to venerable 8.8.8.8 went whackamole round a dozen 192.168 addresses outside our network before barfing. Routing went mad but did occasionally work.