Details would be enlightening : ) Was it the servers or the applications running on the servers that were taken down? Running bare metal servers exposed to the public is a fairly obvious footgun, there should be at least one layer of load balancing in front in addition to the provider’s firewalls. I’d argue MOST publicly exposed servers run by amateurs can be taken down for less than $5. Regardless of them being bare metal at Hetzner or some EC2 instance.
Not my experience at all. Hetzner obviously does offer DDoS protection and responds quickly to that kind of issues. I've also had Hetzner techs proactively contact me regarding attacks on our infrastructure (none of which actually took any of our servers down, by the way). For specialized needs, you can even have your own hardware installed next by your servers in the same rack for a relatively small premium.
It's not just companies using the big 3 as simple VPS providers. A lot of applications are also hugely over-engineered for their actual needs, and unnecessarily ties themselves to the proprietary cloud APIs just for the sake of IAC or just for the sake of the simplicity of having the whole infrastructure at one provider. Or for the sake of using Kubernetes, for which I guess 1/10 of use cases are actually appropriate. I guess part of the problem is that using Big Cloud Provider X is the default in a lot of companies, and alternatives are not even being considered when starting out a new project.
You should be running at least three, preferrably four (3 + hot spare) mirrored instances if you require that level of uptime, regardless of provider or tech (bare metal/VM).
I've been using Hetzner servers for ~15 years with multiple clients and employers, and always been disappointed with other providers compared to what Hetzner delivers. OVH with their frequent network level outages, the 2021 fire and so on. DigitalOcean with their way too frequent and long lasting maintenance windows. And AWS/GCP/Azure with their obscene pricing, ridiculous SLA and occasional hour-lasting outages. One application platform I managed was migrated from DO to Hetzner with huge cost savings, much better uptime and insanely much higher performance running on bare metal servers rather than cheapo VMs. If you need more than two vCPUs and a few gigs of RAM, I see absolutely no reason to use overpriced AWS/GCP/Azure VMs.
This can also be achieved in distributed systems by having each node skip IDs equivalent to the number of nodes in the cluster. E.g. node 1 in a 5 node cluster assigns ids 1, 6, 11 and node 2 assigns 2, 7, 12 and so on.
Many of the statements in that article were misleading or outdated already in 2012, though. E.g. the part about PHP being tied to Apache… At that time I had been using php-fpm behind nginx for years.
So I guess my comment is getting downvoted by people who don't understand «mission critical».
>how can you run mission critical if most cloud services have SLAs two or three nines weaker than needed
That's exactly it. Cloud providers usually provide SLAs in the range of 95-99%. Amazon doesn't provide a full refund until monthly uptime goes below 95 %. Elastic apparently doesn't provide an uptime guarantee at all, they only provide an SLA for support ticket response times. And only on gold and platinum subscriptions.
This incident lasted for more than 24 hours («most» instances restored after 22 hours). It doesn't matter if it's someone else that has to wake up at 3 am to fix the issue, when they're unable to fix it within reasonable time. Mission critical apps simply can't be down for 24 hours. And it's fully possible to design HA elasticsearch deployments. Elastic Cloud just isn't one of them.
Why would framework owners want you to become tightly coupled to the framework? They're basically just offering a set of solutions that come with some tradeoffs, but they don't intentionally create problems for their users.
So, two years after the Snowden revelations, someone’s surprised the US govt uses planes for mass surveillance. And the «secret» program turned out not to be even classified. This is not really a story.
Upgrade to web3 technology now and get 1000x the hosting bills and 1/1000000 the performance! Because… someone will get rich and blockchain is TEH FUTURE!
Using mysql_ functions directly was very outdated and long deprecated when 7 was released. Any reasonably maintained code base had already been using mysqli or PDO for years.
Apache doesn’t need to be in the loop. You can run php-fpm directly behind a load balancer that supports fastcgi, like haproxy. And for depvelopment you can simply use the built in server.
I think the author just missed the point. You’re not supposed to replace all of your CSS with tailwind classes when plain CSS classes make more sense, like the perspective and nth-child examples.
Alt is a required attribute of the <img> tag, so requiring it is just basic validation. C code won't compile if you omit the semicolon after statements, either.
>Thankfully I’m tech-savvy enough to know what I’m doing.
>I’m honestly more concerned by third-party apps dropping support of these older system versions rather than vague security threats.
This guy probably won't notice that someone used one of many available root privilege escalation vulnerabilities against his system before his bank account is emptied.