Applicable to ENA v17.0 P08 upwards. If you are using an earlier version of Entuity, please see this article.
Introduction:
Entuity will start managing a Cisco APIC as soon as it has been added to your network. Please see the sections below for information on how it is managed.
See also this article for help and information on how Cisco APIC data is displayed in Entuity dashboards.
Measuring Cisco ACI health:
There are 10 measurements available from the APIC. These are:
| Metric Name | Metric Description |
|---|---|
| APIC CPU | CPU utilization % of the Cisco APIC. |
| APIC Memory | Memory utilization % of the Cisco APIC. |
| EPG Health | based on the health score of each EPG component, as well as the number of learned endpoints on the leaf. |
| FabricNode Health | fabric node health. |
| FanTray Health | fan tray health. |
| LineCard Health | linecard health. |
| Pod Health | based on the leaf and spine switches health scores, as well as the number of learned endpoints on the leaf switches. |
| PSU Health | power supply health. |
| SupervisorCard Health | supervisor card health |
| Tenant Health | aggregate of the tenant-wide logical object health scores, across the infrastructure they use. |
Entuity uses the APIC concept of health. For all metrics (except for APIC CPU and APIC Memory), APIC health is measured as a percentage:
- 100% health = healthy.
- 0% health = unhealthy.
Cisco advises that if health drops below 90%, then there is a problem. Therefore, Entuity sets 90% as the default health threshold. When a metric's health drops below 90%, Entuity generates an event on that metric.
For the APIC CPU and APIC Memory metrics, Entuity also uses a percentage, but the are not based on health. Instead, they are based on usage:
- 100% usage = unhealthy.
Cisco ACI faults and events:
Cisco ACI faults are an APIC concept. From ENA v17.0 P08 upwards, Cisco ACI faults are mapped to the following Entuity incident and events:
ACI Fault incident, which has the following events:
- ACI Fault - raised to open or update the incident.
- ACI Fault Cleared - raised to clear the incident.
If you want to see details of ACI faults on your APIC objects, however, you can create a custom dashboard that contains an ACI Faults dashlet. In this dashlet, you can see the Cisco ACI faults as displayed by the device. This can be useful for seeing what could be causing a Cisco APIC health-related event or incident. For example, if events and incidents have been raised against a particular Cisco APIC health metric (as described in the section above), you can use the ACI Fault dashlet to view the ACI faults displayed on that component.
Applicable to Entuity v19.0 P03 upwards
From Entuity v19.0 P03 upwards, the Entuity virtualization process (as part of its periodic polling, by default 5 minutes) polls APICs for events and raises them as the event type 'ACI Event'. The source of the event is the APIC device itself, and the identity of the event includes the distinguished name of the source in the APIC plus the event code.
The settings for polling APICs can be configured via the [aci] section of entuity.cfg. Please see this section for further help and information on configuring polling of APICs.
If there are multiple ACI Events from the same APIC with the same event code and same distinguished name (dn), these events are all included as part of the same incident. Events with different event codes or dn will result in different incidents.
To change a health threshold:
- Navigate to the Cisco APIC for which you want to change the health thresholds.
- Open its Threshold dashboard, and ensure the Thresholds Related To field is set to ACI.
- Follow the instructions in this article on overriding thresholds to change the health thresholds for the selected Cisco APIC.
Comments
0 comments
Please sign in to leave a comment.