estel/dify
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

fix infinite loop when clean unused dataset (#24542)

Browse Source 
Signed-off-by: kenwoodjw <blackxin55+@gmail.com>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
This commit is contained in:
kenwoodjw
2025-08-26 15:44:52 +08:00
committed by GitHub
parent 98473e9d4f
commit bfc4fe1a9a
1 changed files with 5 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
api/schedule/clean_unused_datasets_task.py
View File

@@ -45,6 +45,7 @@ def clean_unused_datasets_task():
plan_filter = config["plan_filter"] plan_filter = config["plan_filter"]
add_logs = config["add_logs"] add_logs = config["add_logs"]
page = 1
while True: while True:
try: try:
# Subquery for counting new documents # Subquery for counting new documents
@@ -86,12 +87,12 @@ def clean_unused_datasets_task():
.order_by(Dataset.created_at.desc()) .order_by(Dataset.created_at.desc())
) )
datasets = db.paginate(stmt, page=1, per_page=50) datasets = db.paginate(stmt, page=page, per_page=50, error_out=False)
except SQLAlchemyError: except SQLAlchemyError:
raise raise
if datasets.items is None or len(datasets.items) == 0: if datasets is None or datasets.items is None or len(datasets.items) == 0:
break break
for dataset in datasets: for dataset in datasets:
@@ -150,5 +151,7 @@ def clean_unused_datasets_task():
except Exception as e: except Exception as e:
click.echo(click.style(f"clean dataset index error: {e.__class__.__name__} {str(e)}", fg="red")) click.echo(click.style(f"clean dataset index error: {e.__class__.__name__} {str(e)}", fg="red"))
page += 1
end_at = time.perf_counter() end_at = time.perf_counter()
click.echo(click.style(f"Cleaned unused dataset from db success latency: {end_at - start_at}", fg="green")) click.echo(click.style(f"Cleaned unused dataset from db success latency: {end_at - start_at}", fg="green"))