Hacker News new | past | comments | ask | show | jobs | submit login
Hetzner raises prices while significantly lowering bandwidth (US) (adriano.fyi)
286 points by acaloiar 6 hours ago | hide | past | favorite | 218 comments





As a Hetzner bandwidth enjoyer affected by this, this is why (HN cough) multi-cloud/dedi k3s is great, because if you get rug pulled you just migrate to another provider with better prices.

That said, $1/TB for bandwidth overage seems pretty fair. I empathize with the complaining but if the new price is such a ripoff everyone should be recommending what cloud VM provider they're migrating to for a better deal.


I use OVH (VPS’s specifically), which offers unlimited bandwidth. In my experience they’ve been both reliable and affordable which is a rarity. I run a few applications that require high amounts of bandwidth, so silly caps like the ones that Hetzner are imposing are a non-starter for me

That's what a friend said also. If you look beyond the marketing material, hoever, the ToS says:

> OVHcloud reserves the right to restrict the VPS Service bandwidth to 1 Mbps (1 Megabit per second) until the end of the current billing period in cases of excessive use by the Client

but it advertises with "unmetered"... so is a meter attached by which they can tell whether your bandwidth use is excessive or not? Would they eat those costs for you?

I checked out some numbers. Quoting myself from chat history:

> it begs the question: what's "excessive"? I dunno but if they charge $5/month for the VPS and, while AWS may be ~1/3rd cheaper [than some other thing], that's still on the order of 70$/month. And AWS has insane economies of scale working for them, maybe their cost price is $7/month if they don't need to have a competitive price but that's still a loss then

> I bet you'd win the lawsuit where [OVH] falsely advertised with unmetered 500mbps and a terms of service saying "excessive", so when you transfer 2 TB/day on a connection advertised to be capable of 500mbps×24h = 5.4TB/day... that's reasonable right? But then you're having a lawsuit over a 5$/month VPS


> is a meter attached

Probably not by default, but if your usage starts to saturate their network switches they’ll add one, to figure out who’s disrupting everyone else’s QoS.


> is a meter attached

Yes, of course. Having flow data (or monitoring ports/interfaces) for traffic engineering and management is pretty essential, not least for determining when capacity upgrades are needed.

I understand both sides of the argument here. The idea of offering "unlimited" is appealing because most users of a typical 2GB RAM virtual machine (as an example) consume less than 1TB of bandwidth per month. Offering unlimited bandwidth removes the hassle of overage charges/billing queries and eases customer concern/friction. Both sides benefit from this.

However, on the other hand, is it reasonable for a $5/month virtual machine customer to use 1Gbps 24/7/365, potentially consuming $100–$200 worth of bandwidth?

Should providers avoid offering unlimited bandwidth unless it's truly unlimited? From an engineer's perspective, yes, I agree. But this stance also risks degrading the experience for the 99.5% of "normal" customers—those who don’t exploit this simplification of "free bandwidth"—just to address a handful of users who take full advantage of it.

It's tough, so IME most such providers leave something in their terms that allows them to intervene in extreme cases but typically exercise restraint in doing so, usually only doing it manually if they notice that 'extreme' usage is damaging other users experience e.g it's serious and prolonged usage.


Doing billing like this is awkward as you just don't know what you are going to get. And, in a world where most people are in fact using small amounts of bandwidth, what you have done is caused people who use small amounts of bandwidth to pay MORE than they should, as they are effectively paying the price of the bandwidth for the average user: if you use less than the average, you are subsidizing the people who use more than the average.

Meanwhile, in an ecosystem where everyone isn't already being ripped off with overly-expensive bandwidth, if an ecosystem-level event happens that causes the average user to suddenly use more bandwidth, the service either has to raise rates for everyone or they have to start claiming some uses of bandwidth are "egregious".

The result is then that, to defend the small-scale user from paying even more than the too much you are already charging them (as they are subsiding the larger users), you suddenly start doing traffic analysis with price discrimination by use case, and network neutrality goes out the window :/.

The real reason any of this works is just that people in fact aren't being charged fair prices most of the time, and these unlimited plans let the provider hide that from all involved. If everyone were charged a fair price, not only would heavy users pay a lot and light users pay LESS than they often do today, but everyone would be paying little enough that this idea that it is a big customer "concern" goes away, the same as it is for electricity or water: except in extreme circumstances, no one frets over sudden utility overages.


OVH is the most set and forget experience I've ever had. They email me maintenance notices 3 or 4x a year, but I don't think I've ever had any downtime. It just stays happily humming along for years. I think I pay something like 60 a year for it.

I've only ever been on OVH and was surprised to discover a few years ago that bandwidth is not only unlimited but also costly at most other hosting companies (including cloud ones).

OVH and reliable, in the same sentence? They're cheap, so suitable for projects you don't mind going poof.

Personal anecdote. A few years ago, I lost a lot of sleep on a domain renewal at OVH. Their incompetence was mind-boggling. A less common tld was the only slightly challenging bit. After a week of calling and emailing, and on the verge of the domain lapsing, I gave up and sent someone to the tld registry with cash.

Also, do search for OVH SBG2 should you have missed that.


OVH's infrastructure is absolutely very reliable.

The pain begins when you need support. Just like you, I have lost a lot of sleep over domains held hostage by their incompetence (for almost a year in one instance). Lesson learned, never use OVH for domains.

The support for their dedicated servers is just as bad, mind you, but short of a hardware failure you really don't need them. I have several years of uptime on all my current services.

So for personal projects their vps/dedicated is still a fantastic value.


> OVH's infrastructure is absolutely very reliable.

Well except that time one of their datacenters burned down, likely due to insufficient fire suppression, and the data backups were also lost because they kept them in the same building as the originals.

https://www.datacenterdynamics.com/en/news/ovhcloud-ordered-...

These reports criticize OVHcloud for having no fire prevention system and no power cut-off on the site, for using wooden floors, and for a free-cooling design that created airflows that spread the fire. The reports also say that water was detected near electrical systems before the fire broke out.

It takes quite a while to regain trust after shitting the bed that badly.


