Date: October 21, 2019 (Monday)
At approximately 1:53 PM (MDT), a database cluster member experienced a failure. As designed, the database cluster automatically began its failover process to a partner member, which included transaction playback to ensure data integrity. The database cluster was fully restored and operational by 2:12 PM (MDT). Database cluster failover timing is dependent on the volume of data that is “in flight” during a failure as the “in flight” data needs to be processed by the partner member prior to operational status to insure no loss of data. As can be expected, ImageSilo database clusters have a significant volume of data being continuously added and updated. The failed cluster member was replaced and full redundancy has been restored.
As part of our high availability architecture, hardware failures are expected to occur regularly, and this particular failure recovered automatically as designed. A limited number of customers whose data resides on this particular cluster were affected. We apologize for any inconvenience this interuption may have caused. As always, we sincerely appreciate your business.
ImageSilo Administration
866.374.3569 (303.493.6900)
siloadmin@imagesilo.com