Scribd
Upload
4 views2 pages

GPT

The document outlines test cases and keywords for validating the creation of KF Serving jobs for models using the MLPaas SCM framework. It includes steps for verifying job existence, checking prediction results storage in a PostgreSQL database, and confirming carrier states based on forecast values. The tests are designed to ensure proper functionality and data integrity within the system.

Uploaded by

ganeshlakshman00
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
4 views2 pages

GPT

The document outlines test cases and keywords for validating the creation of KF Serving jobs for models using the MLPaas SCM framework. It includes steps for verifying job existence, checking prediction results storage in a PostgreSQL database, and confirming carrier states based on forecast values. The tests are designed to ensure proper functionality and data integrity within the system.

Uploaded by

ganeshlakshman00
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
You are on page 1/ 2

1>

2>

3>

4>

5>

6>

7>

8>

9>

10>

""""""""1. Validate KF Serving job is getting created for each model using MLPaas
SCM framework""""""""

*** Test Cases ***


Validate_KF_Serving_Job_Creation
[Documentation] Validate KF Serving job is getting created for each model
using MLPaas SCM framework
[Tags] ModelServing
[Timeout] 20 minutes
Open Connection ${MLPaas_Server}
Login to Host ${MLPaas_Hostname} ${MLPaas_Alias} ${MLPaas_Prompt} $
{MLPaas_Username} ${MLPaas_Password}
${model_list} Get List of Models # Replace with actual keyword to get the
list of models
FOR ${model} IN @{model_list}
Validate_KF_Job_Exists ${model}
END

*** Keywords ***


Validate_KF_Job_Exists
[Arguments] ${model}
${job_name} Set Variable ${model}_KF_Job
Run Keyword And Expect Error Execute Command kubectl get deployment $
{job_name} -n kfserving-system return_stdout=True
Should Not Be Equal As Integers ${ERROR} 0

if service control is sending request, then it is pressumed thatmodels are craeted

verify_creation_of_isvc_with_svcctrl
[Documentation] Validate the functionality of ondemand training
[Timeout] 5 minutes
${full_name} ${std_err} ${std_rc}= Execute Command kubectl get po -
n ricxapp | grep "ricxapp-mimomuting-inference" | awk '{print $1}' | head -n1
return_stdout=True return_stderr=True return_rc=True
Log "Logs for service name: "${full_name}
${log_out} ${std_err} ${std_rc}= Execute Command kubectl logs $
{full_name} -n ricxapp --all-containers return_stdout=True return_stderr=True
return_rc=True
Log ${log_out}
Should Contain ${log_out} /servicectrl/inference/create

grep "ricxapp-mimomuting-inference" this has tobe changed, also try keeping it like
argument and give it from Validate_KF_Serving_Job_Creation

*** Test Cases ***


Verify_Prediction_Results_Storage
[Documentation] Verify prediction results are stored in mimo_load_forecast
table in postgres DB
[Tags] DataStorage
[Timeout] 15 minutes
Open Database Connection ${DB_Hostname} ${DB_Port} ${DB_Name} $
{DB_User} ${DB_Password}
${query_result} Query SELECT * FROM mimo_load_forecast WHERE ... #
Replace with your query
Should Not Be Empty ${query_result}

*** Test Cases ***


Verify_Carrier_State_Based_on_Forecast
[Documentation] Verify the carrier is ON/OFF based on the forecast values
and load profile configured in mimo_loadprofile_configuration defined in the
postgres DB
[Tags] CarrierState
[Timeout] 20 minutes
Open Database Connection ${DB_Hostname} ${DB_Port} ${DB_Name} $
{DB_User} ${DB_Password}
${forecast_data} Query SELECT forecast_value FROM mimo_load_forecast
WHERE ... # Replace with your query
${load_profile} Query SELECT load_profile FROM
mimo_loadprofile_configuration WHERE ... # Replace with your query
Validate_Carrier_State ${forecast_data} ${load_profile}

*** Keywords ***


Validate_Carrier_State
[Arguments] ${forecast_data} ${load_profile}
# Your logic to compare forecast and load profile data to determine carrier
state
# Implement your checks here and fail the test case if the carrier state is
incorrect
# For example:
Should Be True ${carrier_state} Carrier state is not as expected

You might also like