> data backups were also lost because they kept them in the same building as the originals.

I asked that question Hetzner support and they admitted doing the same


Better to be honest about having weak backups so users can plan accordingly, than to lie about how safe the data is and lull users into a false sense of security...

the OVH contract relating to automatic backup stipulates that a backup of the VPS server is scheduled daily, exported, and then replicated three times before being available in the customer space, and that the storage space allocated to the 'back-up option is physically isolated from the infrastructure in which the VPS server is set up.'


They can be terrible. But it is dirt cheap. So if you use their stuff in a way that you can recover somewhere else if they have issues, you save a lot.

As you mentioned I would stay away from them for things like domain hosting. Just use them for cheap compute, etc.


Yeah,just hope you never have to interact with support

Shout-out to Microtronixdc, my colocation provider. For their $55/month/U colocation package, they provide unmetered gigabit for no additional cost.

https://microtronixdc.com/


I see terms for hosting and domain registration and managed databases, but not colocation. Do you know if they really not care if you actually use that connection to the stated allowance?

https://microtronixdc.com/products/colocation

Their big selling point is that they truly provide unmetered connections.


Bandwidth for these providers is super cheap. The main cost is power consumption.

Exactly! I had a three node k3s cluster hosted on OVH. When I decided to switch to Contabo, it was as easy as adding the three Contabo nodes to the cluster, and then removing the three OVH nodes (plus updating some DNS rules). It was the easiest and simplest migration I'd ever done. All my services and data just moved automatically and mostly with zero downtime. The only service that experienced a little downtime was Plex, which as I understand does not support high availability. If I ever find a cheaper host, I'll simply switch over. No hassle and no vendor lock in.

As someone who's already using the dedicated server for a ton of things, I have been really grateful. But now, I have a new question, are they going to do this to their dedicated servers as well?

This change only affects VPS services in the US, and for what it's worth, Hetzner does not offer dedicated servers in the US.

I am very ignorant here, so my apologies, but...

When someone runs a dedicated server these days, does this mean a one-off linux install? Or is this more likely to be a docker install so that it's portable?


It's an actual entire machine given to you. I remember there were a few options for me from Ubuntu, Debian to Red Hat to choose from, but all of them would also have preconfigured system users and some level of administration done by the provider.

But other than that, it's an actual bare metal machine and I installed Ubuntu on it and threw in a giant heap of services that have been running on it for more than a year now.


If you could rewind the clock, would you have started setting it up any differently, like in a container?

I am just curious what your options would be now if you wanted to migrate. Would you just copy your bash history to a local text file for reference, and then repeat the steps on a new server?


Generally even in containerized deployments, you run one container per service/process. You wouldn't run everything you’d run on one box in one container.

I definitely recommend using docker compose or similar even in a one node deployment versus just installing and running things on the host linux system like it’s still 1998. Having a single directory to back up and a single file defining all of the services that can easily be redeployed is very convenient.


Replying here as your other question is at max thread depth:

A non virtualized Linux install isn't more locked in than a docker install, as for a bare metal server you are choosing your own OS. I have done the docker thing on a bare metal server, but that's because I wanted to run multiple services on it and isolate them operationally.


> A non virtualized Linux install isn't more locked in than a docker install

Again, sorry for my ignorance here, but if not virtualized, how does one move hosting providers otherwise? My experience is limited to either running all the bash commands in an install readme, or installing a docker image.

So there must be something in-between, to recreate a linux install elsewhere?

> Replying here as your other question is at max thread depth:

btw, you can click on the time of the post, and reply there when there is no reply link in the main thread.


Using dedicated servers doesn't mean you're not using virtualization - it just means you're the one managing it. You control the hypervisor and the vms running on top of it. Because of that, you're actually less tied to a specific hosting provider since you're not reliant on their APIs to set up and manage your infrastructure.

Even if you're not using virtualization there are still plenty of ways to migrate your servers.

One of the most common approaches (which was the thing before docker took over) is managing servers with an IaC approach using tools like chef, puppet, ansible, saltstack etc.

With IaC you define your entire infrastructure in configuration files and deploy those configs to your host. It's a bit like docker swarm but for managing physical and/or virtual servers instead of containers.

Another popular option, often paired with IaC, is to create your own pre-configured *nix images tailored to your needs. For example, you might have specific images set up for your load balancers, db servers, file hosts, or other roles in your stack.

I've worked at a company where we handled migrations using dd. Technically that's also an option. Wouldn't recommend it tho.


If the server hoster supports it (Hetzner apparently does), you can enable KVM and install a previously prepared image. If the server hoster & hardware supports it, you can login remotely to the server management interface (like HP iLO) and install an image this way.

If you don't have above options or simply don't want to do it this way, you can also bootstrap via SSH. But instead of manually typing in shell commands, you will automate it in some way with custom scripts and/or tools like Ansible.


>So there must be something in-between, to recreate a linux install elsewhere?

There's a class of tools like chef, ansible, puppet.

Or you can just package your services into debs and run a Debian repo to install them from, same as the base OS

Possibly there's something closer to dockerfiles out there by now as well

>btw, you can click on the time of the post, and reply there when there is no reply link in the main thread.

Cool, thanks!

Makes me puzzled about the point of concealing the reply links, I guess it just adds a bit of friction?


It's a fire extinguisher feature; for when the thread has too many, too fast replies.

Talos Linux, Flatcar Container linux...

Isn't a Docker image an OS image and a bunch of shell commands?

You can use a configuration manager like Puppet.

https://en.m.wikipedia.org/wiki/Puppet_(software)


For Hetzner you rent actual physical servers. There is nothing virtualized there, it's real hardware without abstractions.

Thanks, I guess what I meant to ask is that is it normal for people to virtualize their own dedicated server these days by default?

Would this be a best practice to avoid hosting vendor lock-in?


There's no lock-in possible. It's a bare metal Linux machine, you do whatever you want with it; you can replace it with the PC under you desk if you want.

If you want to run k3s, k8s or docker, you can, but personally I find those too complicated. NixOS is much easier to deal with, and achieves the same result.


