asked on

Question on File Share Witness and quorum

We have two Exchange 2010 SP3 nodes in a DAG and a FSW. The two MB nodes are in the same Site but the FSW is in another Site. For whatever reason, our FSW became unvailable which should have been OK because the other two nodes were still up and running and we should not have lost quorum but the Cluster service on one of the nodes stopped and the mounted DB's on that node failed over to the other node. Everything is back to normal and I am suspecting a possible NIC glitch with the node that lost quorum and not a FSW issue but just wanted to confirm that in this setup, if the FSW goes down, the Cluster should not be affected?

SOLUTION

Mahesh Sharma

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

ASKER CERTIFIED SOLUTION

Amit

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

shadowtuck

ASKER

Thanks. I think the issue is with the MB node itself, not the FSW as I suspected. I checked the event log on the MB that was complaining and saw Event ID 1135 which I remember from another Cluster I managed was a heartbeat issue. I wound up calling Microsoft on that one and they had me increase the heartbeat thresholds to the values below:

SameSubnetDelay=2000
SameSubnetThreshold=10

This happened during a backup so I am suspecting the heartbeat timed out and the MB server thought quorum was lost and failed over its DB's to the other server. Was just trying to validate my configuration here so thanks.