Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
German discount host Hetzner launches new servers (hetzner.de)
90 points by oellegaard on July 15, 2013 | hide | past | favorite | 83 comments


Be prepared to incredibly unreliable disks. Or better, ask for new disks for your box, even if this will require some additional cost.

Clarification: I had bad experiences with disk reliability, I told different friends, that in turn had very bad experiences including multiple disks in RAID breaking in the same time range.


Besides the mentioned issues with reusing disks, they also had used and maybe still use some faulty series of Seagate drives e. g.:

http://forums.seagate.com/t5/Barracuda-XT-Barracuda-Barracud...

So beware of having those installed, under high I/O they will fail at a random instant.


A Barracuda on my Hetzner server failed 1 hour ago. This is the second time I had disk trouble at Hetzner - the first time both RAID disks failed at the same time.

However, support is acting incredibly quickly, within 5 minutes of sending the support request they got to work on the server.


Both disks failing at the same time might indicate that the problem is with the controller or cables. One of our disks (at Hetzner) died on us a few months ago and unearthed an issue with backups. :(

Because of missing data we asked for a data cable replacement first (instead of just replacing the disk) - and it worked. Even more, it works without problems since. YMMV of course.

BTW, I have worked with many providers but Hetzner is my favourite. Great support and very few problems.


In my case I think it actually was the drives though, it was the same model Barracuda. Looks like they're replacing them with Toshibas now, so I'm hopeful.

> BTW, I have worked with many providers but Hetzner is my favourite. Great support and very few problems.

I'll second this. Support is/was often a horror story for me, but so far Hetzner has been absolutely impeccable.

Contrast this with Mediatemple, where I once had to engage in a lengthy support battle with a chain of incredibly ignorant and condescending support staff before being escalated to a senior person who finally, finally admitted I had found huge problem with the cluster. This took several days of intermittent outages, dozens of emails, and a lot of energy...


I also had a double disk failure on my Hetzner server quite recently. The server didn't respond, and when I hardware reset it began rebuilding the RAID, but it didn't say which disk failed. It happened again very shortly. Hetzner changed the cables and it has now been a month without any issues.

After the disk failures, however, some of the virtual images I had on there had trouble booting. They were complaining about bad blocks. After running fsck they ran again.

The whole experience made me realize that I really need to run on a filesystem with checksumming. Just plain RAID-1 only helps for a limited subset of errors, and only if it is able to detect which disk failed correctly.


I have a couple of ST33000651AS in my server, they failed several times, but the current pair has been running for more than 1 year, I guess they just got a nasty batch.

Needless to say, rebuilding 3 TB raid took a long time and downtime and pain... but all in all, it was worth it, considering the prices.

I can also confirm support was quite fast, even if some of the replacement drives were not new and some even came with bad sectors lol ... such is life.


Odd. I am thrashing the disk of my two hetzner servers with with massive amount of writes and reads for http://searchcode.com without any issues over 2 years. One takes care of HTTP/DB and the other the search index.

The DB server has pumped over 10 TiB of data in the last 80 days (mostly non cached) and the index is rebuilt constantly and well over 1.5 TiB in size.

Anecdotal I know but another data point to consider.


Off topic: Your blog posts regarding building a search engine are quite fascinating, thanks for them!


Thank you. I couldn't find much online like it so wrote it myself. I am hoping a actual expert does the same some day.


That's one awesome project you have right there. I was just thinking of making something like that for F# with data from Codeplex and Github. But I guess your already-existent product blows my non-existent idea out of the water.


Thank you. This sort of feedback keeps me going.


Similar situation here.


I can confirm this, we're replacing at least two disks every month. You really have to design your architecture that allows a server to be taken offline for a few hours for disk replacement. Also, never use raid 0 on Hetzner.


RAID 0 is generally frowned on in pretty much all circumstances. Is there a specific reason you're using it? (Legitimately curious.)


We have a lot of servers running there and they indeed do fail once in a while. We designed it, so that we can recover very fast, by having a standby server ready. Its usually faster to order a new server, than HD swap.


