Compliance Tool for Accelerator Management

The OCP Test & Validation Initiative is a collaboration between datacenter hyperscalers having the goal of standardizing aspects of the hardware validation/diagnosis space, along with providing necessary tooling to enable both diagnostic developers and executors to leverage these interfaces.

Specifically, the ocp-diag-ctam tool provides acceptance testing for Accelerator Management in cloud data centers.

---

This project is part of OCPTV and exists under the same MIT License Agreement.

Getting Started

Prerequisites

Before you begin, ensure you have met the following requirements:

- Python 3.9 or higher is installed
- Python virtualenv and pip is installed
- Install some key libraries:
     sudo apt-get install python3-tk sshpass jq
- Docker is installed (Skip unless you want to create a binary)    

Setup

1. Clone the repo,

   https://github.com/opencomputeproject/ocp-diag-ctam

Setting up the workspace

1. Sample workspace files are present inside json_spec directory. Modify these file as per your infra details.

2. input dir inside json_spec contains sample input file that ctam require to run

3. Create a workspace directory and copy the configuration files from json_spec/input directory into workspace dir.

   • .netrc - contains bmc ipaddress, username and password
   • dut_config.json - contains various params for running the test cases
   • package_info.json - contains details about the firmware
   • redfish_uri_config.json - redfish config file
   • test_runner.json - config file for test run, can be overridden by cli flags
   • redfish_response_messages.json - config file for redfish response messages (default will be picked from json_spec/input).

Flag description

CLI Argument	Type	Definition
-t or --testcase	string	Runs a single test case. Overrides test_runner.json in the workspace
-test_seq or --testcase_sequence	string	Runs a single test case. Overrides test_runner.json in the workspace
-group_seq or --group_sequence	string	Runs no of groups with given sequence
-s or --Suite		Run full ACT Suite
-g or --group		Run tests for a single group. Overrides test_runner.json in the workspace
-d or --Discovery		Path to workspace directory that contains test run files
-l or --list	string	List all test cases. If combined with -G then list all cases of the chosen group
-v or --version		Lists the current version

Running the tool locally

1. Optional: create python virtual environment and activate.

   python -m venv venv
   source ./venv/bin/activate
   

2. Install dependencies

   python -m pip install -r pip-requirements.txt
   

3. For full documentation, from /docs directory, run

   ./make html
   

   Open docs/build/html/index.html for full documentation including architecture and test case details

4. To run suite,

   cd ctam
   python ctam.py -w ..\example_workspace
   

   Logs will be created under example_workspace\TestRuns

5. To list all test cases

   cd ctam
   python ctam.py -l
   

6. To run a specific test case

   cd ctam
   python ctam.py -w ..\example_workspace -t <test case id>
   

   Logs will be created under example_workspace\TestRuns

7. To run test cases of a specifc test group

   cd ctam
   python ctam.py -w ..\example_workspace -g <test group name>
   

   Logs will be created under example_workspace\TestRuns

8. To run test cases with sequence

   cd ctam
   python ctam.py -w ..\example_workspace -test_seq <test case name or id> <test case name or id>
   

   Logs will be created under example_workspace\TestRuns

9. To run groups with sequence

   cd ctam
   python ctam.py -w ..\example_workspace -group_seq <group name or id> <group name or id>
   

   Logs will be created under example_workspace\TestRuns

10. Choose test cases to run by using tags and specifying the tags to include/exclude in test_runner.json

11. Choose test sequence in test_runner.json if you want to run it from test runner config.

Binary

1. One file Binary executable can be created using Makefile, to create binary run following command. This will create binary and sample workspace dir inside the dist folder.

   make build_image

2. You can run the binary the same way running the python file. Just that now python file replaced by binary executalbe. Sample command to list all test cases.

   Note: Please move your workspace directory inside dist directory before running the binary.

   cd dist && ./ctam -l

3. To clear build files, run:

   make clean

Log Files created

1. OCPTV Log file - All logs in OCPTV defined logging format.
2. Test_Score_<>.json - All test cases result + Final score.
3. Test_Report_<>.log - Tabulated report of test run
4. Test_Info_<>.json - Optional log file used by test interfaces (for debug)
5. RedfishCommandDetails/RedfishCommandDetails_<Test_ID>_ <Test_Name>_<>.json - Redfish Commands used & return values (for debug)

Test Runner Knobs

Test runner knobs can be modified in test_runner.json to enable different logging mode.

Variable	Type	Definition
debug_mode	boolean	For debug logs
console_mode	boolean	For console logs
progress_bar	boolean	For for progress bar indicator

Tags

1. Group Tag

• We can give tags at group level also.
• If we provide any tag to a particular group, then all the test case under that group will be considered as the same tag.
• If we run according to group tag, then all the test case will run under the group irrespective of the test case tag.

2. Test Case Tags

• We can assign different tags to different test cases.
• If we run according to test case tag, then all the test cases which assigned with that tag would run irrespective of group tags.

Note: - Tags = Group Tags Union Test Case Tags group tags = ["G1"] and test case tags = ["L1"], so the final tags will be ["G1", "L1"]

Developer notes

VS Code

VS Code is not required for development, however it does have workspace configurations to assist in development.

To use VS Code, open the ctam.code-workspace file.

In lieu of VS Code usage, the following items should be configured for other editors or the developer should perform steps manually to ensure the consistency of the code base.

• automatic file formatting using python black formatter on file save.
  • "--line-length", "120"
• Indent set to 4 spaces
• Auto docstring configured for sphinx, type 3 double quotes below python class or function and the documentation header is automatically stubbed out.
• Automatic mypy checking of code
• Spell checking
• Useful git extensions
• Useful debugger configurations defined in launch.json, extensible
• Useful Code snippets to jump start new test development
  • Snippets for interfaces, test groups and test cases simplify new tests.
  • easy replace of TODO in the snippet creates runnable test case quickly.

Upcoming changes

• More test cases
• Logging improvements
• Ability to set test sequence
• PLDM validator, and auto creation of PLDM bundles with error injection.

Contact

Feel free to start a new discussion, or otherwise post an issue/request.

An email contact is also available at: [email protected]