Looks like the situation is resolved! Here are the details from our Status Alert mailing list:
The Haiku Learning team would like to report that we’ve resolved this morning’s incident and that Haiku Learning is functioning normally at this time.
At 7:52 a.m. PDT, a database replication delay issue with one of our database servers caused 6 minutes of downtime. Around 8:00 a.m. we removed the server suffering from delay out of production. Around 8:32 a.m., as part of an attempt to deal with the underlying cause of the replication delay, we executed a statement that overwhelmed our database server. This resulted in 9 more minutes of downtime. Haiku Learning has recovered and performance is now normal.
We are continuing to monitor the recovery and are taking optimization steps to remedy this issue long-term.
Thank you again for your patience and your understanding.
Best regards,
Justin