Michaelc
asked on
SQL 2005 Cluster both nodes believing it is the active node
On a couple of occasions my two node Active-Passive SQL 2005 Cluster has been completely down because both nodes believe itself to be the active node in the cluster, thus generating an IP conflict on the network.
The only way for me to get the cluster started again is to just restart cluster service on both nodes, one at a time.
It appears that on each occasion, the passive node looses network connection to the active node for a short period and starts th failover process.
Any ideas what is going on and how to prevent this from happening?
The only way for me to get the cluster started again is to just restart cluster service on both nodes, one at a time.
It appears that on each occasion, the passive node looses network connection to the active node for a short period and starts th failover process.
Any ideas what is going on and how to prevent this from happening?
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
I realized the stupidity of my response and set the heart beat network as private.
Think that could be what was causing the problem?
Think that could be what was causing the problem?
Quite possible.
Could review system event log ans cluster log to sure.
If the public nic is pulled on the active node should see the classic cluster fail over to inactive node.
Could review system event log ans cluster log to sure.
If the public nic is pulled on the active node should see the classic cluster fail over to inactive node.
ASKER
The public network is also setup as mixed.