1. Generating customized log events with EHS#


Activity name	Generating customized log events with EHS
Activity ID	09
Short Description	Custom Events can be generated in pySROS script for use case specific conditions and as a trigger for Event Handling System
Difficulty	Beginner
Tools used	pySROS, EHS Module, SR OS YANG Models, SR OS YANG path finder
Topology Nodes	PE2
References	Functions for the event handling system (EHS) Model-driven SR OS interface SR OS Documentation

The ability to raise a custom event via the perform log custom-event command is available in model-driven interfaces. Log events generated in this way pass through the regular event processing pipeline and are visible via all usual mechanisms, including show commands and configured log destinations.

There are 6 distinct events that exist within model-driven SR OS that can be used to send a custom-event out into the world. Each event corresponds to a severity level, ranging from indeterminate for tmnxCustomEvent6 to critical for tmnxCustomEvent1. The other intermediate levels are major for tmnxCustomEvent2, minor for tmnxCustomEvent3, warning for tmnxCustomEvent4 and cleared for tmnxCustomEvent5.

A:admin@g15-pe2# show log event-control "logger"
=======================================================================
Log Events
=======================================================================
Application
 ID#    Event Name                       P   g/s     Logged     Dropped
-----------------------------------------------------------------------
  ...
   2020 tmnxCustomEvent1                 CR  thr          0           0
   2021 tmnxCustomEvent2                 MA  thr          2           0
   2022 tmnxCustomEvent3                 MI  thr          0           0
   2023 tmnxCustomEvent4                 WA  thr          3           0
   2024 tmnxCustomEvent5                 CL  thr          0           0
   2025 tmnxCustomEvent6                 IN  thr          0           0

Custom events can be used as triggers for event handlers, with all parameters passed into the associated EHS scripts. The customizable log events will be output to any configured log destinations (for example, syslog or SNMP traps).

1.1 Objective#

Imagine you’re operating a customer-facing service that suddenly experiences an outage. Your network monitoring system (NMS) detects the issue and triggers an alert.

In this activity, we simulate that automated monitoring system trigger, by using the perform log custom-event command. This command manually raises a custom event, complete with parameters that would typically be provided by the monitoring system (e.g., service ID, interface name, customer subnet). The custom event is taken by the Event Handling System (EHS) and used to call a Python application. In the scenario pictured above, this Python application may instead be called by the NMS or by an operator. We aim to show that as long as the events are available in the system, SR OS can use (custom) events as triggers for pre-programmed responses with the on-box interpreter.

For the activity, once the custom event is triggered, EHS catches it and executes a python-script to automatically instantiate debugging configuration.

This not only mirrors how service providers can automate failure recovery or root cause analysis, but it also shows how the EHS can be leveraged to support intelligent, responsive network behavior through programmable interfaces.

Through the tasks we will see how to pass essential parameters for creating a VPRN service to the EHS by raising a Custom Log Event, and how to use those parameters to configure the service.

classDiagram
  direction LR
  class Custom-Event {
    - Subject
    - MessageString
    - Parameter1()
    - Parameter2()
    - ...Parameter8()

  }
  class EHS-pySROS-script {
    - trigger_event.eventid
    - configure-vprn(Parameters)
    - successful(raise custom-event)
    - failed(raise custom-event)
  }

Custom-Event --> EHS-pySROS-script : triggers
Custom-Event --> EHS-pySROS-script : pass parameters
EHS-pySROS-script --> Custom-Event : can trigger

1.2 Event Handling Script Functionality#

The Event Handling System (EHS) is a framework that enables operators to configure custom behavior on the router.

The EHS introduces user-defined, programmatic exception handling by executing either a CLI script or a Python application in response to a detected log event, referred to as the "trigger".

Python applications are fully supported and leverage the model-driven interfaces available in SR OS along with the pySROS libraries to access and modify both state and configuration data. These applications can also invoke YANG-modeled operations via pySROS API calls, and, if necessary, execute unstructured MD-CLI commands as a fallback.

During the development of an EHS Python application, event attributes are made available through the get_event function in the pysros.ehs module.

pysros.ehs - Functions for the event handling system (EHS)

The pysros.ehs module is available when executing on SR OS only. On a remote machine, the event handling system (EHS) and its functionality are not supported.

1.3 Tasks#

You should read these tasks from top-to-bottom before beginning the activity.

It is tempting to skip ahead but tasks may require you to have completed previous tasks before tackling them.

In the upcoming tasks, you will set up logic that will react to an event and drive an automated response. You will then raise a custom event log containing specific parameters, simulating a monitoring system's function in generating a tailored alert except now there is no need for any additional elements to call your Python application. The raised event itself will act as the trigger for your Python application, which will use the supplied parameters to configure a VPRN service for debugging purposes.

1.3.1 EHS Configuration#

In this task we will ensure the event handling system is ready to respond to an event. The infrastructure of the EHS is already configured on PE2 so that we can focus on the main goal of this activity; in the upcoming tasks you can develop your pySROS code on your group's hackathon VM, under the location

/home/nokia/SReXperts/activities/nos/sros/activity-09/ehs_basic.py

This same ehs_basic.py file has been made available within PE2 via container binds.

Note

In the Hackathon environment the /home/nokia/clab-srexperts/pe2/tftpboot/ directory on your groups Hackathon VM is made available using TFTP to the node PE2. Any file you put in that folder will be accessible from within SR OS using TFTP. This could be used as an alternative to scp or manually copying over file contents for containerlab environnments.

Any modifications you make to that file in your group's hackathon VM will be visible to the router, this is the preferred method of development as it allows you to modify the file under the Linux filesystem while being able to test on the containerlab node.

Connect to PE2 from your group's hackathon VM

ssh admin@clab-srexperts-pe2

In the box below you see that the full configuration for setting up the EHS on the node has already been prepared for you. However, the second step (creating a directory to store the python-script execution results) must be performed manually by you before continuing with the tasks.

EHS Configuration

1- python-script location

2-make-directory for ehs execution results

5-event-handler links the script-policy to the ehs that can be


Do you feel you have achieved something?	Yes No
Was the difficulty level graded appropriately?	Yes No
How do you rate this activity?

1. Generating customized log events with EHS#

1.1 Objective#

1.2 Event Handling Script Functionality#

1.3 Tasks#

1.3.1 EHS Configuration#

1.3.2 EHS Output#

1.3.3 Custom event log parameters#

1.3.4 Use the Parameters of a Custom Event Log for Configuration#

1.4 Summary#