Hi,We are considering using availability groups for our solution. It will have 4 SQL Server 2014 instances running on 4 cluster nodes, spread across two cities (two nodes in each city).All those instances will be online at all times. One will be the primary, two others will be synchronous replicas and one last - asynchronous replica.We are concerned about one scenario: should the link between two cities go down, one of the synchronous replicas, which is in another city would not be able to get updated and the transaction will fail on the primary replica. We can't allow this to happen, as the system must be available for read and write at all times.Our solution must have zero data loss, so I was wondering if there is a fix for this scenario.Thanks.
↧