6.5.1

Table Of Contents
n
vSphere HA attempted to restart the virtual machine but encountered a fatal error each time it tried.
n
Your cluster's shared storage is vSAN and one of the virtual machine's files has become inaccessible
due to the occurrence of more than the specified number of host failures.
n
Restart actually succeeded.
Solution
To avoid virtual machine restart failures, check that virtual machines become protected by vSphere HA
after they are powered on. Also, ensure that your admission control settings match your restart
expectations if a failure occurs. Maximizing the compatibility between virtual machines and hosts in the
cluster can also reduce the likelihood of restart failures.
Note For information on the factors vSphere HA considers for virtual machine restarts, see "Determining
Responses to Host Issues" in vSphere Availability.
Troubleshooting vSphere Fault Tolerance in Network
Partitions
When a vSphere HA cluster experiences a failure of the network that vSphere uses for inter-agent
communication (the management network), a subset of the cluster's hosts might be unable to
communicate with other cluster hosts. In this case, the set of hosts that can communicate with each other
are considered to be in a network partition.
A cluster partition impedes cluster management functions such as vMotion and can impact vSphere HAs
ability to monitor and restart virtual machines after a failure. This condition must be corrected as soon as
possible.
Network partitions also degrade the functionality of vSphere Fault Tolerance. For example, in a partitioned
cluster, a Primary VM (or its Secondary VM) could end up in a partition managed by a master host that is
not responsible for the virtual machine. When a Secondary VM must be restarted, vSphere HA does so
only if the Primary VM is in a partition managed by the master host responsible for it. Ultimately, you must
correct the network partition, but until that is possible, you must troubleshoot and correct any problems
that arise with your fault-tolerant virtual machines to ensure that they are properly protected.
Primary VM Remains in the Need Secondary State
A fault tolerant Primary VM can remain in the need secondary state even though sufficient resources are
available to start the Secondary VM.
Problem
vSphere HA might not restart the Secondary VM of a vSphere Fault Tolerance (FT) virtual machine pair
even though there are sufficient resources available.
vSphere Troubleshooting
VMware, Inc. 49