- when there's a planned outage for maintenance
- when new versions of software are deployed
- as soon as we notice that either service is unavailable for any reason
In case of any unplanned outages, we'll keep you informed here. After the service is back up, we'll tell you exactly what went wrong, we'll investigate the root cause of the problems, and we'll tell you what we're going to do in the future to correct the root cause.
Fog Creek has adopted a policy modeled on Six Sigma's Five Whys to improve the stability and reliability of our production servers in New York and Los Angeles.
Our policy for handling any outage is to investigate the cause, and then ask "why?" repeatedly until we can identify the root cause of the problem. We're going to do this publicly, on this blog, to document what we hope will be a process of continuous improvement in the stability and availability of our servers. We are committed to telling our customers what went wrong, why, and what we're doing to insure that it doesn't happen again.
This blog is maintained on TypePad, which is completely separate from Fog Creek's operations, so it will be available even when we're having trouble.
We strongly recommend that all Fog Creek customers subscribe to this feed in an RSS reader so that if you can't reach our site, you don't have to try to remember this site's URL.