The following figure describes the leaf switch-spine switch topology for 1-HOP or 2-HOP flow telemetry correlation.

The following figure describes the leaf switch-spine switch topology for 3-HOP or 4-HOP flow telemetry correlation.

Supported scenarios: The Cisco Network Insights for Resources topology supports the following scenarios.

VXLAN

• VPC on leaf switch

• Border spine switch

• Border leaf switch

• IR or Multicast underlay

• EBGP or IBGP

• IPv4 underlay

• IPv4 or IPv6 overlay

Legacy spine switch or leaf switch

• VPC on leaf switch

• IPv4 or IPv6

The Cisco Network Insights for Resources (Cisco NIR) is a real-time monitoring and analytics application.

The Cisco NIR app consists of the following components:

• Data Collection—The streaming of telemetry data is done by the Operating System on the fabric nodes. As each data source is different and the format in which data is streamed is different, there are corresponding collectors running analytics that translate the telemetry events from the nodes into data records to be stored in the data lake. The data stored in the data lake is in a format that the analytics pipeline can understand and work upon.

  The following telemetry information is collected from various nodes in the fabric:

  • Resources Analytics—This includes monitoring software and hardware resources of fabric nodes on Cisco DCNM.

  • Environmental—This includes monitoring environmental statistics such as fan, CPU, memory, and power of the fabric nodes.

  • Statistics Analytics—This includes monitoring of nodes, interfaces, and protocol statistics on Cisco DCNM and fabric nodes.

  • Flow Analytics—This includes detecting anomalies in the flow such as average latency, packet drop indication, and flow move indication across the fabric.

• Resource Utilization and Environmental Statistics—Resource analytics supports configuration, operational and hardware resources. Environmental covers CPU, memory, temperature, and fan speed. System analytics also covers, Anomalies, and trending information of each resource and graphing of parameters which help Network operators to debug over period of time.

• Predictive Analytics and Correlation—The value-add of this platform is predicting failures in the fabric and correlating internal fabric failures to the user-visible/interested failures.

• Anomaly Detection—Involves understanding the behavior of each component while using different machine learning algorithms and raising anomalies when the resource behavior deviates from the expected pattern. Anomaly detector applications use different supervised and unsupervised learning algorithms to detect the anomalies in the resources and they log the anomalies in an anomaly database.

The following are the guidelines and limitations for the Cisco Network Insights for Resources (Cisco NIR) application in the Cisco Data Center Network Manager (Cisco DCNM):

• After upgrading Cisco DCNM or Cisco NIR app to new version and before starting Cisco NIR app, make sure the following are set:

  • Navigate to Applications > Preferences from Cisco DCNM Configuration page and modify Telemetry Network Configuration to the desired value.

    Note	The Out-Of-Band is a default value for the interface, which may not be what you set prior to upgrade.

  • Click Submit.

• When you did not modify the telemetry network configuration post Cisco DCNM or Cisco NIR app upgrade and enabled telemetry on any fabrics from Cisco NIR, then the application is not enabled and configured properly.

  • Login to Cisco DCNM active node using SSH client as root. Incase you are already logged into Cisco DCNM active node, change to root.

  • Execute the following command.

    curl -d '{"AppName": "NIR"}' http://127.0.0.1:9595/telemetry/force_cleanup_app

  • Execute the following command.

    curl -d '{"AppName": "NIR"}' http://127.0.0.1:9595/telemetry/force_cleanup_hw_app

  • After few minutes all the fabrics in Cisco NIR Configuration page show a disabled state.

  • Once the status is disabled for all the fabrics, modify Telemetry Network Configuration in Cisco DCNM to the desired value.

  • Enable telemetry on the fabrics from the Cisco NIR Configuration page.

• For Flow Telemetry the Cisco NIR app captures the maximum anomaly score for a particular flow, for the entire cycle of the user specified time range.

• If Strict Config Compliance (SCC) is enabled in a fabric, you can not deploy Cisco NIR on Cisco DCNM.

• To enable telemetry on monitored fabric through Cisco NIR app, you must first delete all existing telemetry configurations on all the nodes in the monitored fabric before you enable this fabric from Cisco NIR app. The telemetry then assigns the receiver IPs to these nodes, which the Health page displays. The telemetry configuration will not push any telemetry configurations to the nodes because they are monitored. Therefore you have to check the receiver IPs from the Health page and must configure the nodes manually.

• After enabling telemetry in Cisco NIR app, to upgrade or downgrade a switch follow these steps:

  • Remove the switch from the fabric that needs upgrade or downgrade. Then upgrade or downgrade the switch image and add it back to the fabric.

  • Or, disable telemetry on the fabrics where switches need upgrade or downgrade. Then upgrade or downgrade the switch image and then enable telemetry on the fabrics.

• IPv6 is not supported for receiving telemetry data for Cisco NIR app.

• The Cisco NIR application requires that physical servers hosting Cisco DCNM computes as VMs are atleast Cisco C220-M4 category. It is also required that a compute be hosted on a data store with a dedicated hard disk of atleast 500GB. See Hardware Requirements.

• For instances where one or more fabrics do not recover from disabling state, you must stop and restart the Cisco NIR application in the Cisco DCNM. This will recover the failed disable state.