If both primary and secondary clusters are down and are brought online again, it may happen that the replication is in error state

If both primary and secondary clusters are down and are brought online again, it may happen that the replication state is displayed as Error in spite of other services on the clusters being in healthy state. This may be caused if the vxnetd start daemon is in hung state on any node of the cluster. (IA-36540)

Workaround:

Restart the nodes on which 'vxnetd start' daemon is in hung state.