I like the spirit of Serveless. I am with you!
Just will give 2 reasons why I did not use serveless all the time.
Reason 1: AWS Lambda charges you every 100 ms, even if your API answer in 6 ms. This could become expensive depending how many million requests do you have. I hope one day AWS charges us…
Hey Kenneth, thank you for your tutorial :-)
I use VS here in Windows and I execute all the my Node code in WSL (ubuntu bash). How can I set my debug to starting a node process from Ubuntu? There’s any attribute at launch.json where I can set something the run cmd of node? Something like that:
cmd: ‘bash -c “node app.js”’
One of the best posts about Lambda. Congrats! Good to know that we have to ping our APIs to keep everything alive every 10 minutes to avoid wiard delays. For sure AWS team will not like this method — and they will probably create a new kind of billing in near future, something like “container uptime price per minute”
What a great post! Every 10 posts that I read about AWS serverless (API Gateway +Lambda) — nine of them is about “wow serverless are great, I am loooving it — you should too” — and there is always one guy with balls to say the truth like you did: Lambda and API Gateway + Lamda are still walking in baby steps, with a lot of downsides — Buuuut, we can not fight against hype, can we?
I feel your pain. We started using ng2 on small isolated projects, but we just gave up — toooo many BC.
How Angular2 core team can have a good night of sleep breaking a lot of things on every RC?? That is bad karma for their entire life, for sure! Lol.
Cool TJ! How much requests per minute or day do you have on your Uptime Service Check? / How much are you paying now, monthly? / How made days did you take to build this service? — I am willing to use lambda in some projects, but it seems a lot easier to use t2.micro, in terms of Develop Time, Monitoring, Maintence and Server Price. I can have a few…