Celerra/Clariion/NS120 - Failover not automatic?

[YokyDoky]

The system has been purchased, installed, data migrated to it, and an mirror DR site brought up on the other side of town. Now the EMC tech tells me failover is not automatic. So, a meteor turns the datacenter into a crater, but since I'm shooting down a zipline in Belize at the time, the DR site just sits there with it's complete replica, happily doing absolutely nothing until someone gets me on the phone or I come home to push the buttons? Really?

Y. Doky - "Manager" of Systems

"A .22 caliber intellect in a .357 Magnum world."

 

[mrdenny]

Correct, something needs to tell the DR site that it is the master now. If you have VMware for your VMs and you are using Site Recovery Manager it can handle this for you. If you have clustered SQL Servers and a single cluster spread between sites the cluster can handle this for you.

You've got a good start, but it looks like you still have some work to do before your DR system is setup.

Denny
MVP
MCSA (2003) / MCDBA (SQL 2000)
MCTS (SQL 2005 / SQL 2005 BI / SQL 2008 DBA / SQL 2008 DBD / SQL 2008 BI / MWSS 3.0: Configuration / MOSS 2007: Configuration)
MCITP (SQL 2005 DBA / SQL 2008 DBA / SQL 2005 DBD / SQL 2008 DBD / SQL 2005 BI / SQL 2008 BI)

My Blog

 

[Cretin]

Some failover is automatic from an EMC Array perspective. You have 2 switches, 2 storage processors, and 2 paths in each array. Failover is automatic on those assuming that your product such as powerpath or multipath is functioning correctly. Failover from a DR perspective is not automatic unless you do something on the WAN side to failover in the event of an outage. The DR array and servers would need some way of telling that they are now primary.

Cretin

 

[comtec17]

All responses here are correct. EMC NS does not automatically failover to a DR site as it is a safety measure to prevent a split brain situation. Can you imagine the headache you could get if the DR site decided to promote itself all the sudden?

Now, you can create a set of monitoring scripts and deploy them on a monitoring station that will automatically promote the DR site in the loss of the primary.