Yep, I had disk issues on a new box too. Hetzner is great though. Also don't be too worried about high ping times, I recently moved http://readable.cc away from Linode and the site is much, much faster (and I'm in Australia, the other end of the planet.)


I can confirm this. Im running many servers with all kind of hosters since the 90s and the only one that had a Harddisk go haywire was the one Server I had with Hetzner.


I've had hard drives go bad in a huge variety of different servers in many different locations. The Planet (now Softlayer), Digital Ocean (VPS ...), NAC.net, Atjeu, and others.

These days I just expect it to go bad, so I use ZFS within FreeBSD to watch for errors, and when they happen I ask them to replace a single drive, and then the other.

So far it has worked very well for me with no data loss. I've been fairly lucky in that both drives have't failed at the same time.


We also had very bad luck with hard drives at Hetzner. Even the replacement disks we got after one gave up had 25000+ hours of spinning time according to SMART.


I have also had many disk failures, including two RAID'd HDs failing within a few days of each other. I think we've lost 5 HDs so far. (We only have 1 server)


I heard this story multiple times, apparently it is possible in the new servers to ask them for new disks and this solves most of the issues.

Also the procedure to get the disk replaced is insane and involves you dealing with disk IDs... It's a RAID, then should just check what is the broken disk and replace it, end of the story.


What makes you say this? I've been running on Hetzner for about half a year now, no issues so far.


My 2¢: I've been on Hetzner for a few months, experimentally, and the sporadic 'offline uncorrectable sector' emails I'm getting from smartd are making me nervous.


Maybe you're not stressing your disks enough. I recently setup a database server and a disk broke (as in becoming unaccessible, not just bad sectors) within a day.


I seem to recall a research by Google which concluded that stress is not really an important factor in disk reliability... Can't find a link now though.

Of course, that doesn't make parent's claim any more true; just pointing out that this is probably not the reason.


is this the google research paper you were talking about: https://static.googleusercontent.com/external_content/untrus... ?


Presumably his bad experiences make him say this.


He only added this clarification after I asked, so my question is not as redundant as it would seem.


A startup I know ran a Dropbox like solution with hundreds of servers mainly on Hetzner servers and had no significant issues with disk reliability although it was not a high stress situation.


I run a hosting service on Hetzner and we haven't had a disk failure on any of our servers for a year or so.


Why?


Hetzner reuses disks to reduce cost. So a "new" server might come with old disks with shorter life-spans. Thats all okay, especially as they are not a 'premium' host. As long as you are aware of this, they are a good host. As others said, you can ask for new disks at a price.

Also be aware that their abuse department shoots first and asks later. I know of servers that were off the grid for multiple days without access because of abuse on a friday - everyone relevant went on weekend.

They _are_ cheap and that comes with a few trade-offs. Doesn't make the product bad, necessarily. If you are cash-strapped and need raw power, buy without hesitation.


When they get an abuse mail for your server, all they do is forward it to you. You then have 24 hours to fill out a form detailing what you did to stop whatever caused the abuse message.

If you ignore the 24 hour limit, it will just escalate the case to a support person, who then decides what to do. There is no automatic shutdown, especially not for multiple days.


This is not true. I've had automatic shutdown without warning. Had to send them some documentation/evidence of solving the issue before they would put me back online on a public IP address.


I had a server shut down after it started scanning other servers from other hosters for vulnerabilities. Taking it offline immediately is what I would expect any sane hoster to do.

Not being available for resolution is a different thing.


I don't see logic behind not powering off an abusive machine, no matter if it's saturday night at 3am. The internet doesn't run on your schedule. If your server has been compromised (or not, heck we all know there are baddies out there), and it's actively hosting a phishing site, taking part in a DoS attack, etc, why shouldn't a provider be allowed to take action to protect their network, other users in general, and at the end of the day, perhaps the reputation of your site as well?

Blaming providers for the inability to secure your server is incredibly irresponsible. Unless service was interrupted because Hetzner misread an abuse report for another server (which is highly improbable and most likely is not what you implied in your original statement), you shouldn't ever be pissed that your host took action to prevent abuse from occurring.


The problem is not that they cut off your (possibly rooted) server. Thats good practice and I don't blame them for that.