Yeah, it's a one-off install. In my case, I did Proxmox[1] for a while with VM's and LXC's, with some of my VM's and unprivileged LXC's running Docker (compose) too because it made the installation of said software easier. It's great, but I'm moving over to switch to Debian with Incus[2] instead of Proxmox. Just for fun mostly.

[1] https://www.proxmox.com/en/ [2] https://linuxcontainers.org/incus/introduction/


Is it an option for you to just move the servers to one of Hetzner's Europe location? I guess a lot of high bandwidth applications don't require low latency.

>multi-cloud/dedi k3s is great

As someone with the exact same setup, thank you for strengthening my confirmation bias.


DataPacket, ReliableSite, HiVelocity, and FDCServers among others have great bandwidth prices.

Can vouch for HiVelocity under the previous owners anyway, not sure what's going on over there now. My companies launch was bandwidth and compute intensive and they handled it well.

It does feel like a case of the Costco hotdog going up to $2 followed by "grrrr. Thats it! I'm..... going to keep buying it because it is still damn cheap!"

I remember reading something about the Costco hot dog story, quite funny IMO, here's what I just found from 2018: "I came to (Jim Sinegal) once and I said, ‘Jim, we can’t sell this hot dog for a buck fifty. We are losing our rear ends.’ And he said, ‘If you raise the effing hot dog, I will kill you. Figure it out.’ That’s all I really needed. By the way, if you raised (the price) to $1.75, it would not be that big of a deal. People would still buy (it). But it’s the mindset that when you think of Costco, you think of the $1.50 hot dog (and soda)." [1].

Turns out Costco has a new CEO this year, and again the hot dog topic came to light apparently, lol. This article is from 2024: "'To clear up some recent media speculation, I also want to confirm the $1.50 hot dog price is safe,' Millerchip said." [2].

Sources:

[1] https://www.425business.com/news/costco-ceo-craig-jelinek-on...

[2] https://www.usatoday.com/story/money/shopping/2024/05/31/cos...


These stories are marketing mythology.

Think about. Imagine they sell 100 hot dogs per hour. A $0.25 difference means $25/hr in a store doing many orders of magnitude more revenue each hour.

It’s nothing. The way they play it up in the media gets a lot of attention and builds goodwill, but it’s entirely meaningless to their bottom line.

It’s amazing that people eat these stories up, though. I’ve heard so many people repeating this story as if it’s some amazing secret.


Wait what? Did they actually raise the hotdog price?

I never buy the hotdog but a price raise is indicative of bad times to come.


Costco hotdogs have been underpriced for decades. It's an intentional choice, a loss leader. It raising would be mostly meaningless.

But no, last I looked a week ago or so it was still the same price.


It’s a warmed hot dog, a bun, some self-serve condiments, and it’s served by people who can do hundreds of them an hour.

$1.50 might very well cover the cost, but even if it’s a loss leader then it’s entirely negligible.

The mythology around the hot dog price makes people think they’re getting a screaming deal, though. It’s perfect marketing material.


I created an account with Hetzner earlier this year, and confirmed my Credit Card with them, but a few second later, they auto-suspended my account before I could log in.

I emailed support, and they bluntly told me to create a new account and this time use real information... Needless to say, I bought compute elsewhere.

I don't know how they're still in business.


I think this might be a cultural thing. HN, SV, and the market for IaaS/SaaS products is a bit of an American monoculture, where "the customer is always right" and there's a strong desire to make the customer happy. I think this is mostly a good thing and especially a good way to build early stage companies, but in my experience it's less present elsewhere.

In some places companies are happy doing their own thing, don't need every customer, don't need to be everything to every customer, and won't fight for business in the same way. Does that limit them? Maybe? But I suspect not enough to be a problem most of the time.


It's not a cultural thing to accuse customers of committing fraud their first interaction.

Not being an absolute insane jerk off is a good expectation of people.


I'm not advocating for anyone being a nasty person, but there are significant cultural differences again for what it means to be rude. In Japan the cultural expectation is that saying "no" is rude in a customer service situation, which is far beyond the expectations of most of the world. This is particularly tricky when the answer is actually "no".

If you're used to American customer service, you may find European customer service to be blunt or curt, and many people would perceive that as rude even though it is not intended that way. Again, if they aren't trying to win every customer, this isn't really a problem.


>This is particularly tricky when the answer is actually "no".

If the answer is actually "no", then the customer service person will tell you it's "a little difficult". You're supposed to understand that that means "no, we can't do it". Of course, many foreigners don't get this, so then they'll change gears and just say "it's impossible".

>If you're used to American customer service, you may find European customer service to be blunt or curt

I don't see accusing your customer of fraud as simple bluntness or curtness.

>Again, if they aren't trying to win every customer, this isn't really a problem.

If they don't want to expand their business outside of Europe, I guess that approach is OK.


"Try again, and this time provide real data" would be considered very rude even in Germany.

Without reading the actual wording sent, and knowing which culture the sender was from, this is all just speculation. I'm more interested in challenging the assumptions that we have about our own expectations in customer service being universal. From the rest of this thread it sounds like there is a commonly held opinion that Hetzner customer service is blunt, and my point is that that may be fine and maybe customers should not expect to be treated in a particular way all the time.

Well, we haven't seen the data.

> It's not a cultural thing to accuse customers of committing fraud their first interaction.

In my experience a lot of German businesses are like that, they'll woosh you away for any slight (perceived or real). So it is definitely a cultural thing. Of course my small sample doesn't mean anything in regards to german culture, but at the very least it proves that this corporate culture is indeed a thing.


Really? In France you're treated most of the time as a potential fraudster first, then maybe when planet aligns as customer in some services like train, tramway, post, banks (the state mandates to justify things like moving and using your own money you already paid taxes on) and various administrations.

> where "the customer is always right" and there's a strong desire to make the customer happy

It’s definitely a culture shock for people from these cultures to encounter businesses who reject their business, whether it’s deemed risky or not worthwhile.

It was eye opening to me to spend time working with cultures where I had to convince businesses to deal with me rather than the other way around. Some people get extremely offended when businesses don’t bend over backward for them, but that’s not going to work when you encounter cultures who operate differently.


> I think this might be a cultural thing.

It's just German bureaucracy. When I wanted to register domains with Hetzner few years back, they asked me to print multiple pages of forms and contract, fill out, sign and... fax it back.


This was between 2000-2010, but when I ran a linux consulting+hosting business our "provision a VPS with a credit card" was where we experienced the largest amount of fraud, by probably 100x. Certainly well over 10x. Might be why Hetzner was touchy?

I was customer for 10 years, but for business I needed a second account. Banned immediately even after I submitted a passport copy. Worked after contacting their support. Still happy customer (with both accounts). I think they're just very strict and get their (un)fair share of stolen credit card or stolen identity signups.

It's always a balance between how much fraud you allow and how many real customers you reject. They set their threshold at an interesting level, but maybe they're happy with that choice.

You're right. But unfortunately this mindset moves us closer and closer to having algorithms exclude people from society and life. And as the stories already told here show, it can happen to any of us for no apparent reason.

It's like, imagine a magic wand that, if waved, would make life a little better for 99% of society, but much worse for 1%. Would it be moral to wave that wand?


Well you got a human answering the support hotline.

I think that alone kinda nullifies the threat of an algorithm. The entire reason why they're such a massive problem is because Google et al. refuse to operate proper support hotlines to help people and even if they do have a support line (Facebook infamously doesn't have one and wants you to go through the courts to contact them), the support staff aren't actually equipped to help people beyond regurgitating canned support page links. You can't solve a malfunctioning algorithm with another algorithm or by forcing a human to behave like an algorithm.

It's not a big secret that the best way to get yourself in front of actual support if GAFAM screws you over is to complain about it on HN because this is where SREs lurk that can actually punt your requests through to people that can look into it.

Hetzner at least gave a direct answer to explain the reason.


How does that human nullify anything if they don't unlock the account? Presuming GP entered real information, which we can only assume at this point

The problem with this mindset (incredibly common in the US; much more so than in the European countries I've lived in, at least) isn't thinking probabilistically, and trying to determine the likelihood of a prospective customer being a fraudster. It's really just that there are only two possible outcomes: Yes and no.

Just by having a third option, most of the downsides of doing the evaluation incorrectly could be mitigated. Of course, that's generally much more expensive (and often uneconomical) than saying no, so it's usually not done.

I've been on the 1% side of things quite a few times due to having new credit and (presumably) various data brokers not knowing every detail about me yet, and the experience really, really sucks.


You're making choices like that every single day. Sometimes with reversed percentages. Even posting this message may be on average a tiny loss for the world. (It's meaningless in the end and burned some energy to give one person a mini dopamine boost)

There’s nothing particularly scary about “algorithms” making these choices since it’s just people at these companies choosing and implementing the algorithms. It wouldn’t get better if those humans weren’t allowed to use algorithms to make these choices since decisions.

Arguably, we wave that magic wand every time we decide not to donate to starving children in third-world countries. Which is pretty often!

The Ones Who Walk Away From Omelas

What type of fraud exactly? You mean like stolen CCs? It feels very medieval as a financial trust system if every little vendor can’t trust payments, even when you pay up front? Like this is in some ways worse than cold hard cash. And then we pay VISA premium on top of that, for the convenience of being mistrusted..

If they pay up front then dispute, the company will suffer extra charges. With enough of reports, their payment processor fee% goes up and impacts all their payments.

Paying up front doesn't really mean much, because if the credit card info is stolen, the actual owner will report the transaction and it will be reversed.

Right but that sucks. So CC companies/banks are simply shifting risk from one side of the transaction to the other. Sure, as a consumer you don’t have to worry because you’ll get your money back. But if the merchant has to worry and reject customers who are “suspicious” to protect themselves then you’re back to square one, except more kafkaesque. That’s why I said cash is better.

I’d rather have $10 permanently lost for a month of VPS than being banned after 5 days setting it up because I’m traveling and my IP is “suspicious”. Which has happened to me.


I wonder if they tried a debit card...

Were you using your real information when you signed up? Were you using a disposable / privacy-first credit card? Did you try to obscure your name or register from a VPN? You're very upset at them accusing you of abuse or fraud, but you haven't actually made it clear whether any of the signals you provided them when signing up were aligned with what they'd usually see from abusive or fraudulent users.

Same with me a few years ago. Their support told me that they didn't want me as a customer. It was my first interaction with them. I swear I'm just a regular nerd with a credit card :D

They are bizarre.


A credit card? That... makes some sense, then. I'd be wary if anyone trying to pay with one of those as well; they're much too rare, and I've heard it's easy to unilaterally reverse charges.

Hetzner is a German company. Credit cards may be common in the USA, but they're not anywhere else.


They are literally the preferred way to pay in the U.S. for damn near everything.

I’ve also had no problems using them in Britain, Canada, Ireland, Korea…


Never had an issue with credit cards on AWS, Azure, GCP, Digital Ocean, Vultr and Scaleway.

If they don't want any customers outside of Germany, they should just say so.

I haven't done business in Germany or used German banks -- is it hard to get them to reverse charges?

Well yes, you go to court if you can't agree with the merchant on anything. They're the competent authority to settle disputes

When I pay for something, now the other party has the money, that's how money works. If I trust them that little, I should probably be using escrow but this costs extra and so it imo doesn't make sense to pay that fee for every transaction - as one does with this chargeback guarantee thing. Probabilistic societies where you're excluded based on bad odds or credit scores is what you get with that system (I've been on the losing end of that, not because I've ever had any debt (too little money) but precisely because they had no positive data because I never needed a loan, and so you can't pay for stuff in certain countries because they require a credit card)


I guess I assumed that consumer protection laws would mean people could reverse charges more easily than they can here in the US (which we can do easily, albeit as a cardholder benefit, not a matter of law). Interesting.

Always felt like they were in the business of blaming and hating their customers. Cloud providers that nitpick and judge every aspect of their customers’ business details and technicalities are a huge operational risk. This archaic practice is the reason generic cloud orchestration was a must, and it’s just not needed anymore.

I don’t care how cheap they are. You get what you pay for.


There’s nothing about cheap that implies terrible customer service. Or rather, the reverse isn’t necessarily true.

Are Hetzner Europe and US run by different companies or something?

My experience in the EU has been nothing but stellar.


I believe so, so that it does not run foul of the CLOUD Act.

When I ordered a VPS at Netcup (a Hetzner alternative) they called me and asked me for the name of the hotel next to my place to check if I really lived at the address I provided. I guess that if i would have needed to look it up, that is, struggled a bit with the answer, they would have denied me as a customer.

I had the same thing, but this was a business address while I was working remotely and had no idea about the area. Told them as much on the phone while looking the answer up on Google Maps. They just accepted that and opened the account.

Just today, a remote colleague mentioned staying at whatever hotel near the office. I had never heard of it: as a local, I don't tend to use the hotels here!

Seems like a really weird choice of question. Thinking of germans, a bakery seems more likely something they'd know of; perhaps a supermarket works more internationally (I'm sure there's exceptions to that as well). Maybe the distance in driving, cycling, or public transport minutes to surrounding cities could be a universal question to ask


If they can look it up then so can you! Maybe it filters out lazy scammers, but it doesn't sound like solid KYC to me.

In fact I don't feel like a hosting service should need to do this at all. If you pay your bills and aren't on the Stasi blacklist you should be good to go. I don't want or expect the likes of Hetzner to be responsible for policing.


As someone who runs mail servers, I wish they'd be MORE strict to stop the drive-by spammers. Every so often our servers get blocked because someone starts sending spam from a machine on the same netblock.

It might be good point-of-difference for some hosting service: have brutal KYC so your netblock is well regarded.


I thought Netcup either colo'ed at or rented servers at, Hetzner? Maybe this was a ways back or have they always had their own DC?

Did you sign up as a private individual or was it a company account? I think they are extremely strict with private accounts, especially outside of EU. Most individuals only get a few resources for a few bucks per month, which probably isn't worth the business at all.

I know a few people who use Hetzner, smaller EU registered companies, zero issues. Also extending limits is just one click and gets approved after a few minutes.

And I totally get why they have to be strict. Blocked IPs are becoming an epidemic, I stopped using Digital Ocean, because I always got bad IPs, that were blocked everywhere. Also Hetzner is suffering from that. People are renting servers for little money and do bad things with the 20 TB traffic they get.


I signed up and immediately got banned because I was accessing through a VPN, which I think is a common problem others have had. I emailed them and their advice was to stop using a VPN and try again.

That seems fair to me though. If that ISP (whatever the WHOIS resolves to for your VPN) gets a high rate of fraudulent customers, how is Hetzner to know if you're the same fraudster again or a new customer? Especially if you share not just the ISP but even your IP address with someone... we also rate limit logins for example based on IP address. If you're coming from the same origin as an attacker, there's no way for me to tell who's who

Perhaps they could implement some way to do a 1-cent bank transfer from an account with a name which isn't on their fraud list, but as a simple step it seems pretty normal to use your regular internet connection (that ties the IP+timestamp combination to a real-world subscriber) and not some anonymity service

Although it would be nice if we could be anonymous on the web, when abuse traffic is involved (more than posting a comment on some forum as an anonymous source or something), I don't know how that'd be possible to reconcile


Why were you using a VPN?

Apple Private Relay is on by default for a lot of Apple products these days. This is why my account was banned.

Had exactly the same experience. I even provided my real drivers license -- something I normally would never do. Didn't convince them I am a real human being that matches what's on the ID.

I never understood why people think so highly of Hetzner.


> I don't know how they're still in business.

They're growing year after year, so clearly they're doing something right.


Because they provide good service to most customers.

Your case is unfortunate, and I totally understand why you'd be taking you business elsewhere, but probably an outlier.

Otherwise, we'd need to ask the same about AWS et. al. as we've definitely seen more than enough wrong account closure complaints on here.


You also don't know the strain brought by continual streams of fraud that their support has to deal with, all for being a relatively cheap option.

Yep. Same thing happened to me. I was not able to get my account working. Ever.

I have all my stuff on vultr now.


OTOH there is Azure where a very sophisticated Twillio phishing service was hosted. When reported to abuse Microsoft replied they were a valid customer. A week later it finally came down.

I'm not sure what was your concern - Hetzner is obliged by law to make sure their signing up a real person or a company and that their servers are not used for malicious intent.

It's pretty obvious what the concern is, no? Hetzner's process seems to be blocking real people with no malicious intent (myself included) from creating accounts, even when we try to provide proof. Obviously this means we end up using a different provider instead. Though in my experience attempting to do business in Germany, this kind of forbidden-by-default perspective is quite common there, so they probably don't see it as a concern, I guess.

Yes, Germany is an extremely low-trust society masquerading as a high-trust society.

Herzner get flack because they are probably the only German entity that the average HN user has dealings with. Anyone who has had exposure to German bureaucracy will recognise the hostility.


They are still in business because 99.99% of customers are not affected by this.

I'm happy customer for years and will be even with these changes.


I've been reading about how great Hetzner for years but i couldn't get past the sigup page where they require a credit card just to create an account.

Honestly if you are providing computer resources, that's pretty standard now and one of the only lines of defence against abuse

Is that a US thing? I'm reasonably sure I did not give them my debit card.

I mean your transaction probably got flagged as fraudulent (like if your postal code didn't match your card), it's not that mysterious.

I think most online operators that have "spend $3 with us" tiers have to be super vigilant about card transactions, and when you fall into the cracks you're a bit SOL.


It's been ages, but once I got something cancelled due to a site not properly distinguishing between shipping address and card billing address.

Because they're so cheap they have to worry about spammers and other sorts of abuse. It's more cost effective for them to refuse dodgy-looking accounts.

The big question is, were you using real information or did you put in fake info?

