Maybe it was just a meteor that hits there Solarpanels.

 

Update from October 10, 12:44 PM: According to Amazon Web Services (AWS), the underlying DNS issue has now been resolved. Most AWS services are back up and running. However, some throttled requests may still occur while the full recovery process continues.

Some services, including CloudTrail and Lambda, are currently still working on retroactively processing backlogged events. Despite the overall stabilization, increased error rates continue to occur when launching new EC2 instances, as well as services like ECS built on EC2, in the US-EAST-1 region. For users who continue to experience issues resolving the DynamoDB endpoint in this region, AWS recommends flushing the DNS cache.

"Technicians were immediately contacted and are actively working to resolve the issue and fully understand the root cause," AWS reports. According to current information, the cause may be related to DNS (Domain Name Service) resolution. This is the process by which domain names are converted into machine-readable IP addresses. This is necessary for computers to communicate with each other. (Sources: allestörungen.de, Amazon Web Services, X) (cln)

Source: www.fr.de/verbraucher/amazon-canva-snapchat-und-co-ueber-40-dienste-von-stoerung-betroffen-zr-93995169.html

 

I GUESS PANIC IS OVER
BUT WHAT TO DO NOW??

 

It was PuPputin fault!

 

As usual when a western thing don't work

 

I wanted to write a long systems guy explanation why this shit is bad in oh so many ways, services going down being just the surface of it but I really can't be arsed. I'm going to meet up with some systems guys still doing, well... systems later on to console them with some drinks and swap war stories.

But in short: Amazon's US-EAST zone is probably the biggest single purpose data centre complex in existence and it hosts a metric fuckton of sites and services. And I think it's still the default when setting up services with Amazon.

And as things go, the problems love to cascade, and just one piece of the puzzle needs to be on AWS for the whole to fuck up, see for instance Barclays Bank today, because their actual money traffic is not on Amazon, but some of the front-end services are, hence their entire retail customer system being unavailable (I'm pretty sure interbank traffic worked just fine, it's just not for us plebs).

I don't work in syseng that much anymore, but even I was indirectly affected... and not just surfing por... reddit on billable time. But this is just endemic in the industry, all eggs in one huge basket.

(and for the syseng/adm types amongst us... yes it was the DNS, it's always the DNS)

 

go figure

 

back to playing with Pianoteq i guess

Spoiler: PulseWave mode on

Let’s craft a quick experimental Pianoteq patch that blurs the line between a grand piano and a synth pad — think “organic-synthetic hybrid.” Here’s a step-by-step approach:
Pianoteq Hybrid Patch: “GranSynth Pad”

1. 
   Start with a Grand Piano model
   
   • Select your favorite grand (e.g., “Steinway D” or “Yamaha C7”) as the base.
     
2. Adjust Hammer & String Settings
   
   • Hammer Hardness: Lower it slightly for softer attacks → gives pad-like smoothness.
     
     
   • String Inharmonicity: Increase subtly → adds a slightly metallic, evolving shimmer.
     
3. Pedal & Resonance
   
   • Sympathetic Resonance: Turn up moderately → creates a rich, lingering tail.
     
     
   • Damper Noise: Lower slightly → reduces “real piano” realism, making it more ethereal.
     
4. Modulate the Sound
   
   • Velocity Curve: Flatten it → soft and loud keys sound closer in volume.
     
     
   • Mod Wheel Mapping: Map mod wheel to hammer hardness or string detuning → allows expressive “swells.”
     
5. Add Effects
   
   • Reverb: Large hall or pad-style reverb, increase decay.
     
     
   • Chorus / Ensemble: Subtle to thicken the sound.
     
     
   • Optional: Use EQ to boost airy highs and warm mids.
     
6. Optional Layering
   
   • Use Pianoteq’s microtuning or dual-layer feature: detune a second layer slightly → creates a lush, dreamy soundscape.
     

 

You ofc can make fun of everything, but there are also some serious services involved.
There are said to be a few people who operate their alarm system via Ring. They may not feel comfortable when they can't just go home to check it.

I would never trust in AWS. A few years ago they had a (really massive) major outage and when they came back some of their clients had lost data. And even AWS backup were lost forever.
The next thing they did was reminding their customers that, by the terms, it was their duty to make a backup of all data on AWS.
Great move.

 

Last edited: Oct 20, 2025 at 5:00 PM

[QUOTE1"Will Kweks, post: 867834, member: 119936"]a metric fuckton[/QUOTE]
Ah, I love this terminus technicus. Put it into my everyday toolbag

 

Oct 20 8:43 AM PDT We have narrowed down the source of the network connectivity issues that impacted AWS Services. The root cause is an underlying internal subsystem responsible for monitoring the health of our network load balancers. We are throttling requests for new EC2 instance launches to aid recovery and actively working on mitigations.



official info: https://health.aws.amazon.com/health/status

 

Spotted on Mastodon:

AWS: We're having an outage that affects only one of the regions in our globally distributed infrastructure platform.

The internet: it's a good thing you don't have, like, one region you consider globally available, that acts as a centralised management point for a lot of your internal workings, right?

AWS: ha ha right yes that would be bad sweatdrop sweatdrop.

 

Unlike CrowdStrike, Amazon stock has yet to be impacted by this outage:
I remember that Crowdstrike stock lost more than $100 in the same day but it doesn't seem to be similar with Amazon.
For now, it's more like business as usual.

 

Amazon is, as they like to say, too big to fail. I mean that literally. This "little" outage will not cause any exodus of customers away from the platform, there will be no pricing adjustment, there will not be any cut into AWS profits.

And even if this would cause businesses to consider their strategies when it comes to services hosting, it will not happen overnight. Migrations take a long time because AWS offer a lot and this takes a long time to replicate or replace, and oftentimes it's eventually decided that it's safer to stick with a known but shit platform, see for instance VMware and their acquisition by Broadcom. Are businesses hostages suffering from Stockholm syndrome with these? Oh, I think I know the answer!

 

I'm just annoyed this makes me look at pictures of the employee's shitty dogs when trying to track orders