Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?

 
 

Collect Analytics Data Overview

Paragon Automation supports collecting analytics data from Juniper Networks, Nokia, and Cisco IOS XR devices.

  • Table 1 lists the data collected from Juniper Networks devices.
  • Table 2 lists the data collected from Cisco IOS XR devices.
  • Table 3 lists the data collected for Nokia devices.
Table 1: Data Collected for Analytics: Juniper Devices

Data

Collected by Default

Ingest Type

Rules Used to Collect Data

Label-switched Path (LSP) statistics [RSVP, Segment Routing (both colored and non-colored)]*

Note:

Currently, Paragon Pathfinder does not support the collection of SRv6 LSP statistics.

Yes

JTI

controller-telemetry /ctrl-label-switched-path

controller-telemetry /ctrl-label-switched-path-aggregation

Logical and physical interface statistics

Yes

JTI

controller-telemetry /ctrl-logical-interface

controller-telemetry /ctrl-logical-interface-aggregation

controller-telemetry /ctrl-physical-interface

controller-telemetry /ctrl-physical-interface-aggregation

Segment ID (SID)

Yes

JTI

controller-telemetry /ctrl-sr-sid

Link Latency

Yes

iAgent

controller-telemetry /ctrl-link-latency

LSP Latency

Yes

Derived from Syslog

controller-telemetry /ctrl-lsp-latency

Quality of Service (QoS)

No

JTI

controller-telemetry /ctrl-egress-queue-interface

controller-telemetry /ctrl-ingress-queue-interface

Label Distribution Protocols (LDP) demands

No

iAgent

controller-telemetry /ctrl-ldp-demand-stats

Netflow Yes Not Applicable controller.telemetry/ctrl-demand-stats

controller.telemetry/ctrl-as-demand-stats

Note:
  • For collecting SR LSP statistics from a Juniper Networks device, the device should be running Junos OS Release 20.2 or a later version.

  • For collecting LSP latency, the lsp-latency-interval must be set. The default value is 180s. You can modify this value; see Modify Pathfinder Settings From the GUI or Modify Pathfinder Settings From the Pathfinder CLI for details.

  • The segment routing LSPs created by using node SIDs might have multiple paths. Therefore, delay for such LSPs is not deterministic. Thus, Paragon Pathfinder cannot calculate and display the LSP delay.

Table 2: Data Collected for Analytics: Cisco Devices

Data

Collected by Default

Ingest Type

Rules Used to Collect Data

Label-switched Path (LSP) statistics [RSVP, segment routing]

Yes

SNMP

The Management Information Base (MIB) used is MPLS-TE-STD-MIB::mplsTunnelTable).

controller-telemetry /ctrl-label-switched-path-snmp-aggregation

Logical and physical interface statistics

Yes

  • SNMP (For devices running IOS XR Release 7.1.0 and prior)

    The MIB used is IF-MIB::ifXTable.

  • OpenConfig (For devices running IOS XR Release 7.1.1 and later)
  • SNMP:

    controller-telemetry /ctrl-logical-interface-snmp-aggregation

    controller-telemetry /ctrl-physical-interface-snmp-aggregation

  • OpenConfig:

    controller-telemetry /ctrl-logical-interface

    controller-telemetry /ctrl-logical-interface-aggregation

    controller-telemetry /ctrl-physical-interface

    controller-telemetry /ctrl-physical-interface-aggregation

Netflow Yes Not Applicable controller.telemetry/ctrl-demand-stats

controller.telemetry/ctrl-as-demand-stats

Table 3: Data Collected for Analytics: Nokia Devices

Data

Collected by Default

Ingest Type

Rules Used to Collect Data

Label-switched Path (LSP) statistics [RSVP, segment routing]

Yes

SNMP

The Management Information Base (MIB) used is MPLS-TE-STD-MIB::mplsTunnelTable).

controller-telemetry /ctrl-label-switched-path-snmp-nokia-aggregation

Logical and physical interface statistics

Yes

SNMP

The Management Information Base (MIB) used is IF-MIB::ifXTable.

SNMP:

controller-telemetry /ctrl-logical-interface-snmp-aggregation

controller-telemetry /ctrl-physical-interface-snmp-aggregation

Note:

The default SNMP collection interval is 300 seconds. You can modify the interval from the ctrl-snmp-frequency field in the Frequency profile tab on the Ingest Settings page (Configuration > Data Ingest > Settings). See Manage Frequency Profiles for details.

Workflow to Collect Device Statistics

Paragon Automation contains a default controller device group and a default controller playbook. Devices must be added to the controller device group so that analytics data can be collected from these devices. The controller playbook defines the rules for collecting the analytics data.

The following is the workflow to collect analytics data in Paragon Automation:

  1. If not added already, add the devices from which you want to collect data, to the Controller device group. See Edit a Device Group.

  2. Run the device collection task to correlate the links to the interfaces displayed in the topology view (Network > topology); see Add a Device Collection Task.

