fix(backend): 部署韧性——模型自愈/就绪门禁/全局代理/启动诊断

- whisper: model.bin 截断/损坏时删目录重下重试一次,修「Unable to
  open file model.bin」死循环;mlx 同样按 config.json 判完整性
- /generate_note 加就绪门禁:本地转写引擎模型没下好直接拦截,返回
  reason=transcriber_model_not_ready,不让任务静默卡在首次下载
- 全局代理:新增 ProxyConfigManager(JSON 配置 + HTTP_PROXY env 兜底)
  + build_openai_client,统一注入代理到 LLM/Groq 客户端;yt-dlp 与
  youtube-transcript-api 也走代理
- build_openai_client 校验 api_key 非空,空 key 给「xxx 的 API Key
  未配置」而不是天书般的 Illegal header value b'Bearer '
- universal_gpt: 模型拒绝自定义 temperature(o1/o3/gpt-5 系列)时
  就地去掉参数重试,不消耗重试预算
- connect_test 改用真实 chat completion 而非 /v1/models 探测
- main.py: lifespan 拆 [startup 1/5..5/5] 分段日志 + 异常清晰定位
- /sys_health 重构为结构化返回 {backend,ffmpeg,db,whisper_model}

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
huangjianwu
2026-05-14 19:01:14 +08:00
co-authored by Claude Opus 4.7
parent 88d25f8cc1
commit 41f17592c2
16 changed files with 534 additions and 92 deletions
+2 -2
View File
@@ -1,6 +1,6 @@
from typing import List
from app.gpt.base import GPT
from openai import OpenAI
from app.utils.openai_client import build_openai_client
from app.gpt.prompt import BASE_PROMPT, AI_SUM, SCREENSHOT
from app.gpt.utils import fix_markdown
from app.models.gpt_model import GPTSource
@@ -15,7 +15,7 @@ class DeepSeekGPT(GPT):
self.base_url = getenv("DEEP_SEEK_API_BASE_URL")
self.model=getenv('DEEP_SEEK_MODEL')
print(self.model)
self.client = OpenAI(api_key=self.api_key, base_url=self.base_url)
self.client = build_openai_client(self.api_key, self.base_url, key_label="DeepSeek 的 API Key")
self.screenshot = False
def _format_time(self, seconds: float) -> str: