fix(ops): keep Ollama health checks on alert fast model
Some checks failed
CD Pipeline / tests (push) Successful in 52s
Code Review / ai-code-review (push) Successful in 9s
CD Pipeline / post-deploy-checks (push) Has been cancelled
CD Pipeline / build-and-deploy (push) Has been cancelled

This commit is contained in:
Your Name
2026-05-05 23:16:15 +08:00
parent 83daeb3f87
commit 1cc215ec30
3 changed files with 6 additions and 2 deletions

View File

@@ -3201,6 +3201,7 @@ bash scripts/ops/ollama-topology-check.sh
- `interactive` / `healthcheck` / `alert_fast` 保持 GCP-A 優先
- `code_review` / `rag` / `embedding` / `deep_rca` / `image_analysis` / `hermes` 改為 111 優先
- 111 不可用時才回 GCP-B避免 GCP-A/B 在告警 canary 期間被 7B/14B/32B 模型污染
- `OLLAMA_HEALTH_CHECK_MODEL` 改為 `gemma3:4b`,避免 health probe 自己把 `qwen2.5:7b-instruct` 載入 GCP-A
驗證: