fix(ollama): 恢復 GCP 優先路由(ADR-110 正式路由)
nginx proxy 架設完成後恢復原設計: GCP-A (110:11435 → 34.143.170.20:11434) → primary GCP-B (110:11436 → 34.21.145.224:11434) → secondary 111 (192.168.0.111:11434) → 兜底 OLLAMA_URL=http://192.168.0.110:11435 OLLAMA_SECONDARY_URL=http://192.168.0.110:11436 OLLAMA_FALLBACK_URL=http://192.168.0.111:11434 已用 kubectl set env 熱更新,不動 image tag。 兩台 GCP Ollama 均 200 OK(10 個模型各)。 Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
@@ -17,14 +17,13 @@ data:
|
||||
# 服務端點 (非機密)
|
||||
# 2026-04-16 ogt + Claude Sonnet 4.6: 改指向 111(GPU 機,RTX)
|
||||
# 188 = CPU-only Ollama,推理極慢(>60s);111 有 GPU,avg 10s
|
||||
# 2026-05-03 ogt: ADR-110 Ollama GCP 三層容災(GCP-A → GCP-B → Local HDD)
|
||||
# 2026-05-04 ogt: ADR-110 修正 — K8s pods → GCP-A/B:11434 = connection refused(外網路由不通)
|
||||
# K8s 可達:111(內網),不可達:GCP-A/B(外網 port 11434 被擋)
|
||||
# 修法:111 升為 primary;GCP-A/B 保留為 secondary/tertiary,待 nginx proxy 架設後恢復可用
|
||||
# 長期目標:在 110 架設 nginx proxy 轉發 GCP-A/B,ConfigMap 改指向 110:11435 / 110:11436
|
||||
OLLAMA_URL: "http://192.168.0.111:11434"
|
||||
OLLAMA_SECONDARY_URL: "http://192.168.0.110:11435"
|
||||
OLLAMA_FALLBACK_URL: "http://192.168.0.110:11436"
|
||||
# 2026-05-04 ogt: ADR-110 三層容災正式路由(nginx proxy 架設完成後恢復 GCP 優先)
|
||||
# GCP-A(via 110:11435) → GCP-B(via 110:11436) → 111 兜底
|
||||
# 110 nginx proxy 轉發:11435 → 34.143.170.20:11434, 11436 → 34.21.145.224:11434
|
||||
# K8s pods 不可直連 GCP:11434(NetworkPolicy 外網 egress 只開 443)
|
||||
OLLAMA_URL: "http://192.168.0.110:11435"
|
||||
OLLAMA_SECONDARY_URL: "http://192.168.0.110:11436"
|
||||
OLLAMA_FALLBACK_URL: "http://192.168.0.111:11434"
|
||||
OPENCLAW_URL: "http://192.168.0.188:8088"
|
||||
KALI_SCANNER_URL: "http://192.168.0.112:8080"
|
||||
SIGNOZ_URL: "http://192.168.0.188:3301"
|
||||
|
||||
@@ -68,11 +68,11 @@ spec:
|
||||
- name: TELEGRAM_ENABLE_POLLING
|
||||
value: "true"
|
||||
- name: OLLAMA_URL
|
||||
value: "http://192.168.0.111:11434" # 2026-05-04 ogt: 111 primary(K8s 內網直連,GPU RTX)
|
||||
value: "http://192.168.0.110:11435" # 2026-05-04 ogt: GCP-A primary via 110 nginx proxy(11435 → 34.143.170.20:11434)
|
||||
- name: OLLAMA_SECONDARY_URL
|
||||
value: "http://192.168.0.110:11435" # 2026-05-04 ogt: GCP-A via 110 nginx proxy(11435 → 34.143.170.20:11434)
|
||||
value: "http://192.168.0.110:11436" # 2026-05-04 ogt: GCP-B secondary via 110 nginx proxy(11436 → 34.21.145.224:11434)
|
||||
- name: OLLAMA_FALLBACK_URL
|
||||
value: "http://192.168.0.110:11436" # 2026-05-04 ogt: GCP-B via 110 nginx proxy(11436 → 34.21.145.224:11434)
|
||||
value: "http://192.168.0.111:11434" # 2026-05-04 ogt: 111 兜底(K8s 內網直連,GPU RTX)
|
||||
- name: OPENCLAW_DEFAULT_MODEL
|
||||
value: "qwen2.5:7b-instruct"
|
||||
- name: OPENCLAW_TIMEOUT
|
||||
|
||||
Reference in New Issue
Block a user