Why i try to select native filters options results in Network error: Network error while attempting to fetch resource in superset5.0

[sunilreddy0720]

I have a dashboard with more the 10 filters, when i try to select filters value continously results in Network error: Network error while attempting to fetch resource
below is my screenshot which i'm getting in superset5.0

when i select some filter value when issue occurred it loads again normally like below

@dosu help us solving this

[sunilreddy0720]

@dosu help us on this

[dosubot[bot]]

This is a common issue when dashboards have many native filters. Each filter triggers a separate database query to populate its dropdown values, and with 10+ filters firing concurrently, the backend can get overwhelmed — resulting in timeouts that surface as "Network error while attempting to fetch resource" [1].

Here are the most likely causes and solutions:

1. Increase timeout settings

The default SUPERSET_WEBSERVER_TIMEOUT is 60 seconds [2]. With many filters, queries may collectively exceed this. In your superset_config.py:

SUPERSET_WEBSERVER_TIMEOUT = 120  # or higher

Also ensure your reverse proxy (Nginx, etc.) timeout is set higher than this value.

2. Increase worker capacity

More Gunicorn workers/threads help handle concurrent filter requests [3]:

SERVER_WORKER_AMOUNT = 12
SERVER_THREADS_AMOUNT = 15

3. Enable caching

Configure Redis caching to serve filter values from cache on subsequent loads, reducing database pressure.

4. Use filter dependencies

If some filters are related, configure parent-child dependencies between them. This serializes queries (reducing concurrent load) and narrows child filter queries using parent filter values [4].

5. Set a fetch values predicate

On your datasets, configure the "Autocomplete query predicate" (fetch_values_predicate) to limit the scope of filter value queries [5].

6. Reduce filter row limits

In superset_config.py, lower the max rows fetched for filter dropdowns:

NATIVE_FILTER_DEFAULT_ROW_LIMIT = 500  # default is 1000

Database-specific notes:

• DuckDB: Doesn't support concurrent queries — pin to duckdb-engine==0.14.1 [6]
• BigQuery: Make filters depend on a time filter to pass partition filters [4]

Could you share which database you're using and what your current timeout/proxy settings look like? That would help narrow down the root cause further.

To reply, just mention @dosu.

---

Share context across your team and agents. Try Dosu.