Update at 10:58 AM Eastern: We have initially identified the cause of these two incidents as a code change deployed yesterday by Automattic to WordPress.com. We believe the change unintentionally generated uncached database query activity at a level that began degrading performance, and eventually resulted in the database being unable to accept new queries, which caused the first incident where VIP sites were unavailable.
While we were able to get WordPress.com database systems to recover initially, we had not yet tracked down the code change, and so database performance again degraded to the point of being unavailable. Shortly after that we reverted the related code, which we believe will prevent any further recurrence.
We’ll review our code review process with the aim of avoiding a repeat of this situation in the future.
We apologize for the inconvenience.
Update at 10:01 AM Eastern: this issue recurred around 9:50 AM Eastern, recovering by 10:00 AM Eastern, and we posted a second Lobby update noting that.
Original Update: Today around 9:05 AM Eastern there was an issue on WordPress.com that briefly affected the availability of VIP sites. All sites returned to normal operation around 9:10 AM Eastern.
We are investigating the root cause and will update this post with more information as soon as it’s available.
Our initial understanding is that this issue did not affect VIP Go sites.
If you have any questions, please contact us via support ticket.
You must be logged in to post a comment.