6.5

Since a managed switchover cannot be performed during synchronization, it is important to review the queue
information prior to attempting a managed switchover. If the queues are large, le operations on the active
server are high and for this reason it may be prudent to delay a managed switchover due to the length of time
required to completely clear the queue. Queue lengths can be viewed in the Data: Trac/Queues page of the
vCenter Server Heartbeat Console.
Prior to performing work on the Primary server, a managed switchover can be triggered by selecting the
Secondary server and clicking Make Active on the Server: Summary page. This changes the server roles such
that the active server becomes passive and the passive server becomes active. This means users are able to
work continuously while the Primary server is o line.
When the Primary server is back up and running, the managed switchover can be triggered again so that the
passive server becomes active and the previously active server becomes passive.
Note The managed switchover process may be performed at any time as long as the systems are fully
synchronized with respect to data les and registry replication. Switchovers cannot be performed if either server
is in an unsynchronized or unknown state.
Recovering From a Failover
A failover diers from a switchover. A switchover is a controlled switch (initiated manually from the vCenter
Server Heartbeat Console, or initiated by vCenter Server Heartbeat when precongured) between the active
and passive servers. A failover happens when any of the following fail on the active server: power, hardware,
or VMware Channel communications. The passive server waits a precongured time after the rst missed
heartbeat before initiating a failover. When this period expires, the passive server automatically assumes the
active role and starts the protected applications.
The following recovery scenario is based on a vCenter Server Heartbeat conguration with the Primary server
as active and the Secondary server as passive. A failover has occurred and the Secondary server is now running
as the active server.
Note When failover conditions, such as a power failure, cause failures in both active and passive servers, a
condition may result that causes both servers to restart in passive mode. In this situation, manual intervention
is required. See “Two Passive Servers,” on page 85.
Procedure
1 Review event logs on both servers to determine the cause of the failover. For assistance, use the Log Collector
(launched from the Taskbar icon) tool to collect information and send the output to VMware Support.
2 If any of the following issues exist on the Primary server, performing a switchback to the Primary server
may not be possible until other important actions are carried out. Do not restart vCenter Server Heartbeat
until the following issues have been resolved:
■
Hard Disk Failure – Replace the defective hard disk.
■
Power Failure – Restore power to the Primary server.
■
Malware – Clean the server of all malware.
■
Communications – Replace or repair the physical network hardware.
■
Blue Screen – Determine cause and resolve. As required, submit the dump le to VMware Support
(www.vmware.com/support) for analysis.
3 Run the Congure Server wizard and verify the server identity is set to Primary and the Active Server is
set to Secondary. Click Finish to accept the changes.
4 Disconnect the VMware Channel network cables or disable the network card.
36 VMware, Inc.
Administrator Guide