6.6

Upon detection of missed heartbeats vCenter Server Heartbeat performs the following steps:
1 As soon as the active node detects that the VMware Channel is experiencing missed heartbeats, it
determines if a valid failover target (the passive node) is present.
Simultaneously, once the passive node detects missed heartbeats, it determines if it is a valid failover
target.
2 Next, the active node determines if it is synchronized with the failover target (the passive node). If
synchronized, it continues to the next step. If it is not synchronized, it vetos a failover.
Simultaneously, the passive node checks to see if it is synchronized with the active node. If synchronized,
it continues to the next step. If it is not synchronized, it vetos a failover.
3 At this point, both the active and passive nodes check the configured value of the Failover timeout and
start a "Heartbeat lost" countdown. Both nodes start the countdown at approximately the same time.
4 Failover or isolation of the active node is postponed until the configured Failover timeout value (in seconds)
has elapsed and it is during this period that both nodes accomplish steps 1 & 2.
5 Once the configured Failover timeout period has elapsed, the active node assumes the VMware Channel
is lost and attempts to ping the failover target (passive node) via the Public network. If the ping is successful,
active node isolation is vetoed. If the attempt to ping the failover target is unsuccessful, the active node
proceeds to the next step.
Simultaneously, the passive node assumes the VMware Channel is lost and attempts to ping the active
node via the Public network. If the ping is successful, failover is vetoed. If the ping attempt is unsuccessful,
the passive node proceeds to the next step.
Note If the nodes have reached this point, then neither node can see the other node.
6 Both the active and passive nodes check their connectivity to the Public network. If the active node has
lost connectivity to the Public network, it isolates itself by making itself passive (potential active).
Should the active node reconnect with the passive, it becomes active again. Otherwise, it remains passive.
If the passive node has lost connectivity to the Public network, it vetos a failover.
Application Monitoring
vCenter Server Heartbeat monitors protected applications and services to ensure that applications are available
to clients by monitoring application and service status.
The Applications: Summary page displays the active node identity, application state and health, details of
application types, and their corresponding running status and health.The lower pane provides an Applications
Log that allows viewing of application events as they occur. This page also provides controls to edit, remove,
start, and stop applications, and to configure and edit the configuration of all protected applications.
Viewing Application Status
After an application successfully starts and is running, you can view application status in the Applications pane
of the Applications: Summary page. If an application fails, right-click the event in the Applications Log and click
on Properties to invoke the Event Properties dialog and investigate the failure.
34 VMware, Inc.
vCenter Server Heartbeat Administrator's Guide