Managed Object -
FailoverClusterManager(vim.vcha.FailoverClusterManager)
- Property of
-
ServiceContent
- See also
-
VchaClusterHealth
- Since
- vSphere API 6.5
Managed Object Description
FailoverClusterManager provides operations to manage a vCenter
High Availability Cluster (VCHA Cluster).
A VCHA Cluster consists of three VMs. One is the Active vCenter VM that
serves client requests. Second is the Passive VM that is identical to the
Active vCenter VM in terms of resources and capabilities. Passive VM
constantly receives updates from Active VM and takes over the role of
Active vCenter VM in the event of failover. Third is the Witness VM that
acts as a quorum VM in a VCHA Cluster. Sole purpose of Witness VM is to
avoid classic split-brain problem in a VCHA Cluster.
A VCHA Cluster has following states -
1. Healthy - All three nodes in a VCHA Cluster are healthy and connected.
State replication between Active and Passive node is working and both
nodes are in-sync.
2. Degraded - A VCHA Cluster is said to be in degraded state when it has
lost one of the three nodes. Node loss can be due to various reasons and
as a result the lost node is not visible to other two nodes. If an Active
node is lost, Passive node will take the role of Active node. If Passive or
Witness node is lost, Active node will continue to serve requests. A VCHA
Cluster can also be in degraded state if state replication fails between
Active and Passive nodes.
3. Isolated - All three nodes are isolated from each other. If this happens
while VCHA Cluster is in enabled mode, Active node stops serving client
requests. If nodes are isolated in a disabled VCHA Cluster mode, Active
node continues to serve client requests.
A VCHA Cluster has following modes -
1. Enabled - State replication between Active and Passive nodes is
enabled and automatic failover happens if Active node fails while the
VCHA Cluster is in a healthy state.
2. Disabled - All three nodes are part of VCHA Cluster but state
replication and automatic failover is disabled.
3. Maintenance - All three nodes are part of VCHA Cluster but automatic
failover is disabled while state replication continues. Active node continues
to serve client requests even if Passive and Witness nodes are lost.
Properties
|
Name
|
Type
|
Description
|
| disabledClusterMethod | xsd:string[] |
A list of method names that must not be called and will throw
a fault due to some other method running that the disabled method
can cause side-effects for.
This list may include the following methods:
GetClusterHealth will also be disabled if Deploy is in progress.
As with other disabled methods there will be no property updates
on this property when called with non-zero property collector versions.
|
Methods
getClusterMode
Returns current mode of a VCHA Cluster.
- Required Privileges
- System.Read
Parameters
Return Value
|
Type
|
Description
|
| xsd:string | |
Faults
|
Type
|
Description
|
| RuntimeFault | Thrown if any type of runtime fault is thrown that is not covered by the other faults; for example, a communication error. |
Events
Show WSDL type definition
GetVchaClusterHealth(getClusterHealth)
Returns last known health of the VCHA Cluster.
- Required Privileges
- System.Read
Parameters
Return Value
Faults
|
Type
|
Description
|
| RuntimeFault | Thrown if any type of runtime fault is thrown that is not covered by the other faults; for example, a communication error. |
Events
Show WSDL type definition
initiateFailover_Task
Allows a caller to initiate a failover from Active vCenter Server node
to the Passive node. By default it is a forced failover. The planned
flag can be used to initiate it as a planned failover.
For forced failover, Active node immediately initiates a failover. This
may result into a data loss after failover.
For planned failover, Active node flushes all the state to the
Passive node, waits for the flush to complete before causing a failover.
After the failover, Passive node starts without any data loss.
A failover is allowed only in the following cases -
1. Cluster's mode is enabled and all cluster members are present.
2. Cluster's mode is maintenance and all cluster members are present.
API throws an exception in all other cases.
- Required Privileges
- Global.VCServer
Parameters
| Name | Type | Description |
| _this | ManagedObjectReference |
A reference to the
FailoverClusterManager
used to make the method call.
|
| planned | xsd:boolean |
- if false, a failover is initiated immediate and may
result in data loss.
if true, a failover is initated after the Active node
flushes its state to Passive and there is no data loss.
|
Return Value
Faults
|
Type
|
Description
|
| RuntimeFault | Thrown if any type of runtime fault is thrown that is not covered by the other faults; for example, a communication error. |
Events
setClusterMode_Task
setClusterMode method allows caller to manipulate the mode of a
VCHA Cluster
Following mode transitions are allowed -
enabled -> disabled - Allowed only in healthy and degraded states.
enabled -> maintenance - Allowed only in healthy state.
disabled -> enabled - Allowed only in healthy state.
maintenance -> enabled - Allowed only in healthy state with all nodes
are running the same version.
maintenance -> disabled - Allowed only in healthy state with all nodes
are running the same version.
All other transitions are not allowed.
VCHA Cluster configuration remains intact in any of the cluster modes.
- Required Privileges
- Global.VCServer
Parameters
Return Value
Faults
|
Type
|
Description
|
| RuntimeFault | Thrown if any type of runtime fault is thrown that is not covered by the other faults; for example, a communication error. |
Events