Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?

 
 

Configure Paragon Insights Notification for LSP Gray Failures

Paragon Insights can notify Paragon Pathfinder if there are gray failures in Label Switched Paths (LSPs) or nodes connecting the LSPs in an MPLS network. Gray failures or short-lived failures are mainly caused by interface-related issues such as link flaps, traffic exceeding bandwidth thresholds, link or interface errors, network delay, or packet loss and node-related issues such as routing daemon CPU utilization exceeding threshold, routing protocol issues, memory leaks, chassis temperature rising beyond a set threshold, or line card failures. When an LSP consists of a faulty link due to a gray failure, Paragon Insights can be configured to notify Paragon Pathfinder which marks the link or node for maintenance and removes the maintenance mode status when the cause of gray failure is resolved.

Before you begin to configure functions, you must have completed configuring the interface.statistics/check-interface-flaps rule. The interface.statistics/check-interface-flaps rule periodically collects link flap counts and notifies Paragon Pathfinder when the link flap counts exceeds a pre-defined threshold.

The following checklist ensures that you have configured the interface.statistics/check-interface-flaps rule in Rules page:

  • Selected sensor and configured sensor frequency in interface.statistics/check-interface-flaps rule in Sensor tab.

  • Added a field to take flap count from sensor in Fields tab.

  • Added a field to set flap threshold in Fields tab.

  • Added a field to capture interface name in Fields tab.

  • Added a Term to check if link flaps are increasing compared to threshold value for a given time range in Triggers tab.

  • Added a Term to check if link is stable in Triggers tab.

Paragon Insights has a number of pre-defined rules that collect interface statistics such as flap count, administrative state of the interface, neighbor interface states, and so on. Let us take the example of how you can configure a function in Paragon Insights that sends Paragon Pathfinder a maintenance mode alert when flap count increases consistently.

After you configure the interface.statistics/check-interface-flaps rule, you can configure functions to generate or remove link maintenance mode in the Functions tab of the rule. The Python function catalog notify_pathfinder.py has pre-defined functions for generating and removing maintenance modes. You can make a reference to these functions and set arguments (parameters).

To add a link maintenance function:

  1. Go to Configuration > Rules in the left navigation bar.

    The Rules page appears.

  2. Search for interface in the search box of Rules page.

    This brings up all pre-defined rules related to interfaces. Check for interface.statistics/check-interface-flaps and click on the rule.

  3. Click the Functions tab.
  4. Click the Add Function button.
  5. Enter a name for the function such as generate_maintenance.
  6. Enter the description of the function in the Description field.
  7. In the Path to Function field, enter notify_pathfinder.py.
  8. In the Method Name field, enter generate_link_maintenance.

    You can click the script icon near the field to open the catalog of functions available in notify_pathfinder.py. This shows a set of python functions with arguments (parameters). You must define the arguments mentioned in the function you choose.

  9. Click the Add Arguments button to enter all the arguments specified in the generate_link_maintenance function that is listed in the notify_pathfinder.py catalog.

    If you enable the Mandatory button, you must enter a value for the argument when you call the function.

  10. Click Save and Deploy.

In the Triggers tab, you can call the function name you added and enter the value for the arguments (function parameters).

To call the generate maintenance mode function for link flaps in a trigger:

  1. Click Triggers tab.

  2. In the link_flaps trigger, click on the Term you configured to check if link flaps exceed the threshold limit in a given time.

  3. Under the Then statement of the Term, click the Add Function button.

  4. Select the function name you configured to generate maintenance link from the drop-down menu.

    All the arguments (function parameters) you entered in the Functions tab appear here.

  5. Enter $ followed by argument name in the value field for each argument. You can also select a suitable input from the drop-down menu. For example, $interface-name for the argument ifd_name, $mnt_time for argument mnt_time, and so on.

  6. Click Save and Deploy.

You can follow the same instructions to add remove_link_maintenance function in Functions tab and define argument values for that function under the Term in Triggers tab that checks if an interface is stable.

The four main functions that enable Insights to trigger notifications are generate_link_maintenance and remove_link_maintenance for any interface related issues and generate_node_maintenance and remove_node_maintenance for any node related issues.

To check nodes or interfaces that are under maintenance, navigate to Network > Topology in the left navigation bar. Click ellipse (...) in the panel with node, link and tunnel tabs. Select Maintenance in the drop-down menu. The Maintenances page displays details of links and nodes kept under maintenance.

Related Documentation