End-to-end regression validation for Elasticsearch plugins (Lance Vector and Cloud IAM). Use when validating changes to prevent regressions, after modifying plugin code, before committing or creating PR, when user asks to validate test or regression test the plugins, or any mention of lance-vector or cloud-iam validation. Provides automated test scripts covering 75 test cases across P0/P1/P2 priorities including kNN search, OSS integration, memory leaks, authentication, role mapping, and edge cases.
Run P0 critical tests:
./project_starter.sh -d && bash .claude/skills/es-plugins-validation/scripts/run_p0.sh
| Component | Value |
|---|---|
| ES Distribution | build/distribution/local/elasticsearch-9.2.4-SNAPSHOT |
| OSS Bucket | denny-test-lance |
| OSS Credentials | ~/.oss/credentials.json |
| Startup Script | ./project_starter.sh |
| Basic Auth | elastic / Summer11 |
| ID | Description | Method |
|---|
| R1 | Zero-score bug | All hits._score > 0 |
| R2 | Memory leaks | Heap growth < 30% |
| R3 | OSS env vars | Set before ES starts |
| R4 | Arrow types | pa.string() for _id |
run_p0.sh - Critical TestsRuns 21 P0 tests (blocking regressions). Use after any code change.
run_p1.sh - High Priority TestsRuns 38 P1 tests (major functionality). Run daily or per PR.
run_all.sh - Full RegressionRuns all 75 tests. Run weekly or pre-release.
test_memory.sh - Memory Leak DetectionMonitors heap during 1000 searches.
create_test_data.py - Generate Lance DatasetsCreates test data in OSS bucket.
| Category | Tests | P0 | P1 | P2 |
|---|---|---|---|---|
| Lance Vector (LV) | 32 | 10 | 14 | 8 |
| Cloud IAM (CI) | 26 | 6 | 14 | 6 |
| Integration (INT) | 7 | 3 | 3 | 1 |
| Configuration (CFG) | 5 | 1 | 2 | 2 |
| Error Handling (ERR) | 5 | 1 | 3 | 1 |
Total: 75 tests (21 P0, 38 P1, 16 P2)
This skill is generated from reg_validation_guide.md (OFFICIAL guide). To update:
reg_validation_guide.mdpython3 .claude/skills/es-plugins-validation/scripts/sync_from_plan.py