mirror of
https://github.com/meta-llama/llama-stack.git
synced 2025-12-27 09:01:59 +00:00
Add support for server:<config> format in --stack-config option to enable seamless one-step integration testing. This eliminates the need to manually start servers in separate terminals before running tests. Features: - Auto-start llama stack server if target port is available - Reuse existing server if port is already in use - Health check polling with 2-minute timeout - Custom port support via server:<config>:<port> - Clean test output with background server execution - Backward compatibility with all existing formats Examples: pytest tests/integration/inference/ --stack-config=server:fireworks pytest tests/integration/safety/ --stack-config=server:together:8322 Test Plan: - Verified server auto-start with available ports - Verified server reuse with occupied ports - Verified health check polling via /v1/health endpoint - Tested custom port configuration - Confirmed backward compatibility with existing config formats |
||
|---|---|---|
| .. | ||
| client-sdk/post_training | ||
| common | ||
| external-provider/llama-stack-provider-ollama | ||
| integration | ||
| unit | ||
| verifications | ||
| __init__.py | ||
| Containerfile | ||
| README.md | ||
Llama Stack Tests
Llama Stack has multiple layers of testing done to ensure continuous functionality and prevent regressions to the codebase.
| Testing Type | Details |
|---|---|
| Unit | unit/README.md |
| Integration | integration/README.md |
| Verification | verifications/README.md |