Database Service Disruption
RESOLVED
2015-MAY-15 08:01UTC (RESOLVED)
Our systems engineering team has completed a post mortem / root-cause analysis. The conclusion is that our C2 MySQL Cluster detected multiple conflicting transactions and automatically shutdown in order to avoid a split-brain situation which would result in data loss. This is a “by design” fail-safe feature built-in to all master – master / active – active topology clusters. The feature exists to protect your data against corruption in the case that conflicting transactions can not be resolved within a certain time slot. Our team has concluded that a momentary network disruption while nodes were attempting to resolve the conflict caused split-brain protection to be triggered. This is a very rare event and we do not expect it to happen again any time soon. We apologize for any inconvenience this outage may have caused and we are doing our very best to make sure that the impact of events like these are minimized.
2015-MAY-15 07:33UTC (OPEN)
Our C2 Database cluster has been temporarily affected by site-wide interruption of service. Sites that are hosted by this cluster may have experienced a 10 minute interruption for uncached hits or when logged into wp-admin. Our systems engineering team is currently running a root-cause analysis to find out why this event happened.
For real-time status information, please visit the Pressidium Cloud Status page at https://status.pressidium.com
