Cisco UCS Monitoring Extension
Create Analytics Metrics
Copy the queries from this Gist and create analytics metrics from them. Your metric names should match the values in the gist.
Refer to the Create Analytics Metrics From Scheduled Queries documentation for details on how to do this.
Create Analytics Health Rules
Create health rules from the scheduled queries in the previous section. You may use the AppDynamcis Config Exporter* tool to import the health rules from this Gist into your controller. Config Exporter has limited analytics functionalities in Controller version 4.5.14, and may be completely depreciated in version 4.5.16 - so your mileage may differ depending on your controller version.
The other alternative is to manually create the health rules. Again,the health rules names should be exactly the same otherwise, you will be required to update the dashboard JSON file.
"name" : "UCS_StorageDisk_Health",
"name" : "UCS_Blade_Server_Health",
"name" : "UCS_RackUnit_CPU_Health",
"name" : "UCS_NumberOfFaults",
"name" : "UCS_RackUnit_PSU_Health",
"name" : "UCS_Chassis_PSU_Health",
"name" : "UCS_FanModule_Health",
"name" : "UCS_Adaptor_vNICs_Health",
"name" : "UCS_Chassis_IOM_Health",
"name" : "UCS_Fabric_Interconnect_Health",
"name" : "UCS_RackUnit_PSU_Health"
A full list of the health rule names, including UCS servers’ power supply and temperature telemetry baselining can be found in this Gist.
Using UCS_StorageDisk_Health as an example:
Condition
Monitoring the monitor
UCS Monitoring extension performs a health check on itself, ServiceNow connectivity (if in use) and connectivity to UCS Manager.
Navigate to the application that contains the tier ID you provided in the config.json file and the create the following health rules using exact name:
SNOW Connectivity Health
Health Rule Name: SNOW Connectivity Health
Condition
- No warning condition
- A metric value of 1 indicates failure
UCS Connectivity Health
Health Rule Name: UCS Connectivity Health
Condition
- No warning condition
- A metric value of 1 indicates failure
UCS Machine Availability Health
Health Rule Name: UCS Machine Availability Health
Condition
- No warning condition
- A metric value of 1 indicates success
Upload the dashboard
- Download the dashboard JSON file from this Gist.
- Get your analytics applicationName from the controller. Navigate to Analytics - Alert & Response - Health Rules. Select any health rule and note down the value under “Policy Executed On” column as indicated in the screenshot below
-
Open the UCS_Dashoard.JSON file in your favourite text editor and find and replace all instances of “AppDynamics Analytics-249” with the value from step 2.
-
Upload the dashboard
-
Update any missing health status manually - including the UCS extension health check status
-
Make it your own
Role Based Access Control
If you’d like to restrict access to the UCS dashboard, you’d need to create a UCS Dashboard viewer role. Note that this role needs to be able to query the faults, PSU and temperature schemas respectively. Follow the screenshots:
Navigate to the Administrator’s page and Add a role:
Under Applications, select Analytics and the application that contains the tierID you provided in the config.json file
Under Analytics, Events, Select Custom Analytics Events
Then add all 3 UCS schemas
Finally, assign the role to a user or group - depending on your need.
*Config Exporter - Ask your AppDynamics representative to give you the Config Exporter tool if you don’t already have it. Config exporter is used to migrate configuration between controllers or applications. The configuration can be imported directly into another controller/application or it can be download as a file.