New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
Exactly!
How do you accomplish everything CF have to offer for free or at a low price in a way that’s immune from going down?
Luckily there are only a handful of Tier 1 providers that are US-based, so there's that.
Cloudflare just went down for a loooooong time, sooooo
not too immune
People put everything including their personal websites behind Cloudflare because they're scared of DDoS attacks, when in reality Cloudflare isn't needed because nobody will waste their money attacking websites with such low traffic, which is how we got to a point like today.
Only if your business is in an industry like finance, gambling, telecommunications, gaming, and government should you be worried but if that's the case you wouldn't be using the free or cheap options.
Never said it is
Any dev knows for something as complex as this, 100 % uptime just isn’t possible
I’d be constantly worried if I used a provider billing for bandwidth overages without anything like Cloudflare in front.
Sounds like an AWS or any other cloud thingy to overcharge thousands of dollars
This is the same as telling, I do not have any websites, so, my websites weren't down. An outage could be (and it is) happening anytime in any service, either small or big. Google had outages, AWS had, Meta, now CF. The reason to use cloudflare is not to prevent being down, but if their service fits you and has value for the money it asks.
I use heavily CF and this is tie only reason that I never had in mind for choosing to use them. I like the convenience to change DNS whenever I want it and not to wait even days for propagation, their free CDN and caching service saves a lot of bandwidth and server resources, their argo/tunnel service is an excellent way to have your domain behind a NAT only server, their ssl is much more convenient in ma cases than LE (it is permanent and works fine with proxy servers), I can use page rules in DNS level, something very useful a lot of times and the list goes on. And this for their free service and only for my small own needs - lot of people benefits much more...
This was a major outage. Cloudflare had a very long time to present such an outage - if any in this scale. The last really big one that lasted ~5 hours and affected over 50% of their global infrastructure was on April 2020 that affected only their API and dashboard, and on Oct 2022 for 6 hours when their cache system broke and returned error in ~5% of the calls.
Sadly this is the problem with the 'centralised' solutions (like this). When everything is concentrated in 'one hand'.
No 'redundancy'. And when no alternative solutions, only one 'big company'.
My services, users also not affected, I don't use them.
Good thing their 1.1.1.1 DNS resolver didn't goes down, I them set at the router level.
RFO: https://blog.cloudflare.com/18-november-2025-outage/
Looks like they architected the system so that it relies on couple services which then can bring the whole thing down for a few hours. It's kinda inevitable when the system grows in complexity, it starts forming internal dependencies and single config propagates to multiple places with unforeseen results. This is why I like and prefer simplicity than complexity. If you aren't serving millions of users, which I think the vast majority of LET users don't, then you don't need to rely on so many external services and various others layers of abstraction.
That's the thing, Cloudflare is apparently very good at marketing because it convinced everyone and their mother to start using their free services for anything from basic domain and DNS hosting (walled garden ecosystem) to DDoS protection for sites that are either already protected from their hosting provider or aren't even remotely popular that it would make any noticeable difference.
I glanced over the CEO's report and found "Machine Learning" mentioned about where the core of what happened is ...
That said, they at least seem to not keep their workings hidden.
It´s really impressive how much of them modern internet depends on Cloudflare edge network, when they go down, it highlights how hard it is to desing true redundancy at a global scale
Not the redundancy is the problem; it is the fact that many chose just one same provider.
When AWS burps half the Internet gets a cold, also.
It kind of is. Making your network fully redundant at all levels isn't typical with LET hosts, they don't own the DC's to have such total control. Then there's geographical redundancy.
Burps cause a cold? Dafuq analogy is that? Stick to automobile or water analogies, please.