Let ThreadPool exit gracefully. (#3653)

### What problem does this PR solve?

#3646

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
This commit is contained in:
Kevin Hu 2024-11-26 16:31:07 +08:00 committed by GitHub
parent 5c59651bda
commit 0891a393d7
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
5 changed files with 37 additions and 39 deletions

View File

@ -65,7 +65,3 @@ class Crawler(ComponentBase, ABC):
elif self._param.extract_type == 'content':
result.extracted_content
return result.markdown

View File

@ -64,9 +64,9 @@ def build_knowledge_graph_chunks(tenant_id: str, chunks: list[str], callback, en
BATCH_SIZE=4
texts, graphs = [], []
cnt = 0
threads = []
max_workers = int(os.environ.get('GRAPH_EXTRACTOR_MAX_WORKERS', 50))
exe = ThreadPoolExecutor(max_workers=max_workers)
with ThreadPoolExecutor(max_workers=max_workers) as exe:
threads = []
for i in range(len(chunks)):
tkn_cnt = num_tokens_from_string(chunks[i])
if cnt+tkn_cnt >= left_token_count and texts:

View File

@ -88,12 +88,12 @@ class MindMapExtractor:
prompt_variables = {}
try:
res = []
max_workers = int(os.environ.get('MINDMAP_EXTRACTOR_MAX_WORKERS', 12))
exe = ThreadPoolExecutor(max_workers=max_workers)
with ThreadPoolExecutor(max_workers=max_workers) as exe:
threads = []
token_count = max(self._llm.max_length * 0.8, self._llm.max_length - 512)
texts = []
res = []
cnt = 0
for i in range(len(sections)):
section_cnt = num_tokens_from_string(sections[i])

View File

@ -366,7 +366,7 @@ class OllamaChat(Base):
keep_alive=-1
)
ans = response["message"]["content"].strip()
return ans, response["eval_count"] + response.get("prompt_eval_count", 0)
return ans, response.get("eval_count", 0) + response.get("prompt_eval_count", 0)
except Exception as e:
return "**ERROR**: " + str(e), 0

View File

@ -492,6 +492,7 @@ def report_status():
logging.exception("report_status got exception")
time.sleep(30)
def analyze_heap(snapshot1: tracemalloc.Snapshot, snapshot2: tracemalloc.Snapshot, snapshot_id: int, dump_full: bool):
msg = ""
if dump_full:
@ -508,6 +509,7 @@ def analyze_heap(snapshot1: tracemalloc.Snapshot, snapshot2: tracemalloc.Snapsho
msg += '\n'.join(stat.traceback.format())
logging.info(msg)
def main():
settings.init_settings()
background_thread = threading.Thread(target=report_status)