  3. For a Juniper Networks device, configure the devices to send JTI telemetry data and RPM statistics; see Configure Routers to Send JTI Telemetry Data and RPM Statistics to the Data Collector.

  4. For a Cisco IOS XR device, configure OpenConfig or SNMP for collecting interface statistics:

    1. Access the CLI of the Cisco IOS XR device.

    2. Do one of the following:

      • To configure OpenConfig on Cisco IOS XR devices, add the following configuration:

      • To configure SNMPv2 on Cisco IOS XR devices, add the following configuration:
      • To configure SNMPv3 on Cisco IOS XR devices, add the following configuration:
        Note:

        The following is a sample configuration.

      After you add the SNMPv3 configuration, configure the SNMPv3 parameters from the Devices page (Configuration > Devices). See Configure SNMP Ingest for details.

    3. Commit the configuration.

  5. Add optional rules to the controller playbook in Paragon Automation.

    The rules for collecting LSP statistics, interface statistics, SID statistics, and link latency are added to the controller playbook by default.

    If you want to collect QoS and LDP demand statistics, you must add the rules for collecting these statistics manually. To add rules to the controller playbook.

    1. Click the Configuration > Playbooks icon in the left navigation menu.

      The Playbooks page appears.

    2. Click the Controller playbook.

      The Edit Playbook Controller page appears.

    3. In the Rules field, click and select the following rules:

      • controller-telemetry/ctrl-egress-queue-interface and controller-telemetry/ctrl-ingress-queue-interface for collecting QoS data.

      • controller-telemetry/ctrl-ldp-demand-stats for collecting LDP demand statistics.

    4. Click Save and Deploy.

      A confirmation message appears that the playbook is successfully saved and deployed on the devices in the controller device group.

      For more details on editing a playbook, see Edit a Playbook

    5. Pause the current playbook instance assigned to the controller device group and add the new playbook instance. see Manage Playbook Instances for details.

    6. (Optional) Delete the paused playbook instance.

  6. Commit the configuration.

  7. Add the following data rollup summarization profiles to the controller device group, if you added rules for QoS data in step 5:

    ctrl-egress-queue-interface-rollup-profile and ctrl-ingress-queue-interface-rollup-profile

    controller-default-rollup-profile is added to the controller device group by default.

    For details about data summarization profiles, see Data Summarization Overview and Apply Data Summarization Profiles.

  8. Add devices to the collector device group to enable collecting the device data.

    Do one of the following to add a device to a device group:

    Note:

    The Junos telemetry interface (JTI) sensors use the system ID parameter of Juniper Networks devices to collect data from the devices. So, if not already configured in Paragon Automation, configure the system ID for the Juniper Networks devices. The system ID should be in the following format: <host_name>:<jti_ip_address> and should have the same value as configured on the device; see Edit Devices for configuring the system ID of a device in Paragon Automation.

Sample Configuration for collecting Telemetry from Cisco IOS XR Devices

The subscription name on a Cisco device must be derived from the sensor-name of the Paragon Insights rule. The subscription depends on the sensor-name you want to execute on the Cisco device. For example, if the sensor-name is /interfaces-test, the subscription name needs to be configured in IOS-XR as hbot_interfaces_test_.

When Paragon Automation sends the subscription request to the Cisco device:

  • A prefix hbot_ is added to the sensor-name, and

  • all the non-letter and non-digit characters are replaced with an underscore (_).

Note:

The conversion will take place only if the sensor-name is configured with a leading slash (/).

The following is sample configuration of the Paragon Insights rule for obtaining telemetry from Cisco IOS XR devices:

Since the sensor-name is /jtimon_sr_te, the Cisco IOS configuration will be:

Data Summarization and Retention Policy

Paragon Automation uses summarization (rollup) profiles to rollup data so that the data is stored in a smaller disk space and performance of the time-series database (TSDB) is improved. Raw network and device data are grouped together into hourly and daily data by using aggregate functions. For more information on data summarization, see Data Summarization Overview.

By default, the data is retained for the following duration:

  • Daily rollup data is stored for 1000 days.

  • Hourly rollup data is stored for 180 days.

  • Raw data is stored for 14 days.

You can modify the default duration for which the rollup data is retained from the Ingest Settings page Administration > Ingest Settings > Retention Policy.

The network archive task uses the rolled-up data for providing network statistics. Also, statistics are displayed on the topology graphs by fetching long range data from rolled-up measurements based on the rollup-query-cutoff-interval Paragon Pathfinder setting.

rollup-query-cutoff-interval defines the time period for which raw data or rolled-up data is fetched. If data is queried for a time period less than or equal to the value set in rollup-query-cutoff-interval, raw data is fetched, If data is queried for a time greater than rollup-query-cutoff-interval, hourly rolled-up data is fetched. The default value is set to 7 days; that is, by default, rolled-up data is fetched when queried for data greater than 7 days. For information about setting the rollup-query-cutoff-interval parameter, see Modify Pathfinder Settings From the GUI or Modify Pathfinder Settings From the Pathfinder CLI.