Event 1135, 7031, or 7036 when the cluster service stops in Windows Server 2012

Applies to: Windows Server 2012 DatacenterWindows Server 2012 DatacenterWindows Server 2012 Standard

Symptoms


Consider the following scenario:
  • One of the network adapters fails in a Windows Server 2012 cluster that has built-in load balancing and failover (LBFO) to switch independent teaming and active-standby mode.
  • The server that has the network adapter is removed from the cluster membership after five seconds.
  • The cluster service on the server is automatically restarted after 60 seconds.
In this scenario, the server continues to work as expected even without the un-recovered network adapter. Meanwhile, packet loss occurs on the server.

Note This issue does not occur in Windows Server 2012 R2.

Events that resemble the following are logged when this issue occurs:

Cause


Failover is known to be less graceful in active-standby mode, and this is a known issue in a Windows Server 2012 environment that has LBFO teaming.

Workaround


Because the active-standby mode provides no benefits over the active-active mode, from the Microsoft Windows Clustering stand-point, we do not recommend an active-standby configuration. There are currently no changes to the current version as running active-active mode is the recommended teaming configuration.

Status


Microsoft has confirmed that this is a problem in the Microsoft products that are listed in the "Applies to" section.