The "don't do it if it's going to make you miserable" stance is a nice switch from the usual "startups are suffering so go read some Marcus Aurelius" attitude we get around here.
I am not sure if I buy the "it's nobody's fault" explanation. From what I gather, what they built simply did not work, even though it was clearly feasible. They had a failure of engineering and could not afford to rebuild the entire app. I know nothing about the specifics of the architecture, but I do know first-hand that if using Rails as the basis of an efficient and high-capacity SaaS backend, it is the wrong decision.
It would be a terrible shame if you tanked a viable business because of availability concerns, because these sorts of problems are well solved. Highly available systems are designed in such a way that any individual component may fail without bringing the overall system down. (ie. "Kernel issues? Detect, remotely reboot server." Yes! There's free code for that. No. Many run of the mill sysadmins probably aren't familiar with it.)
Hi, y'all. Amy here. If you've got a real question, I'd be glad to answer it.<p>EDIT: As of 4pm eastern I'm off to meet my interior designer and then attend a joinery demo at a wood shop so it may be a bit but I will absolutely read & respond to your questions later or tomorrow. My goal is to share. And maybe mock people a little, if they can't think up an original insult when they attack me.