3.1.4.3.7.3 CLUSCTL_CLUSTER_CHECK_VOTER_EVICT

The CLUSCTL_CLUSTER_CHECK_VOTER_EVICT cluster control code queries the server for whether evicting the designated configured node from the cluster or changing the cluster quorum configuration such that the designated resource is no longer participating in quorum, as defined in section 3.1.1.10, will cause the loss of quorum.

Protocol version 2.0 servers do not support this control and MUST fail this method using error code 0x00000001 (ERROR_INVALID_FUNCTION).

The client MUST NOT invoke this method if the server is in a Shared Disk quorum type cluster, as defined in section 3.1.1.10.

The client MUST provide either a resource or node ID, as specified in sections 3.1.4.2.15 and 3.1.4.2.49 respectively, as a null-terminated Unicode string in the buffer that is designated by lpInBuffer. The server MUST fail this method by using ERROR_INVALID_PARAMETER if the resource specified is not the one currently configured as the quorum resource, as specified in section 3.1.1.10, or the node ID does not correspond to the ID of a configured node in the cluster.

 After successful completion of the method, the server MUST write a 32-bit integer containing one of the values from the CLUSTER_QUORUM_VALUE enumeration (as specified in section 2.2.2.10) to the buffer that is designated by lpOutBuffer, indicating whether quorum will be lost. If the lpOutBuffer buffer is not large enough to accommodate the enumeration value, error 0x000000EA (ERROR_MORE_DATA) MUST be returned.

 The server SHOULD accept a CLUSCTL_CLUSTER_CHECK_VOTER_EVICT cluster control code request if its protocol server state is read-only, and the server MUST accept the request for processing if it is in the read/write state, as specified in section 3.1.1.