Database shard failover

Incident Report for Braze, Inc.

Resolved

On Sunday, January 18th at approximately 9:08pm ET, one of database shards on a cluster that hosts less than 10% of our customers experienced a failure, affecting data collection from the SDKs intermittently for approximately 3 minutes for these customers. Our secondary database quickly took over for the failed component and the system healed itself. Our DBAs are investigating the source of the failure.

Posted Jan 18, 2015 - 21:08 EST