High Availability and Disaster Recovery for EV implementation

[JaneSmith]

Hi,

We are in the process of planning for EV 9 implementation and looking for an HA solution as well as a DR solution. 

Our ‘”production” EV infrastructure is as follows:

•         Single Site Configuration, with three physical EV servers configured to archive mail from Exchange 2010.

•         Clustered SQL 2008 Server which will be geo-clustered to replicate to DR site

•         Vault Store Data on NAS (CIFS Share), Indexes and SQL DB’s on SAN

Our typical clustering technology is MSCS.  I was thinking of clustering the three EV servers using MSCS for high availability for production.

For our DR, we want manual intervention.  I had these options in mind:

·         Creating three servers at the DR site and using Update Service Locations/Building blocks to move services and tasks

·         Add 3  additional passive nodes to the cluster or geo-cluster using w2k8

Is USL/ building blocks + cluster even supported?  I heard that with w2k8 clustering, all nodes are no longer required to be on the same LAN segment…?  Any ideas on the best HA and DR for this?

Thanks!

[JesusWept3 · Accepted solution]

Personally i would suggest engaging Symantec Professional Services for these type of indepth questions.
USL though in a single site environment that uses MSCS is redundant, however since you are going to be able to rollover to a complete second site thats clustered, USL would be appropriate at this point, just as long as DNS is changed properly.

Remember though, that USL is not HA, and when you figure in DNS changes, caches on machines etc, you may not meet the availability SLA using that solution