The problem is that relevant persons assisting you in resolving the problem (e.g. giving access using a secure connection) are not reachable over the weekend and take their time. That is a problem of the provider.


that is the biggest issue with unmanaged providers. people don't recognize what "unmanaged" means. You pay for a hosted server, you're not paying for an on-call admin to fix your joomla on turnkey ubuntu setup.


I am also not calling for that. They provide physical boxes, so I expect them to be on call when there is an issue with a physical box. Especially when talking about providers of hetzners size.


You assume a definitive definition of "abusive".


and you don't?

things that warrant disruption of service and are defined as abuse:

1) any type of DoS attack

2) phishing

3) malicious activity (ssh brute force, port scanning without consent, etc)

4) cp

5) anything that generates a court order with jurisdiction, including DMCA if you have anything to do with the US

6) spam/spamvertised site

there are more things that can be defined as abuse, but the others are more or less discretionary, imo.


Having worked an abuse desk for quite a while, I can tell you a non-trivial number of reports are clueless reporters or people trying to get competitors/enemies/etc. boxes shut off.


Currently, I rent a managed server from Hetzner for my SaaS business. However, it has some limitations and I'm thinking about switching to a root server. Thing is, I don't want to deal with security and updates and administration and that stuff.

Can anyone recommend a solution? Is there a good software that works as good as their managed version, e. g. keeps itself updated and has a more or less tight security package out of the box? Or do I need to hire a system administrator?

I do have some Linux/command line skills, but I don't want to put a lot of time in it.


If you want a root server hire an admin. Especially if you want a root server from Hetzner. We had two incidents where our machine either lost power because of a problem in the datacenter or was shut down outside of a predefined service window. Hetzners reaction to the resulting problems* was that they won't do anything because it is a root server. They didn't care that they were the source of the problem. We changed the hoster afterwards.

* First time was only a hanging connection to the server, because the machine was still performing disk checks (they didn't tell us that, we had to request local access), second time the hdd crashed and we learned the hard way that they are unable to configure a software raid properly (but tell you that the root servers are preconfigured with raid), which resulted in a complete data loss. External backups saved us here.


three thing you need to always expect when dealing with a provider who hosts your stuff:

1) offsite backups

2) potential that things go sideways during maintenance

3) offsite backups

Them external backups are always worth the cost!


I think you forgot another one:

4) offsite backups


Be careful and ask support to confirm all the software in your stack will be runnable. We tried to launch an SQL Server (MSSQL) -backed system on one of their root servers and their support said "the Windows licenses we use do not allow installing MSSQL - you will have to buy your own licenses for Windows and SQL Server".


You need a special version of windows to run MSSQL? In any case, isn't it normally the customer's responsibility to provide licenses for software they install and run?


It appears I misremembered this a little bit - the issue was with license ownership, not license type. The answer we got from Hetzner support was: " we don't offer MSSQL licenses, if you want it you need your own license for MSSQL and also for Windows server 2008 R2, because you can't use your own licenses with our Windows license. "

And yes, it is reasonable to expect the customer to provide the licenses for the software, but hosting companies can also sell monthly software usage to their customers - we ended up choosing a host that did exactly this. The monthly cost we pay for Windows + MSSQL will not reach the total cost of a full MSSQL license for three years.


Oh I didn't know that Microsoft changed their licensing to allow that, I was under the impression that only select partners (Amazon, et al) were large enough and Microsoft cared enough about to permit that.


This was on the frontpage just 13 hours ago... https://news.ycombinator.com/item?id=6042442


Looks like hn do distinguish between http and https in submissions. This thread is about the same Hetzner page as the other thread, only the old thread was about the https url.

This: http://www.hetzner.de/en/hosting/produktmatrix/rootserver-pr...

Old: http://www.hetzner.de/en/hosting/produktmatrix/rootserver-pr...



Pity Hetzner's policy on user submitted content is really strict. A single malicious user can get all of your servers take offline for 'breach of terms'.


they usally give you 24hrs to fix it


cite?



If any abuse complaint goes 24 hours without a response your server will be powered down. If you are hosting user generated content that will generate a lot of complaints, you need to be responsive.


