Only for roaming customers though. Here in Europe a customer in their home country can only use their home network unless they're calling emergency services. Only when roaming multiple options would be available.
So I wouldn't expect all that much extra load really.
At least you can drown your network sorrows in (relatively) affordable pints of Guinness and the healthy craft beer scene.
Separate note, but I am astonished by how expensive London is - I can pay engineers Bangalore level salaries but they have to deal with Chicago level CoL.
Ireland's the same now :( Cost of living is insane, especially rent. It's still aftershocks from the financial crash in 2007, the whole economy was leaning on the housing development industry and that has never recovered. So there's a huge backlog in new developments.
Diff with Rogers is that they took out their entire network: cellular, home/biz internet, home phone, corporate circuits (including MPLS links), most cable TV, a bunch of their broadcast radio (AM/FM) network just dead dead dead.
Well, their towers were sorta up (as they couldn’t remotely turn them off since the network was down), so if you had a Rogers SIM, a call to 9-1-1 wouldn’t failover to other networks because the device made just enough of a handshake to try and fail on the Rogers network. A flaw in GSM I reckon.
Apparently the workaround was to remove/disable your SIM and hope another network has a stronger signal.
Oh, and the CTO was on holiday and had no idea for a while because… their phone was on roaming with Rogers and therefore dead.
I wonder if Rogers still does planned-in-advance multi-stage potentially-enterprise-breaking updates on Fridays
In a financial company I worked at we would do some of the biggest, riskiest changes at 5pm on a Friday (or Saturday evening if we were worried about impacting international trades). The logic being that we would have the most time to fix things before markets open monday.
Our release window is Saturday morning. All the support people are on, most users are not, gives us 36hrs.
We absolutely do not release during week if we can help it. But we are traditional ERP application so pretty much everything we do is contrary to the HN/modern zeitgeist :-)
Being able to release safely during the week is super important for eg financial services, for fairly obvious reasons.
In trading and market making contexts for instance, we release 100s of times a day — including Fridays. This includes bog-standard infra changes like roleswaps and server rebuilds. The releases that happen on weekends tend to be highly disruptive infra changes, eg unexpected changes to some kind of physical connectivity where we’re not comfortable with carrying weekday risk.
We didn’t explicitly set to to optimise QoL for engineers (the real driver for safe intraday change was being responsive as a business) but not usually being on call on weekends was a big plus.
The '22 Rogers outage, hah. As I recall it didn't affect me at all since I was at home and work in Vancouver all day… but it was a great excuse for not responding to workplace on-call messages which I got in the evening
> Well, their towers were sorta up (as they couldn’t remotely turn them off since the network was down), so if you had a Rogers SIM, a call to 9-1-1 wouldn’t failover to other networks because the device made just enough of a handshake to try and fail on the Rogers network. A flaw in GSM I reckon.
Didn't know that part, amazing.
It sounds kind of like connecting to a WiFi access point which has a broken/non-working uplink to the Internet. Modern smartphones pretty much automatically detect and avoid such APs, and indeed the whole SSID if they need to, but it sounds like the stuck-in-1985 2G baseband layer has no equivalent connectivity check.
I mean, incredibly critical personnel probably should be! There may only be a few dozen such people, but I wouldn't want the added chaos caused in the event of a Rogers outage if I couldn't get in touch with the key decision makers and most critical operations engineers because of the very outage they're meant to fix. And in the e-sim era that is hopefully very cheap and without any real downsides.
Reminds me of recent outages in Russia due to buggy rollouts of Great Russian Firewall aka Sovereign Internet. Were there any state-level infrastructure updates planned recently?
We also have mobile internet disabled/throttled sometimes when there are drone attacks or large international forums. Weak-minded people with Internet dependency like to complain about this online as if their online game is more important than an international forum.
Serious question: who gets to decide that some international forum is more important than residents’ use? - be it games, video calls, or whatever else.
The government has an authority to decide, according to the laws? By the way they also often block roads for security of important foreign guests and cause lot of traffic jams.
Well, yes. It’s the same kind of issue. AFAIK, these things are decided opaquely and do not follow any formal rules - which would make them ripe for abuse.
The GGP seemed to be ridiculing “those gamers who think their ping is more inportant”, but there is no way for a person to tell if the network limits, or road blocks for that matter, are reasonable or an instance of corruption.
Seems to be the norm, unfortunately. I have a day’s worth of emails that were never delivered a few weeks ago due to an issue with Apple’s Hide My Email service, and AFAIK there hasn’t been any statement from Apple on the matter.
Data is working on Vodafone mvno. Can't call Out or text, Can't make calls On o2 or EE either. Edit EE working. Edit all mobiles now seem to be working OK.
Even with a ported number, inbound call routing still heavily relies on the "number range" owner to direct the incoming call to the correct network.
If the original number range owner has their subscriber database go down, they can't do the lookup for the network to direct the incoming call towards, so it can cause disruption. The same is true if the incoming signalling endpoints are unavailable, as the incoming call requests won't be responded to.
> A map showing reports of EE outage reports made to DownDetector suggests that those in London, Birmingham, Manchester and Glasgow are the worst affected.
No. Those are the most densely populated areas of the UK - obviously they appear as bright red spots on the map.
Another Down Detector bullshit article.... it's getting incredibly tiring. Every time a provider (Phone, Internet or even cloud services) suffer issues ALL of them are reported as down.
Honestly I know this sound snarky but it’s 100% true - Three has become so unbelievably bad in East London lately I’d struggle to know if it was affected by this outage or if it was business as usual for them. I could go on about how broken their billing and app and site were but meh… need to change provider.
You reminded me that a couple of jobs ago (I tend to stay with jobs for a long time) we discovered that a local paging tower had failed because one of our techs could not receive pages at home. Our assumption was that we were basically the last people using their service in the area.
I see we’re still using Down Detector as a source for stories, which may as well be called People Are Talking on Twitter Detector. It doesn’t do anything smart, it’s just looking for keywords on social media, sometimes by coincidence this indicates an actual outage.
Was about to say the same. So many of my colleagues looki] for officially confirmed statuses if a cloud provider, AI, blockchain or github starts acting funny and troubleshooting problems thinking it's their own systems. My response is: did you check for people talking on X? About half an hour later there's an official incident.
Yeah, it's far from ideal, but in my experience its accuracy is better than most anything else readily available, including the official status pages maintained by most tech companies.
Yeah, and not only do you get to see if it's down or not (reddit infamously always says it's up even when there are issues), but you also get to see the raw data of reports. Ofttimes I've seen the trend go up and realized it's a very recent issue - even before downdetector itself recognizes it as such.
Human reading > DD reading >> "All our services are operational" when they're absolutely f--ing not.
It's fine in simpler cases but as we're seeing here, is absolutely useless at providing information on failures in complex systems. It appears that it was one network suffering issues but this was then reported as "EVERY MOBILE NETWORK IN THE UK IS DOWN" because people just go "I tried to make a call and it didn't connect". That could be anything from a single cell tower being hit by a truck to a nationwide power outage.
Google can detect flu break-outs much faster than the CDC for example, because people tend to search for symptoms before they let anyone know officially, visit a doctor, etc.
I do think there's a better than average chance that WW3 starts not with an open nuclear exchange as our parents imagined, but instead with a substantial cyberattack which shuts down power, water treatment, communications, hospitals, public transportation, etc. This might even be deniable / grey zone for a few hours or days while the belligerent parties use the chaos to accomplish some Blitzkrieg style attacks.
They did try, but ukraine were suspicious something was going to happen (massive army suddenly forming near their border) and they had spent time securing stuff
They did, to some extent, though not to the degree I described.
As an American, I would be more worried about China than Russia though. They makes a lot of our hardware and firmware, giving them plenty of chances to embed killswitches and zero-days. They have possibly the most successful industrial espionage program in the world, giving them the opportunity to find vulns in other systems and embed agents inside critical platforms. They have deeply internalized the concept of fighting where their enemy is weakest not where they are strongest, so they have likely invested in attacking the American military at home rather than on the field.
That doesn't seem to be stopping Russia from engaging in the energy war, where they're "wasting" incredibly valuable and rare precision missiles attacking substations and power generation.
So far, this has been only moderately successful in impacting the military, because most of Ukraine's military production is not actually located in Ukraine, and because they've gotten quite good at repairing their electrical grid.
This was generally true of allied strategic bombing campaigns in WWII as well. Simple adaptations like building walls between sections of the plant to require more bombs to attack any given target, hardened shelters for skilled employees, and staging parts outside of the plant enabled some targets to maintain better than 50% uptime during aggressive bombing campaigns. Look into the Oil campaign[1] for more details.
It is interesting to see how precision missiles and cheap drones may change this in the future.
Only if they absolutely need it. Nobody would spoil an asset like that. Maybe they would turn off mobile in Taiwan if they control their network. I didn't check which technology provider they use.
No mobile network can be an inconvenience but who has to respond will have the means to communicate no matter what. Furthermore every single common person will feel a personal level of danger and they won't simply shrug about the destiny of a remote island somewhere on the map.
tbh breaking public internet/phones will probably be done by a local government trying to do something nasty. Stop people coordinating and turning up places they might get in the way.
I think my tin foil hat was askew. There. All better.
https://www.bbc.co.uk/news/articles/cnvmvqrnq7go
> A spokesperson from BT, which owns EE, apologised and said the firm was "currently addressing an issue impacting our services".
> Vodafone and Three have confirmed to the BBC they do not have network issues.