501–510 of 1001 posts
How much longer are we going to tolerate this marketing bullshit about "Designed to provide 99.999999999% durability and 99.99% availability"?
Every week or so we interview a company and ask them if they have a fall-back plan in case AWS goes down or their cloud account disappears. They always have this deer-in-the-headlights look. 'That can't happen, right?' Now imagine for a bit that it will never come back up. See where that leads you. The internet got its main strengths from the fact that it was completely decentralized. We've been systematically erodin…
Planning for an AWS outage is a complete waste of time and energy for most companies. Yes it does happen but very rarely to the tune of a few hours every 5-10 years. I can almost guarantee that whatever plans you have won’t get you fully operational faster than just waiting for AWS to fix it.
US-East-1 is more than just a normal region. It also provides the backbone for other services, including those in other regions. Thus simply being in another region doesn’t protect you from the consistent us-east-1 shenanigans. AWS doesn’t talk about that much publicly, but if you press them they will admit in private that there are some pretty nasty single points of failure in the design of AWS that can materialize…
>US-East-1 is more than just a normal region. It also provides the backbone for other services, including those in other regions I thought that if us-east-1 goes down you might not be able to administer (or bring up new services) in other zones, but if you have services running that can take over from us-east-1, you can maintain your app/website etc. I haven’t had to do this for several years but that was my experien…
Every week or so we interview a company and ask them if they have a fall-back plan in case AWS goes down or their cloud account disappears. They always have this deer-in-the-headlights look. 'That can't happen, right?' Now imagine for a bit that it will never come back up. See where that leads you. The internet got its main strengths from the fact that it was completely decentralized. We've been systematically erodin…
Planning for an AWS outage is a complete waste of time and energy for most companies. Yes it does happen but very rarely to the tune of a few hours every 5-10 years. I can almost guarantee that whatever plans you have won’t get you fully operational faster than just waiting for AWS to fix it.
If your resilience plan is to trust a third party, that means you don't really care about going down, does it?
Besides that, as the above poster said, the issue with top tier cloud providers (or cloudflare, or google, etc) is not just that you rely on them, it is that enough people rely on them that you may suffer even if you don't.
Every week or so we interview a company and ask them if they have a fall-back plan in case AWS goes down or their cloud account disappears. They always have this deer-in-the-headlights look. 'That can't happen, right?' Now imagine for a bit that it will never come back up. See where that leads you. The internet got its main strengths from the fact that it was completely decentralized. We've been systematically erodin…
The internet is a weak infrastructure, relying on a few big cables and data centers. And through AWS and Cloudflare it has become worse? Was it ever true, that the internet is resilient? I doubt. Resilient systems work autonomously and can synchronize - but don't need to synchronize. * Git is resilient. * Native E-Mail clients - with local storage enabled - are somewhat resilient. * A local package repository is - so…
Now we have computers that shit themselves if DNS isn’t working, let alone LANs that can operate disconnected from the Internet as a whole.
And partially working or indicating this it works (when it doesn’t) is usually even worse.
Every week or so we interview a company and ask them if they have a fall-back plan in case AWS goes down or their cloud account disappears. They always have this deer-in-the-headlights look. 'That can't happen, right?' Now imagine for a bit that it will never come back up. See where that leads you. The internet got its main strengths from the fact that it was completely decentralized. We've been systematically erodin…
Planning for an AWS outage is a complete waste of time and energy for most companies. Yes it does happen but very rarely to the tune of a few hours every 5-10 years. I can almost guarantee that whatever plans you have won’t get you fully operational faster than just waiting for AWS to fix it.
It shouldn’t, but it does. As a civilization, we’ve eliminated resilience wherever we could, because it’s more cost-effective. Resilience is expensive. So everything is resting on a giant pile of single point of failures. Maybe this is the event to get everyone off of piling everything onto us-east-1 and hoping for the best, but the last few outages didn’t, so I don’t expect this one to, either.
Doesn't help either. us-east-1 hosts the internal control plane of AWS and a bunch of stuff is only available in us-east-1 at all - most importantly, Cloudfront, AWS ACM for Cloudfront and parts of IAM.
And the last is the one true big problem. When IAM has a sniffle, everything else collapses because literally everything else depends on IAM. If I were to guess IAM probably handles millions if not billions of requests a second because every action on every AWS service causes at least one request to IAM.
US-East-1 is more than just a normal region. It also provides the backbone for other services, including those in other regions. Thus simply being in another region doesn’t protect you from the consistent us-east-1 shenanigans. AWS doesn’t talk about that much publicly, but if you press them they will admit in private that there are some pretty nasty single points of failure in the design of AWS that can materialize…
What it turned into was Daedalus from Deus Ex lol.
One main problem that we observed was that big parts of their IAM / auth setup was overloaded / down which led to all kinds of cascading problems. It sounds as if Dynamo was reported to be a root cause, so is IAM dependent on dynamo internally? Of course, such a large control plane system has all kinds of complex dependency chains. Auth/IAM seems like such a potentially (global) SPOF that you'd like to reduce depende…
Many AWS customers have bad retry policies that will overload other systems as part of their retries. DynamoDB being down will cause them to overload IAM.
>We recommend customers continue to retry any failed requests.