After running a disaster recovery operation, the recovery plan displays the following state:
“Run recovery again. The recovery operation was not able to complete all of the steps at the protected site. If the plan was run while the sites were disconnected, run the plan again immediately. If the sites were connected during recovery, there might be errors with the VMs. Check the plan history to view errors that might need to be corrected before running the plan again. Running recovery again will not affect the recovered VMs at the recovery site.”
The recovery plan history reports errors similar to:
“Cannot unmount datastore ‘#####’ from host ‘######’. Unable to communicate with the remote host, since it is disconnected.”
VMware Live Recovery 9.0.2
One or more ESXi hosts at the protected site (DC) were in a Not Responding or Disconnected state during the recovery operation.
This prevented Site Recovery Manager (SRM) from communicating with the affected host(s) and unmounting the protected-site datastores required to complete the recovery plan.
During disaster recovery:
Virtual machines from Site A are powered on at Site B without protection.
When Site A becomes available again, protected-site virtual machines and datastores must be powered down and unmounted before protection can be reversed.
If any protected-site ESXi host is disconnected, datastore unmount operations fail, causing the recovery plan to remain in a Recovery Required state
Example diagram :
Cause Validation :
Review of the vmware-dr.log file confirms datastore unmount failures caused by host disconnection. The following log entries indicate HostNotConnected faults at the time of recovery:
Log path : less /opt/vmware/support/logs/srm/vmware-dr.log
2026-01-13T21:00:23.433+02:00 verbose vmware-dr[01283] [SRM@6876 sub=DrTask ctxID=### opID=74e#####c-retrieveHistoryWindow] [5247f] Task 'dr.#######' completed with result: (dr.recovery.RecoveryResult) [--> (dr.recovery.RecoveryResult) {--> runKey = ######,--> operation = "failover",--> options = (dr.recovery.RecoveryOptions) {--> note = <unset>,--> syncData = false,--> plannedFailover = true,--> migrateEligibleVms = true,--> skipProtectionSiteOperations = true,--> autoAnswerPrompts = false,--> recoveryPoint = <unset>--> },--> plan = 'dr.recovery.RecoveryPlan:####aa########',--> planName = ######",--> planDescription = "",--> user = "O#####r",--> errorIpCustomizedVms = 1,--> poweredOffVms = 0,--> warnings = <unset>,--> errors = (vmodl.MethodFault) [--> (dr.fault.MultipleFault) {--> faultCause = (vmodl.MethodFault) null,--> faultMessage = <unset>,--> faults = (vmodl.MethodFault) [--> (dr.storage.fault.UnmountDatastoreFault) {--> faultCause = (vmodl.fault.HostNotConnected) {--> faultCause = (vmodl.MethodFault) null,--> faultMessage = <unset>--> msg = "Unable to communicate with the remote host, since it is disconnected."--> }, faultMessage = <unset>,--> hostName = #######,--> host = 'vim.HostSystem:e#######5',--> datastoreName = "#########",--> datastore = 'vim.Datastore:ed######5'--> msg = "Cannot unmount datastore '#######' from host '######'. Unable to communicate with the remote host, since it is disconnected."--> },--> (dr.storage.fault.UnmountDatastoreFault) {--> faultCause = (vmodl.fault.HostNotConnected) {--> faultCause = (vmodl.MethodFault) null,--> faultMessage = <unset>--> msg = "Unable to communicate with the remote host, since it is disconnected."--> },--> faultMessage = <unset>,--> hostName = "#########",--> host = 'v######',--> datastoreName = "######",--> datastore = '##########--> msg = "Cannot unmount datastore '#########' from host '########'. Unable to communicate with the remote host, since it is disconnected."--> }--> ]--> msg = "Cannot unmount datastore '#######' from host '#########'. Unable to communicate with the remote host, since it is disconnected. Cannot unmount datastore '#####' from host '#######'. Unable to communicate with the remote host, since it is disconnected."Resolve the Not Responding or Disconnected state of the affected ESXi host(s) at the protected site.
Once host connectivity is restored, rerun the recovery plan to allow SRM to successfully unmount the protected-site datastores and complete the recovery operation.