They're still in business because they're cheap and pretty reliable. But being cheap means there are things you can't get with them but can with others. For example, you can't pre-pay for the entire year ahead of time.


You can actually pre-pay them. Under the transactions section in your account, it says:

“If you make advance payments by bank transfer to our bank account, the amount will be posted as a credit on your account. We will automatically deduct this credit on your account when we process future invoices.”


They must have changed that. I remember reading on their site at some point that they didn't accept that.

> I don't know how they're still in business.

They have 400 GBit/s of DTAG transit.

Showing that they are rich as fuck without saying it.


[flagged]


What led you to assume that GP misrepresented their identity? The way I read the comment - they put their real info but was accused of putting fake info.

How would you even attempt to rectify that?


Send them a copy of your ID?

doesn't always work apparently https://news.ycombinator.com/item?id=42269262

That's a particularly, but unfortunately good example of how awful they are as a company. The sooner they get pushed out of business the better.

As a customer of Linode, I feel like I'm getting a lot, maybe too much, for the money I pay.

For $5 per month, I have a CPU running continuously near 100% utilization, training and retraining L1/L2/L3-CPU-cache-resident transformers, looking for patterns in futures and options markets.

This kind of extreme resource utilization is becoming more common, and these businesses have to adapt to stay profitable.

I expect Linode to change the price on me, eventually.


Why is this attitude so commonplace?

Servers are worthless to a hosting company without utilization. It’s in their best interest to have them pegged 100%. Like airplanes - they don’t make money when they are empty.

Why do so many in this thread think “I am using 100% of what I pay for! They are bound to change it soon!” That’s not how it works. If I offer you service for a fee, I’m going to allow you to use 100% of that service for that fee.


Completely backwards. A server that is being paid for but not utilized is the best-case scenario for a hosting provider – revenue the same, lower costs (electrical/cooling/etc).

in mba brain, but hosting providers want as much utilization because it means people are going to use more over time.

On the other hand airlines regularly overbook flights, and residential ISPs don’t have nearly enough bandwidth for every customer to max it out.

This is absolutely untrue in both a naive and more nuanced sense. Naively, it's not like an airplane, it's like an airplane rented by the month: the owner of the airplane collects rent whether or not the airplane is flying. More nuanced, many VPS providers don't sell you 100% CPU utilization; they sell you a compute budget. AWS actually enforces this; I don't know about other providers. But if you're exceeding your compute budget and the provider doesn't enforce the budget limit, you can expect the system to break at some point.

I'm not sure why you compare this to airplanes. Airplanes fuel consumption is very expensive if the plane is empty relative to profit. They need to put bums in seats for the flight to be viable commercially.

On the other hand, a flat out rack takes more power than an idle one. So this isn't the same thing at all.


Yeah, the GP comment is so dumb. Most commercial aircraft are owned by a few leasing companies. They would love if lessors didn't fly the aircraft, but paid leasing fees!

No competitively priced VPS* is charging a price that works when everyone is fully utilizing their entire quota at 100%...

It's not nefarious: you pay a bit more than the compute you use would cost on a fractional basis, and in exchange the cost of entry is dramatically reduced, spikes in usage get absorbed, etc.

It's a win for everyone involved unless usage patterns shift and suddenly there's never a surplus to go around. At that point prices will quickly climb to roughly what dedicated resources cost.

(*and frankly it's not just VPS, a lot of cloud services rely on everyone not trying to max out their quota at the same time to even function, let alone profit.)


What's the justification for this approach? Buy an old NUC with some cheap Celeron in it, install Hamachi if you need remote access, and it'll pay for itself in a couple months.

Seeing as they are paying $5 a month, how do you expect buying a NUC to pay for itself in a few months? Where are you finding NUCs for $20 with free electricity?

Any decommissioned office PC from eBay will be faster than $5 linode. For example search for optiflex https://www.ebay.com/sch/i.html?_nkw=optiplex+pc&_udhi=30 They're not too power hungry either if you make sure not to go for i7s

But it'll cost $100+/yr in electricity.

You have a CPU? Or a core, that is 1 of many cores on a system where your VM can be and will be pre-empted by the host node’s scheduler without your VM having any control or insight into the process?

Off topic, your website is interesting, I wish I knew how to interpret it "commitment of traders". Maybe there is a blog post somewhere?

Hetzner, in volume, is about 5x cheaper than limped for my workloads. Limped is 2-3 times cheaper than AWS, but AWS has a few things going for it that make it worth it for some workloads.

It's most likely a vCPU. So even the caches are shared.

I don’t find that to be all that impressive.

I’m immediately saving money with the server I built out of mostly used parts and threw in my closet compared to VPS solutions.

The only reason it’s near 100% utilization is because $5 VPS instances have barely any computing power assigned to them.

For the same price as one game server I’m running something like 5-8 VMs at once. I can utilize 128GB of RAM and 6/12 real CPU cores (Ryzen 3600).


What is your annual electricity cost for that server in your closet?

Completely unrelated, but I’m surprised how many people actually use the Ryzen 3600, from desktops to servers, it seems to be everywhere.

Don't think we're close to that point yet. You can still get the same server for free from oracle free tier if you're willing to put up with god awful enterprisey control panel.

Also that linode CPU is virtualized (i.e. at least some of that cache is shared).


Oracle has people jumping all kinds of hoops to get that service too. Just like Hetzner. Took me a few tries with different credit cards then tried to get one for my friend with their card and nothing would work. Great free service though. That ampere vm with 24 gb ram is quite capable.

Couldn't you just buy a used decent PC and run this on its CPU + GPU and get faster/better results?

I have been a fan of Hetzner for many years. $1/terabyte overage charges seems reasonable.

I live in Arizona, but have always rented Hetzner servers or VPNs in Germany. I imagine the service is much the same using servers in the USA?



Lots of dupes of this one today

107 points/50 comments https://news.ycombinator.com/item?id=42264668

88 points/38 comments https://news.ycombinator.com/item?id=42264789


What are some examples of applications people are running that:

1. Require 20TB of bandwidth / month

2. That bandwidth can't be shielded by Cloudflare and others?

Is it like... real time video streaming? Gaming servers? I can't imagine a web app getting anywhere close to that.

