woongmin commited on
Commit
62bd0bb
·
1 Parent(s): b34afb9

[Delta-02] Drill C retry/backoff 또는 graceful fallback 동작 확인용 코드 제거

Browse files
app/embedding_utils.py CHANGED
@@ -3,19 +3,10 @@ import nltk
3
  import numpy as np
4
  from openai import OpenAI, RateLimitError
5
  from sentence_transformers import CrossEncoder
6
- from tenacity import (
7
- retry,
8
- wait_exponential,
9
- stop_after_attempt,
10
- retry_if_exception,
11
- before_sleep_log,
12
- )
13
- import logging
14
  from app.config import OPENAI_API_KEY, OPENAI_EMBEDDING_MODEL
15
 
16
- logger = logging.getLogger(__name__)
17
-
18
- client = OpenAI(api_key=OPENAI_API_KEY, timeout=1.0)
19
 
20
  # Cross-encoder 리랭커 (첫 호출 시 모델 다운로드)
21
  reranker = CrossEncoder("BAAI/bge-reranker-v2-m3")
@@ -61,7 +52,6 @@ def _is_rate_limit_retryable(e: Exception) -> bool:
61
  retry=retry_if_exception(_is_rate_limit_retryable),
62
  wait=wait_exponential(multiplier=1, min=2, max=30),
63
  stop=stop_after_attempt(2),
64
- before_sleep=before_sleep_log(logger, logging.WARNING),
65
  )
66
  def _openai_embed(text: str) -> List[float]:
67
  response = client.embeddings.create(
 
3
  import numpy as np
4
  from openai import OpenAI, RateLimitError
5
  from sentence_transformers import CrossEncoder
6
+ from tenacity import retry, wait_exponential, stop_after_attempt, retry_if_exception
 
 
 
 
 
 
 
7
  from app.config import OPENAI_API_KEY, OPENAI_EMBEDDING_MODEL
8
 
9
+ client = OpenAI(api_key=OPENAI_API_KEY)
 
 
10
 
11
  # Cross-encoder 리랭커 (첫 호출 시 모델 다운로드)
12
  reranker = CrossEncoder("BAAI/bge-reranker-v2-m3")
 
52
  retry=retry_if_exception(_is_rate_limit_retryable),
53
  wait=wait_exponential(multiplier=1, min=2, max=30),
54
  stop=stop_after_attempt(2),
 
55
  )
56
  def _openai_embed(text: str) -> List[float]:
57
  response = client.embeddings.create(
app/main.py CHANGED
@@ -26,23 +26,14 @@ from .config import (
26
 
27
  import pdfplumber
28
  from openai import OpenAI, RateLimitError, AuthenticationError
29
- from tenacity import (
30
- retry,
31
- wait_exponential,
32
- stop_after_attempt,
33
- retry_if_exception,
34
- before_sleep_log,
35
- )
36
- import logging
37
  import hashlib
38
  import time
39
  from concurrent.futures import ThreadPoolExecutor
40
 
41
- logger = logging.getLogger(__name__)
42
-
43
  app = FastAPI(title="DocuFlow AI Agent")
44
 
45
- client = OpenAI(api_key=OPENAI_API_KEY, timeout=1.0)
46
 
47
 
48
  @app.get("/health")
@@ -62,7 +53,6 @@ def _is_rate_limit_retryable(e: Exception) -> bool:
62
  retry=retry_if_exception(_is_rate_limit_retryable),
63
  wait=wait_exponential(multiplier=1, min=2, max=30),
64
  stop=stop_after_attempt(2),
65
- before_sleep=before_sleep_log(logger, logging.WARNING),
66
  )
67
  def _call_llm(messages: list, model: str) -> str:
68
  response = client.chat.completions.create(model=model, messages=messages)
 
26
 
27
  import pdfplumber
28
  from openai import OpenAI, RateLimitError, AuthenticationError
29
+ from tenacity import retry, wait_exponential, stop_after_attempt, retry_if_exception
 
 
 
 
 
 
 
30
  import hashlib
31
  import time
32
  from concurrent.futures import ThreadPoolExecutor
33
 
 
 
34
  app = FastAPI(title="DocuFlow AI Agent")
35
 
36
+ client = OpenAI(api_key=OPENAI_API_KEY)
37
 
38
 
39
  @app.get("/health")
 
53
  retry=retry_if_exception(_is_rate_limit_retryable),
54
  wait=wait_exponential(multiplier=1, min=2, max=30),
55
  stop=stop_after_attempt(2),
 
56
  )
57
  def _call_llm(messages: list, model: str) -> str:
58
  response = client.chat.completions.create(model=model, messages=messages)
docs/04-weekend2/deployment-checklist-lab.md CHANGED
@@ -119,9 +119,17 @@
119
  {"detail":"Request timed out."}
120
  - 복구 시간(분): 3분
121
 
122
- - [ ] retry/backoff 또는 graceful fallback 동작 확인
123
  - 관찰 내용:
124
- - 복구 시간(분):
 
 
 
 
 
 
 
 
125
 
126
  ---
127
 
 
119
  {"detail":"Request timed out."}
120
  - 복구 시간(분): 3분
121
 
122
+ - [x] retry/backoff 또는 graceful fallback 동작 확인
123
  - 관찰 내용:
124
+ - 변경한 것들 - 1) timeout=1.0 — OpenAI 클라이언트에 1초 타임아웃을 설정해서 거의 확실히 타임아웃이 발생하게 했습니다. - 2) before_sleep_log — tenacity가 재시도하기 직전에 서버 로그에 이런 메시지가 남습니다:
125
+ `WARNING - Retrying _call_llm in 2.0 seconds as it raised <예외>.` - 3) import logging + logger — 위 로그를 출력하기 위한 설정.
126
+
127
+ - 결과:
128
+ - API Timeout Error 발생하면서 재시도 없이 바로 실패함.
129
+ - rate limit 또는 timeout 재현: timeout=1.0으로 강제 타임아웃 재현 성공
130
+ - retry/backoff 동작 확인: 타임아웃은 retry 조건에 포함되지 않아 즉시 실패. rate limit만 재시도 대상임을 확인
131
+ - 사용자에게 "Request timed out." 메시지가 500으로 그대로 노출됨
132
+ - 복구 시간(분): 4분
133
 
134
  ---
135