I’m writing to report and explain a limited incident this morning that affected a small subset of users. One of our application server pools did not complete its regularly scheduled maintenance on time and was prematurely opened for new connections. As a result, the new users directed to this pool were not able to register successfully, and returning users experienced significant delays (up to 1 min) at login.
No data was affected and no transfers were affected, and all has returned to normal.
A combination of vigilant reporting, experienced design and architecture, and good fortune has allowed us to avoid most issues (even narrowly contained incidents such as this one), but we think it is important to report them to our user base when we do experience a problem.
Please let us know if you see any other issues and we’ll respond as soon as possible.