I run a mid sized NFT art creation website that generates both images and GIF's (https://mintables.club) and with over 100000 users at peak, it only needed about 1.5TB of bandwidth.


I have 2 services that run above those figures:

1. the demo instance of my OSS software which is a bring your own storage Dropbox like UI for SFTP, S3, FTP, and every protocols imaginable: https://github.com/mickael-kerjean/filestash People tend to come in the demo to upload / download tons of stuffs

2. the docker registry for my oss stuff since I was kicked out of the docker open source program and now need to find a new place to store all the images. 10 millions downloads over the last few years, it does add up very quickly way above the 20TB limits if your image isn't super slim and try to selfhost everything


> the demo instance of my OSS software which a bring your own storage Dropbox like UI for SFTP, S3, FTP, and every protocols imaginable: https://github.com/mickael-kerjean/filestash People tend to come in the demo to upload / download tons of stuffs

Do you have problems with illegal content shared over your service? I wouldn’t ever offer something like this because I wouldn’t want to deal with someone uploading child porn or stuff like that.


Not really scared as I take action the second I hear about nasty things. The most recent events were people creating shared links pointing to FTP servers that had revenge porn and sharing those on a whacky telegram account, another annoying one is people who keep trying to brute force other people accounts on cloud providers even though there's a protection in place to throttle to 20 connection attempt per second, effectively degrading the free service for everyone else and getting me kicked off from cloud provider like the most recent one being Hostinger

github's registry (ghcr.io) is the other big free one

BitTorrent seedbox. If you are apart of a private torrent tracker you may download 5tb and upload 10tb of data per month to be in good standing within that tracker’s community

I've always wondered how ratios like that are supposed to work. Surely the ratio has to be 1.0, averaged over all users.

most trackers have some kind of bonus point system

Good thing I pay only 30 EUR a month for symmetric 1 gbit fiber. Right now with BF you can find BF deals cheaper than 2 USD a month, and then there's Usenet.

Is this legitimate data or pirated media?

Yeah, this is it.

Structural monitoring in engineering for example. You need high quality pictures and have lots of drone data and files need to move to hyper scaler or on premise for GPU processing.

Cloudflare is not trustworthy and you don't always need to pay the "protection" fee. I.e. B2B where a day offline doesn't matter.

Cloud pricing is insane compared to Hetzner with regards to storage and CPU power. If you have some Linux knowledge it's perfectly stable. Support is top notch.

For 50€ per box you get 10gbe which is also nice. Not everyone needs 100% IaC like the big hyper scaler offer or geo redundancy etc.pp... even Hetzner has an API and you can even terraform their dedicated servers to some degree. It's 15 to 30 minutes vs. seconds but who cares.


> video streaming?

Pet peeve but yes. It should not, in 2024, be considered a niche use case to deal with.. video. We’ve become accustomed to YouTube and their impossible-to-beat free hosting. But it’s really pricey to do streaming and so yes, we should expect large bandwidth being available. “Who needs X” is a question that should be reversed, instead we should ask “why not?”. When we have good affordable infra, we get cool new stuff and everyone benefits.


There are many data streams you may want to process that take lots of network traffic. CT logs, monitoring aggregators, web crawlers, etc. For the traffic you initiate, there's no proxying/caching you can do.

Besides needing it: Not having to fear that an attacker spamming the machine with Mail or web requests or something incurs a bug traffic bill gives better sleep at night.

PeerTube comes to mind.

blockchain nodes

This has nothing to do with Hetzner. It's because of the US tariffs.

"I’ve been a big fan of Hetzner. Unfortunately they’ve made a feeble attempt to dress this change up in the name of “fairness”."

Hetzner is a company know for it's precise pricing structure. An increase in prices would be correlated to an increase in costs, and in fact the next paragraph Hetzner writes:

"With the new tariff structure, we want to make conditions for our customers around the world as fair as possible "

Bottom line, the US imposes tariffs, this increases the prices of imported products, of which Hertzner is one, (Servers from Europe)

How can you write an entire article to complain about a price increase and not see that it was actually your country that increased the price.


They're not importing the VMs from Europe. And there are no new tariffs yet, inauguration is in January so there are only various announcements so far.

They're using tariff to mean price, which is an unusual choice and likely because it was written by someone from Germany. This would not necessarily sound out of place to a German speaker as "Tarife" is a common way to describe differently priced plans for any kind of service.


It's British English. Tariff is used to mean import duties, business prices (e.g. "phone tariff") and also prison sentences (e.g. "whole life tariff").

Though given the political context, it would probably have made sense to use a different word.


It's most likely imprecise wording due to a linguistic false friend from German (Hetzner is a German company); see also my other comment.

It's not a false friend, it is the same word.

Not only do they share an etymology (the word "tariff" in English and "tarif" in German both come from the same Arabic word), they also mean the exact same thing. It's also used in a bunch of other European languages in essentially the same way, with either a similar or identical spelling.

It's only US English where some of the meanings that are still common elsewhere are no longer used. It's actually kind of amusing that Americans would struggle to understand the meaning of an English word, but a French or Italian speaker would understand it. I wonder if there are any other words like this.


Having the same etymology and even having the exact same spelling in two languages does not disqualify a word from being a false friend at all!

Take "gift", for example: "Gift" means "poison" in German, despite both the English and German word deriving from the same Proto-Germanic root, meaning "to give".

> (the word "tariff" in English and "tarif" in German both come from the same Arabic word), they also mean the exact same thing

In this very thread you can find a pretty good counterexample to that proposition, at least in American English.


"Tarif" indeed roughly translates to "plan" in this context and can be used for all kinds of plans/rates/tariffs, not just the import duty kind.

No new tariffs yet. My bad. Misread

Sorry, is Hetzner a company from Canada, China, or Mexico? Are digital goods & services indicated to be under tariff as well?

Ah my bad. Tariffs not implemented yet, and restricted to those areas. Might have been a bad choice of word by hetzner

Because of announcements of tariffs?

I hope this experience teaches you to be more polite and open-minded even when you think you know what you are talking about.

Yeah. Being wrong is always humbling.

That said, sometimes I'm right so I find a balance in the upswings and downswings of confidence


Probably more likely has something to do with increased energy costs.

Pretty sure VA/US energy prices are a lot cheaper than German ones.

I think it wouldn't be very relevant. A surge in US energy prices would be sufficient to cause a rise in US compute prices.

I think the US has been scaling compute for both AI and mining the last year, so it makes sense


Cool: then maybe they can precisely explain the full breakdown of these changes instead of hiding behind a word that explains nothing.

Used Linode for years. Brilliant service. Not sure how the Akamai takeover will pan out.

Anyone else looking to migrate. https://www.serverhunter.com/


Ironically one of the reasons I abandoned Linode for Hetzner was because I assumed Akamai would ruin it. I still have stuff on Linode and I don't think Akamai ruined it.

With that said, Hetzner's prices are still cheaper and I've never had issues with their service, but I feel gaslit by this announcement.


As an age-old Linode customer from the UML days, I'd say that Akamai has done a good job with it. I've had nothing to complain about.

Well the Akamai takeover was 2 years ago, so fairly certain whatever you have now is how it ended up panning out.

I did panic a bit when it happened but my Linode just keeps working so I’m not unhappy.


Was it two years ago?? I could have sworn it was less than a year ago. Anyway yeah I’m happy with Linode

As a long term linode user, I think the enshittification is well underway :(

It’s kind of not the brightest to both raise prices and reduce services at the same time, who’s in charge over there? Maybe I’m missing something ?

hetzner is really known for its german bandwidth prices, a change in the US is fairly insignificant IMO. for most applications you could just put a free CDN in front of the cheaper german service to reach the US

I personally use 0TB per month across 6 CPX21 servers (I know I’m over-provisioned; that’s not the point).

Kinds of is the point tho, you're hogging resources you don't actually need (not talking about traffic here ..).

People probably spam-provisioned cheap CPX boxes to get cheap bandwidth.

Also, complaining about a "large 27.52% price increase" is kind of absurd when the absolute value of the increase is just under 2€.


> Kinds of is the point tho, you're hogging resources you don't actually need (not talking about traffic here ..).

That seems backwards? They're not hogging anything, they're paying for network they're not using; this is excellent for Hetzner.


I know it's a very simple question but what is the cost of bandwidth? I appreciate that a server is a machine with parts that degrade. I also appreciate that lines need to be laid for the most part and there is a need for a return on digging up the streets to lay cable.

So is bandwidth going to the cable?

I am asking that because I never quite understood 3g/4g/5g costs. In my youth it was $0.10US for a text message. Now in my country, no one would pay to send a standard text. It's free even on PAYG.

I'd imagine there is similarity?


On a Hetzner level, you don't quite know, as bandwidth deals between large hosting companies and IP transit providers like Hurricane Electric, Cogent, Zayo, etc, are typically protected by non-disclosure agreements.

Unless you are a big enough fish (FAANG size) in the internet pond to do free peering with all the big boys, the contract with be negotiated for a certain amount of dedicated bandwidth (specified in megabits or gigabits per second), not like the terabyte like in the consumer LTE context.


Cloud providers make their money with the high bandwith costs. I can get 10 gbit/s line for ~$3,000 a month with a dedicated server. That's 3 PB of data and according to AWS cloudfront calculator it would cost ~$90,000 with them...

So yeah, people get ripped off, just like with SMS in the past. Most don't know this or think it's fine, because "cloud" has benefits that justify the costs in the end ("I don't have to pay for sys admin!").


My first smartphone had a data plan which cost 1024€/Go. And I was happy about how cheap it was (it got more expensive shortly after).

Cellular towers aren't cheap: you need to install quite a lot of them, especially in big cities!


Those are some steeeep drops. I'm curious how they settled upon such numbers.

Presumably by looking at their competition, and realizing they’re still a better option.

Staying competitive swings both ways.

To be fair their pre-change allowances were insanely generous.

The bandwidth reductions look like they bring the US in line with Singapore, i.e. typically 20TB of bandwidth in the EU and 1TB in Singapore, US moving from 20 to 1.

I wonder if their cost base has changed significantly. Singapore having just 1TB included makes sense because in Singapore traffic is charged differently and costs a huge amount for all companies. I wonder if the colo provider for Hetzner in the US has screwed them over in some way.


My take on this is that there’s a price increase on power consumption and Hetzner decided to bump the prices now; perhaps they are facing a power price increase in January 2025. Do Ashburn or Hillsboro publish their power prices and have they increased in the past year?

They specifically mention tarifs, maybe its political. Can't expect that kind of thing to get thrown around without international response.

Is anyone old enough to remember 1and1? Similar arc?

Maybe I'm too young to remember the good times, but to me 1and1 were just the European GoDaddy. Mostly targeting small businesses who didn't know any better and ripping them off with bad services at inflated prices, through a lot of well targeted marketing. Selling things like email with a 250MB inbox and 2MB attachment limit far beyond the time when that was a reasonable offering, and at a far higher price than that was worth (being worth roughly zero).

They started with free .com registration.

They grew their business with this offer, and then degenerated into what you described.


1and1 were also notorious for aggressively trying to collect on "debts" - more than one person I know had 1and1 send debt collection letters to their parents over renewal fees for hosting plans they didn't even want.

They still exist, called ionos now: https://www.ionos.com/

Yes, I should have been more clear. It was another Germany? based provider that gave excellent, even free prices... until they didn't.


Yes this hit me. We use a ton of traffic. Any alternatives? Although the pricing isn't too bad even with the increase, nice to have backups.

We use Hivelocity. Not exactly the same service (bare metal).

Checks out

[flagged]


> 1890k =-0;.9l, km8nubh6ydes3w2q1

What is the meaning of your comment?


dead drop maybe?

Hetzner suck. So do Vultr (OVH?). People learn about the bad things cloud providers do but still go crawling back



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: