As the support team for CX as Code, we are sometimes pulled in to examine a customer's environment or one of our own internal environments to understand what is happening in their Terraform/CX as Code environment.
As a result of these requests, we have written several Python and Jupyter Notebook scripts to process and analyze the data. We are releasing these into our lab environment "AS-IS" with no guarantee of support. Our goal is to share these scripts so that the large CX as Code community can use them as they see fit.
All requirements were captured in a Python requirements.txt file. The libraries can be
installed using pip install -r requirements.txt.
All paths used in these notebooks are read from commonlib/config.Config class. This class will read and write based on three environment variables:
export TERRAFORM_LOG_PATH="" #Location of the log file
export NORMALIZED_TERRAFORM_LOG_PATH="" #Output path fo the normalized Terraform log data
export NORMALIZED_GENESYS_SDK_PATH="" #Output path ot the normalized SDK data
The sdk-plan-notebooks directory contains two files: plan-analysis.ipynb and sdk-notebook.ipynb.
The plan-analysis.ipynb file is a Jupyter notebook that will take the STDOUT from the running of terraform plan or tofu plan and parse the results to identify long plan resolution times, drift detection, etc...
The sdk-notebook.ipynb' will parse the output from a STDOUT run of a TF_DEBUG=json terraform apply --auto-approve` and will break down the API calls being made in the plan.
The notebooks/common-lib contains Python functions used to carry out various functions in processing.
The generator folder is a small Python script used to generate a large number of resources. This was so we could use it to create enough to parse and log the output.