Amazon Partially Down

[GRBurst]

On the webstore, you will be facing the following message “We’re sorry, an error has occurred. Please reload this page and try again.”

Moreover, multiple reports at https://downdetector.com/status/amazon/

And on twitter and social media.

[edhelas]

Happy to see news about the Amazon deforestation catastrophe on HN

[ac2022]

Pretty funny, I was going to order a balance board that I had been eyeing for a bit. Went to Amazon but it was not loading the page correctly, so I went to company's website to look up specs and found out they had 10% off coupon & no taxes. So I purchased it from them.

Wondering how much Amazon outages can actually increase revenues for smaller companies.

[tyingq]

AWS us-east-1 had issues last Thursday also

[UI_at_80x24]

These 2 service issues also seem to match my organizations issues using MS Azure VDI's.

Crashing, Saving files, and general instability of the platform. I have wondered if Azure is doing some load-balancing on AWS.

[elaus]

For me (central Europe) it was down about half an hour. I can't remember the Amazon store website being down that long before. Curious to see what caused this.

[rasz]

>“We’re sorry, an error has occurred. Please reload this page and try again.”

seeing this exact message on reddit when trying to post

[antisthenes]

First noticed it when it had vertical text alignment on the product pages a few hours ago across all browsers.

[mholt]

Still down here in western US. Some product pages don't even show that error; they're just empty.

[GRBurst]

It is strange that there is no statement from Amazon, yet. Or did I miss it?

[Bender]

Could it be a coincidence that this occurred after John Oliver's report [1] on tech?

[1] - https://www.youtube.com/watch?v=jXf04bhcjbg

[pc86]

Not only could it be, it definitely is.

[dnadler]

I just recently published a project and had been mulling over whether I should move it to AWS or keep hosting it on the server in my basement...

Seems like I made the right choice!

In all seriousness, I don't think I'd rely on a single provider for anything these days, unless I really don't care when it goes down. That's pretty easy and cheap to do with most things, but I'm not quite sure how to approach a backup database in an economical way.

[Freebytes]

When infrastructure costs are your largest business expense, then having two separate providers for a duplication of services is not feasible. You must also then have methods by which to determine when the service is unavailable and seamlessly fail over to the service that is actively running.

In cases where data integrity is important, this could also result in splits where data gets left behind. It is often better to fail than to have situations where the data does not match.

[dnadler]

Well, to address the costs issue, you could run hot-hot and have half of your resources on each provider, and in the event of failure, scale the functioning one up to meet demand. That's obviously a decent amount of effort to set up, though.

[OJFord]

Is your basement not also a single provider?

[dnadler]

Yes definitely, what I meant is that I would probably use AWS as a fallback for my basement, or maybe the other way around. I didn't make that clear at all.