The i7-920 is almost five years old at this point and only clocks a 5k on passmark. Very strange offering.

That being said, I'm sure the 4770 is great for Minecraft or as a testing server (lack of ECC ram makes them a bad idea for serious use).


I'm not sure, but isn't the 920 also a lot more power hungry? That could explain the higher price.

Or they just quickly whipped up some consumer-grade Haswell servers to offer while working on more reliable offerings of the higher end segments.

Hetzner usually discards their old servers on their server bidding robot [1], I'm surprised these 920's haven't been moved there yet.

[1] https://robot.your-server.de/order/market


This is most likely a mistake on their site. The i7-920 doesn't even support this much RAM. For >32GB it has to be some kind of xeon, maybe sandy bridge E.



Is Amazon using EEC ram in their servers? Has anyone ever asked if they do?


Amazon runs nothing but Xeons and Opterons, which require the use of ECC memory.


I became interested of finding the answer. It seems that answer is no.

http://serverfault.com/questions/433221/does-amazon-ec2-and-...

>AWS does not comment on this question. The general consensus (based on seeing this question asked on SO, HackerNews, Reddit and so on) is that, unless AWS specifically tells you the instance type is running ECC memory, you should assume it is not.

>Although unscientific and inconclusive I suppose this makes sense - ECC is a selling point, and the fact that AWS does not plaster "all our instances run ECC" on their info pages certainly suggests that they are not.


Xeons support both EEC and non-EEC ram. It' depends on the north bridge chipset on the motherboard if you must use EEC.

Some Opterons require EEC, some don't.


That is some impressive pricing. 32 GB ram, 20 TB Traffic at 49€ / month is cheap.

Is there any way of getting VMware or Xen on thus? I have a project ( http://www.opentestsearch.com/ ) that currently uses 9 small virtual machines, running some different operating systems and software that can't easily run on the same server. Currently colo has been the only viable option. Cloud servers would be too expensive and no cheap dedicated host I know about support virtualizations systems like VMware or Xen.


I have succesfully run Xen on 2 of their root servers for 3-4 years (with 5 to 8 virtual machines on each), ending this episode only recently (not because of technical difficulties). There were some occasional problems with Xen itself (some virtual machines went to zombie state from time to time - I hope Xen is more reliable nowadays), but nothing in connection to a hosting provider.


You can install anything you want on them. Ask them to attach Lara (their KMVoIP system) and give them the ISO to boot the machine... I had VMWare on one of these boxes previously, and they had the installer up and running for me within an hour or so of asking...


I successfully ran ESXi on it, without any issues. I had burn the image and insert it into the server, had to pay a small fee for that, but nothing noteworthy. I think I had to get a different NIC as well, an Intel one, costed like 30 eur or something like that.


Not sure about Xen or VMWare, but I use their auto-installer for getting Ubuntu on their servers, after which you can just install KVM.


Unfortunately I also need some Windows servers :( But running the Linux ones on one server, and the two Windows ones somewhere else is of course an option.


You could probably install Ubuntu, install xen-hypervisor on it and use that Ubuntu as your Dom0. Haven't tried though.


Ok, i am going to ask, any reason to go with Hetzner when your have OVH's DC in Canada offering similar if not better Server for more or less the same price?


If your customers are in Europe and you want everything to be fast.


I've seen a lot of posts on various hosting forums that OVH's Canadian offering is pretty subpar as far as reliability of the network goes. One report (admittedly I think it was a bit old) showed a traceroute that went from Canada to Europe then back to where the user was.

Perhaps that's all been fixed now though.


happy OVH (Canada) customer here - good price, good h/w, no problems so far (just one box so definitely on the anecdotal side)


I don't see anything from OVH that comes close to an i7 with 32GB RAM for the price of Hetzner. Am I missing something?


As today seems to be advertising day, I'll plug HostEurope, http://www.hosteurope.de/, another large german datacenter operator from Cologne.

I've got a virtual server running there flawlessly for the last five years. Customer service is amazing (all communication is english and german), and I can only remeber one single downtime for a few hours in all the years.



Can anyone name any big sites hosted on Hetzner datacenters?


Did anyone benchmark the SSDs yet?




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

Search: