Celery task update_storage_usage_cache has no time_limit — worker may hang indefinitely

[Joaquinriosheredia]

Hi,

While analyzing the codebase with python-vibe-guard, I found a potential
issue in api/caching/tasks.py line 117.

The Celery task update_storage_usage_cache has no time_limit or
soft_time_limit defined:

@app.task(max_retries=5, default_retry_delay=10)
def update_storage_usage_cache(target_id, target_guid, per_page=...):
    storage_usage_total = compute_storage_usage_total(...)

compute_storage_usage_total executes a paginated SQL query over file
versions. If the query blocks (database lock, large table, index contention),
the Celery worker will hang indefinitely — there is no timeout mechanism
to recover it.

Suggested fix:

@app.task(
    max_retries=5,
    default_retry_delay=10,
    soft_time_limit=270,
    time_limit=300,
)
def update_storage_usage_cache(...):

With a SoftTimeLimitExceeded handler to log which target_guid was
being processed when the timeout fired.

Tool used: python-vibe-guard —
AST-based runtime anti-pattern scanner validated on 902 real repositories.