AWS: Inaccurate Estimated Billing Data – $1.7 billion • Hacker News Board

Re: AWS: Inaccurate Estimated Billing Data – $1.7 billion

#791

Post by dawnerd » Thu, Jul 23, 2026, 2:04 AM UTC

myself248 wrote: post #790

dawnerd wrote: post #768

Earlier quoted context omitted.

I’ve never had a limited card get rejected. Granted I use ones from privacy.com. That’s also what I have hooked up to my aws / amazon.

They are also not all that limited. I used a Privacy.com card for a New York Times trial subscription, which of course I could not cancel online, so I just let it expire. A few days past the renewal date, I saw a $waytoomuch charge on my card. I wrote to Privacy, and this was their reply: > Jun 1, 2022, 5:21 PM EDT > I've been reviewing your dispute and wanted to touch base with you to explain what happened. > It app…

That seems like it shouldn't be even allowed.

Re: AWS: Inaccurate Estimated Billing Data – $1.7 billion

#792

Post by cosmoz » Sat, Jul 25, 2026, 12:29 PM UTC

[flagged]

Re: AWS: Inaccurate Estimated Billing Data – $1.7 billion

#793

Post by xp84 » Sun, Jul 26, 2026, 9:33 PM UTC

el_memorioso wrote: post #454

xp84 wrote: post #370

Earlier quoted context omitted.

Somehow I highly doubt anyone will leave AWS over this unless their use of AWS is way more low-complexity than the average account. People make similar pronouncements after every us-east-1 outage makes the news, but I feel like AWS would be going out of business by now if people followed through. It reminds me of airlines, where after a particularly grueling irregular ops experience, a few dozen people file off the p…

Airlines are all subject to a lot of the same factors, but there are unequivocally better and worse performers in terms of on-time arrivals, by a lot. Take a look at the Air Travel Consumer report for details.

> unequivocally better and worse performers in terms of on-time arrivals, by a lot.

I don't think it's a very good metric to compare them with, though. Those numbers are going to be heavily weighted by conditions at their hub and focus city airports, and speaking more broadly, the cities they serve. Meaning, things that aren't really in their control, and which are either in your control (often, you get a choice of which connecting hub to fly through), or is a fixed risk to you because of your origin/destination, no matter what airline (e.g. if you're flying from/to Orlando in the summer, expect a risk of thunderstorms interfering. Or snow in winter in Chicago.)

If I'm travelling from, say, LAX-PDX, United's systemwide on-time performance is not the data I'd care about, because it's probably affected far more by the conditions in their O'Hare and Newark hubs, which I'm not visiting.