Early Tuesday morning our user database kicked the bucket. Technically, one of our underlying database files became corrupted, but the result was that no one could sync. And for a company in the sync business, that's no good.
Of course we expect these kinds of problems to crop up from time to time, and we back up our data accordingly. Unfortunately, it turned out to be a lot harder and more time consuming to restore from our backups than we had planned.
The good news is we were able to get back up and running without spilling a drop of data. The bad news is it took us almost a day and a half, and we were offline during that time. And frankly, that's embarrassing. We hold ourselves to a higher standard than that, and we failed to meet it.
So, what to do? Well, a couple of things.
First, we're revamping our internal backup and restore processes so we'll be able to recover from problems like this much more quickly in the future.
Second, we're putting in place redundant systems so problems like this don't even require a restore from backup; instead they simply trip a switch that causes the redundant system to take over with no interruption in service.
And finally, we're saying thank you to our customers and trial users for your patience. We were offline for a day and a half, so we've added three days of service to every customer's paid subscription and every trial user's free trial period. We haven't looked it up, but we believe that's a better service credit you'd get from any of the big guys who offer Service Level Agreements.
Regardless, if you're a paid customer and you'd like a refund, just let us know at firstname.lastname@example.org. We'll refund your purchase, no questions asked.
We really value what we do, and we believe our customers do too. We had a predictable problem and failed to solve it in a satisfactory amount of time. We'll do our best to make sure that doesn't happen again.