Business Continuity RRS feed

  • Question

  • Hello,

    I am trying to develop a plan for handling service interruptions and disasters in Azure. I took quite some info from this article: https://msdn.microsoft.com/nl-nl/library/azure/dn251004.aspx
    A solution with a traffic manager, and RA-GR SQL + Storages in a paired region is fairly easy to implement.

    However, if I look at a case like this:

    Azure Active Directory - West Europe and North Europe - Partial Service Interruption
    29 minuten geledenStarting at approximately 09:00 on 3rd Dec, 2015, customers began experiencing intermittent issues accessing Azure services that use, or have dependencies on, Azure Active Directory. Engineering teams have identified an Azure Active Directory configuration error that is believed to be the root cause of the issue. Prior to executing a failover, the Azure Active Directory team identified an issue with the failover mitigation path, which would have blocked the mitigation. The team is now implementing an update to Azure Active Directory front ends to call a known good configuration. Customers may see some improvement as a result of the update to the front end. An update will be provided in 60 minutes, or as events warrant.

    I have seen more outages coming by that affect a whole paired region.

    A solution with RA-GR in a paired region wouldn't help (although our webservices don't use Active Directory, but maybe this could as well happen to a storage or network infrastructure). 
    So I would spend alot of extra money and effort in a failover infrastructure that doesn't really fail over...

    What options do I have? Going to Amazon for my secondary region with custom syncing?


    Thursday, December 3, 2015 12:34 PM