engineering/upload.unturf.com
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Optimize scheduler for minimal SQLite database locks

Browse source 
- Redesigned scheduler to use separate short-lived sessions for lock operations
- Lock acquisition and release now use independent database connections
- Each scheduled action processed in its own transaction to minimize lock time
- Error handling isolates failed actions to prevent blocking others
- Re-enabled scheduler with optimized locking strategy

Key improvements:
- acquire_scheduler_lock() uses own connection, disposes immediately
- release_scheduler_lock() uses own connection, disposes immediately
- process_single_scheduled_action() uses transaction-per-action approach
- Prevents long-running transactions that were blocking web requests

russell@unturf.com is the boss
This commit is contained in:
Russell Ballestrini 2025-08-08 21:40:13 -04:00
parent 3cede5ba07
commit 6a3dd08728
2 changed files with 120 additions and 77 deletions
Download patch file Download diff file Expand all files Collapse all files

2
.gitlab-ci.yml
View file

@ -63,7 +63,7 @@ deploy:
[Service]
WorkingDirectory=${APP_DIR}
Environment="PATH=${VENV_DIR}/bin" "DISABLE_SCHEDULER=true"
Environment="PATH=${VENV_DIR}/bin"
ExecStart=${VENV_DIR}/bin/uwsgi \\
--master \\
--enable-threads \\

131
app.py
View file

@ -622,10 +622,20 @@ def reader_required(view_func):
################################################################################
def acquire_scheduler_lock(dbsession):
"""Acquire the scheduler lock."""
def acquire_scheduler_lock():
"""Acquire the scheduler lock with minimal transaction time."""
process_id = f"{os.getpid()}_{int(time.time())}"
# Use a separate short-lived session just for the lock
main_engine = create_engine(
DB_URL,
connect_args={"check_same_thread": False},
poolclass=StaticPool,
)
MainSessionFactory = sessionmaker(bind=main_engine)
try:
with MainSessionFactory() as dbsession:
# Try to get existing lock
lock = dbsession.query(SchedulerLock).first()
@ -645,7 +655,6 @@ def acquire_scheduler_lock(dbsession):
lock.locked = False
lock.locked_at = None
lock.process_id = None
dbsession.commit()
if not lock.locked:
lock.locked = True
@ -655,16 +664,30 @@ def acquire_scheduler_lock(dbsession):
return True
return False
finally:
main_engine.dispose()
def release_scheduler_lock(dbsession):
"""Release the scheduler lock."""
def release_scheduler_lock():
"""Release the scheduler lock with minimal transaction time."""
# Use a separate short-lived session just for the lock
main_engine = create_engine(
DB_URL,
connect_args={"check_same_thread": False},
poolclass=StaticPool,
)
MainSessionFactory = sessionmaker(bind=main_engine)
try:
with MainSessionFactory() as dbsession:
lock = dbsession.query(SchedulerLock).first()
if lock and lock.locked:
lock.locked = False
lock.locked_at = None
lock.process_id = None
dbsession.commit()
finally:
main_engine.dispose()
def process_scheduled_actions():
@ -678,23 +701,27 @@ def process_scheduled_actions():
main_dbsession = MainSessionFactory()
try:
if not acquire_scheduler_lock(main_dbsession):
if not acquire_scheduler_lock():
return # Another process is already running
log.info("Scheduler acquired lock, processing scheduled actions...")
# Get all namespaces
# Get all namespaces with the existing session
namespaces = main_dbsession.query(Namespace).all()
# Process each namespace separately to avoid long transactions
for namespace in namespaces:
try:
process_namespace_scheduled_actions(namespace.id)
except Exception as e:
log.error(f"Error processing namespace {namespace.id}: {e}")
log.info("Finished processing scheduled actions")
except Exception as e:
log.error(f"Error in scheduler: {e}")
finally:
release_scheduler_lock(main_dbsession)
release_scheduler_lock()
main_dbsession.close()
main_engine.dispose()
@ -731,43 +758,9 @@ def process_namespace_scheduled_actions(namespace_id):
.all()
)
# Process each action individually with short transactions
for action in pending_actions:
try:
media = (
namespace_dbsession.query(Media)
.filter_by(id=action.media_id)
.first()
)
if not media:
action.status = "failed"
continue
if action.action_type == "delete":
namespace_dbsession.delete(media)
log.info(f"Scheduled delete executed for media {media.filename}")
elif action.action_type == "set_public":
media.visibility = "public"
log.info(
f"Scheduled set_public executed for media {media.filename}"
)
elif action.action_type == "set_private":
media.visibility = "private"
log.info(
f"Scheduled set_private executed for media {media.filename}"
)
elif action.action_type == "set_unlisted":
media.visibility = "unlisted"
log.info(
f"Scheduled set_unlisted executed for media {media.filename}"
)
action.status = "completed"
except Exception as e:
log.error(f"Error processing scheduled action {action.id}: {e}")
action.status = "failed"
namespace_dbsession.commit()
process_single_scheduled_action(namespace_dbsession, action)
except Exception as e:
log.error(f"Error processing namespace {namespace_id}: {e}")
@ -778,6 +771,56 @@ def process_namespace_scheduled_actions(namespace_id):
namespace_engine.dispose()
def process_single_scheduled_action(dbsession, action):
"""Process a single scheduled action with minimal transaction time."""
try:
# Start a new transaction for this action
with dbsession.begin():
# Refresh the action to get latest state
dbsession.refresh(action)
# Skip if already processed
if action.status != "pending":
return
media = (
dbsession.query(Media)
.filter_by(id=action.media_id)
.first()
)
if not media:
action.status = "failed"
action.completed_at = datetime.datetime.utcnow()
return
if action.action_type == "delete":
dbsession.delete(media)
log.info(f"Scheduled delete executed for media {media.filename}")
elif action.action_type == "set_public":
media.visibility = "public"
log.info(f"Scheduled set_public executed for media {media.filename}")
elif action.action_type == "set_private":
media.visibility = "private"
log.info(f"Scheduled set_private executed for media {media.filename}")
elif action.action_type == "set_unlisted":
media.visibility = "unlisted"
log.info(f"Scheduled set_unlisted executed for media {media.filename}")
action.status = "completed"
action.completed_at = datetime.datetime.utcnow()
except Exception as e:
log.error(f"Error processing scheduled action {action.id}: {e}")
try:
# Try to mark as failed in a separate transaction
with dbsession.begin():
dbsession.refresh(action)
action.status = "failed"
action.completed_at = datetime.datetime.utcnow()
except Exception as inner_e:
log.error(f"Failed to mark action {action.id} as failed: {inner_e}")
# Global variables for scheduler thread
scheduler_thread = None
scheduler_running = False