@alex – I’m not sure I’d worry too much about picking one solid hosting provider over another in terms of reliability. You’ll find good and bad reviews for each one. I think it’s more likely a breakdown in your stack will cause downtime/poor performance than a network outage. Asking a hosting provider to debug those issues can be frustrating and/or time-consuming, even for managed hosting.

And yes, Scout can be noisy, but it sounds like you’ve been able to go in and tune your triggers.

If you don’t then there is a risk that by the time you realise that your backup doesn’t include that little bit of incredibly important data hidden away over there … its too late to do anything about it.

I am sure there is some smart @rse SysAdmin along the lines of “You don’t test backups, you test restores”

A family friend’s office building were he rented space for his small business burned down this past winter, and thanks to his weekly off-site backup, his insurance business was up and running the following day with only the loss of a few days work. His weekly off-site backup plan seemed excessive when you look purely at the chance of his place burning down, but when the impossible happens you sure look smart for having implemented those good habits.

@Patrick, thanks for the in depth look at your process. As I’ve begun learning Rails lately, I’m impressed with how you and other in the Rails community implement great processes like this (and share your thinking behind them with the community to learn from).