AWS Lambda Silent Crash – A Platform Failure, Not an Application Bug [pdf]

(lyons-den.com)

115 points nonfamous | 2 comments | 15 Jul 25 01:22 UTC | HN request time: 0.415s | source

Show context

WatchDog ◴[15 Jul 25 08:51 UTC] No.44569244[source]▶

This is a simple misunderstanding of the Lambda execution lifecycle, but it's maybe not a great look for AWS support, that no one there was able to explain the problem to this guy.

I guess it's possible someone did explain it to him, and David hasn't mentioned that in his diatribe.

replies(2): >>44569370 #>>44571140 #

1. jdeastwood ◴[15 Jul 25 09:16 UTC] No.44569370[source]▶

>>44569244 #

I actually witnessed almost exactly the same thing happen in my org recently. AWS support was far _too_ nice and actually did escalate to the service team (or claimed to).

There was a few days of back and forth which should have been resolved by saying "please read the how lambda works" section of the documentation.

replies(1): >>44571455 #

2. the_plus_one ◴[15 Jul 25 14:22 UTC] No.44571455[source]▶

>>44569370 (TP) #

It's been several years at this point, but I used to be a cloud support engineer for AWS (i.e. the first line of defense against support tickets). We were trained to always be kind to the customer, even if we're frustrated with them, so a response like "please read how the lambda works" probably would've gotten me reprimanded. Customers can rate your response from one to five stars, and if I remember correctly, anything less than four required a written explanation to your manager as to what went wrong.

That job was a glorified call center; I didn't last very long.

↑