Ask HN: Has AWS ever surprised you with a bill?

5 points by noway_bro ↗ HN
I’m validating a small tool that could’ve caught a $42k DynamoDB spike before it snowballed — within 10 minutes, via Slack alert.

Trying to talk to 10 devs or engineers who’ve been hit by unexpected AWS charges (Lambda loops, S3 egress, log retention, etc).

Not selling anything. Just want to understand:

- What caused it? - How you found out? - What you tried after? - Would a $15/mo tool with IaC cost-diff + Slack alerts have helped?

If that’s you (or your team), I’d love to hear from you.

8 comments

[ 2.3 ms ] story [ 42.2 ms ] thread
(comment deleted)
Yeah, it's a meme at this point that if you're learning AWS, expect to be homeless because you left the EC2 on. Maybe not that bad, but I feel like most people have the logging to detect spikes etc, except those who don't know how to do logging.
Yep — we got hit with a ~$1.8k bill spike on AWS due to a Lambda function caught in a recursive loop during a staging deploy. It kept calling itself due to a misconfigured trigger — we didn’t catch it for ~6 hours because it wasn’t high-traffic, just persistent.
No, only Datadog has done that to me.
DynamoDB got me for a couple hundred in college after I used it for a hackathon. Fortunately support was kind to me and wiped it after I explained.
Nope.

I personally don't use any AWS constructs, just reserve EC2s as needed and host everything myself.

"Not selling anything" is a wild take on "would you pay $15 per month".
hi there, this is a problem I have been thinking about a lot lately as a recent entrepreneur. I would love to chat more and collaborate, please feel free to mail me (email in profile). thank you.