Introduction
The Cisco UCS Management Pack comprises several key features, which are listed here briefly. Subsequent chapters in this document elaborate these features further.
This chapter includes the following sections:
One Monitor per UCS Domain Fault
This management pack implements one monitor per UCS domain fault. This provides improved power of customization at the fault level. You can override parameters, such as changing the priority and severity, enabling or disabling monitors from the Operations Manager console interface. Apart from the console interface, all these customizations can also be applied using much more powerful Operations Manager cmdlets. PowerShell scripts could be developed to operate on multiple monitors at a time. Monitors are not implemented for informational messages from UCS domain. Informational messages will not be captured by the management pack in Operations Manager.
UCSM FSM faults are transient faults, therefore not supported by this version of Cisco UCS Management Pack. For a complete list of FSM faults in the UCS Domain not supported in the Management Pack, go to the following URL:
http://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/ts/faults/reference/2-0/UCSFaultsErrorsRef_20/UCS_FSMs.html
All monitors corresponding to UCSM faults of type ‘Configuration’ are disabled by default. Only the monitors with severity as ‘Critical’ are enabled by default in the Management Pack. Users can customize rest of the monitors as per their environment/requirement.
Knowledge Articles in Alerts
Knowledge articles are available for all the monitors implemented in the Management Pack. Whenever there is an alert, you can check the knowledge article associated with the alert from the Alert Details section or Health Explorer. Knowledge articles provide the details of fault, such as fault code, name, cause, explanation and list of steps to resolve it.
Installing the Management Service outside the Resource Pool
This release of the Management Pack supports installing the Management Service outside the Resource Pool. Once the Management Pack is imported in the Resource Pool, user can choose to install the Management Service on the Agent Managed Computer (on trusted or untrusted boundaries) or Gateway Server.
Event Based Discovery
Event based discovery is a unique feature in this Management Pack, by which the object discoveries executes only when there is some change detected in the inventory of UCS Domain. Intelligence is built in the Cisco UCS Management Service to detect any change in inventory and raise appropriate events to trigger the Object discovery and pull the changes in Operations Manager.
Real Time UCS Fault monitoring
This Management Pack supports real time UCS monitoring, which means that any fault on the UCS Domain is captured in the Operations Manager within a few seconds.
Dedicated Service Machine for UCS Domain Monitoring
When you choose a service machine while adding a UCS Domain in the Add Monitoring Wizard, it means all the monitors, object discoveries, and rules required to monitor an UCS Domain execute on the Service Machine dedicatedly and do not use the resources on any other computer for monitoring this particular UCS Domain. The Management Servers however, will continue to analyze the return data from the Service Machine and store them in the database. A single Service Machine can also monitor multiple UCS Domains.
While adding a UCS domain for monitoring using Add Monitoring wizard, the selected service machine should have Cisco UCS Management Service installed and running on it.
Power State Change Notification for Blade and Rack Servers
From Release 3.1(2), management pack generates warning and critical alerts when power state on associated blade and rack servers changes to the following:
Power states for warning alert:
- Test
- Off
- Offline
- Offduty
- Power-save
Power states for critical alert:
- Unknown
- Degraded
- Error
- Failed
- Not-supported
Depending on the severity of the alert, health state of the server also changes. When the power state of the server is restored to On, Online, or Ok state, alert is closed in operations manager and health state is reset to the state before the power state alert. This functionality is not applicable for unassociated servers. If a server is unassociated, any existing alert on the server related to power state closes in Operations Manager.
Proxy Server Support
Effective with Release 3.1.1, you can configure the proxy server while adding a UCS Manager to the management pack using the Add Monitoring Wizard.
PowerShell Cmdlets
Effective with the release 3.1(1), Cisco UCS Management Pack supports the use of cmdlets that can be imported from a PowerShell Module. You can use these cmdlets to perform actions such as discovery of chassis along with enabling or disabling monitors on components.
PowerShell 3.0 must be installed on machine to use these cmdlets.
Additional Features
Aside from the features listed above, the Cisco UCS Management Pack comprises the following additional features:
- A KVM Console can be launched on a Service Profile, Blade Server, or Rack Unit.
- Tech Support utility facilitates collecting logs for diagnosing an issue.
- Run-As Account association in the Add Monitoring Wizard simplifies the overall process to start monitoring a UCS Domain.
- Supports more secure account distribution.