The value of our product depends on a reliable cloud service and it is our responsibility to do a good job of keeping that service available especially on busy holidays. The Thanksgiving load uncovered some software problems that we should have already discovered and fixed. We deeply apologize for this failure.
The most important thing we are doing now to make sure this doesn’t happen again is testing. We beefed up our development servers to match our production servers so that we can run load tests that significantly exceed our highest projections for usage levels during busy holidays. This testing will expose any additional system limitations and then we will fix them. We’ll keep doing this until we know the system can handle the busiest days.
Our system already includes geographically separate redundant servers (Southeastern US and Midwestern US) so that even if the Internet for a whole region of the US goes down, our cloud service will still be available, after the few minutes it takes to automatically switch servers from one region to the other.
On a related note, in 2018 we will add some networking features to our mobile apps and controllers that will allow users to monitor and control their Flame Boss controllers even if they have no Internet. You will be able to access your controller similar to the way computers use wireless printers – local and direct.
Very sorry again for our service outage, and thank you for the feedback.