What is the default limit set for requests per minute/second

[Devika7733]

Bug description

I am running load testing on superset using k6 operator. I am testing with 5 concurrent users and getting logs as "Traceback (most recent call last):
File "/app/.venv/lib/python3.11/site-packages/flask/app.py", line 1482, in full_dispatch_request
rv = self.preprocess_request()
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/.venv/lib/python3.11/site-packages/flask/app.py", line 1974, in preprocess_request
rv = self.ensure_sync(before_func)()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/.venv/lib/python3.11/site-packages/flask_limiter/extension.py", line 1160, in _check_request_limit
raise e
File "/app/.venv/lib/python3.11/site-packages/flask_limiter/extension.py", line 1157, in _check_request_limit
self.__evaluate_limits(endpoint, all_limits)
File "/app/.venv/lib/python3.11/site-packages/flask_limiter/extension.py", line 1141, in __evaluate_limits
raise RateLimitExceeded(
flask_limiter.errors.RateLimitExceeded: 429 Too Many Requests: 50 per 1 second
2025-10-14 05:38:03,787:WARNING:superset.views.error_handling:HTTPException
Traceback (most recent call last):
File "/app/.venv/lib/python3.11/site-packages/flask/app.py", line 1482, in full_dispatch_request
rv = self.preprocess_request()
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/.venv/lib/python3.11/site-packages/flask/app.py", line 1974, in preprocess_request
rv = self.ensure_sync(before_func)()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/.venv/lib/python3.11/site-packages/flask_limiter/extension.py", line 1160, in _check_request_limit
raise e
File "/app/.venv/lib/python3.11/site-packages/flask_limiter/extension.py", line 1157, in _check_request_limit
self.__evaluate_limits(endpoint, all_limits)
File "/app/.venv/lib/python3.11/site-packages/flask_limiter/extension.py", line 1141, in __evaluate_limits
raise RateLimitExceeded(
flask_limiter.errors.RateLimitExceeded: 429 Too Many Requests: 50 per 1 second
2025-10-14 05:38:03,826:INFO:flask-limiter:ratelimit 50 per 1 second (10.40.22.127) exceeded at endpoint: global
2025-10-14 05:38:03,826:WARNING:superset.views.error_handling:HTTPException
Traceback (most recent call last):
File "/app/.venv/lib/python3.11/site-packages/flask/app.py", line 1482, in full_dispatch_request
rv = self.preprocess_request()
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/.venv/lib/python3.11/site-packages/flask/app.py", line 1974, in preprocess_request
rv = self.ensure_sync(before_func)()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/.venv/lib/python3.11/site-packages/flask_limiter/extension.py", line 1160, in _check_request_limit
raise e
File "/app/.venv/lib/python3.11/site-packages/flask_limiter/extension.py", line 1157, in _check_request_limit
self.__evaluate_limits(endpoint, all_limits)
File "/app/.venv/lib/python3.11/site-packages/flask_limiter/extension.py", line 1141, in __evaluate_limits
raise RateLimitExceeded(
flask_limiter.errors.RateLimitExceeded: 429 Too Many Requests: 50 per 1 second
2025-10-14 05:38:03,829:INFO:flask-limiter:ratelimit 50 per 1 second (10.40.22.127) exceeded at endpoint: global
2025-10-14 05:38:03,829:INFO:flask-limiter:ratelimit 50 per 1 second (10.40.22.127) exceeded at endpoint: global
2025-10-14 05:38:03,829:INFO:flask-limiter:ratelimit 50 per 1 second (10.40.22.127) exceeded at endpoint: global
2025-10-14 05:38:03,829:INFO:flask-limiter:ratelimit 50 per 1 second (10.40.22.127) exceeded at endpoint: global
2025-10-14 05:38:03,829:WARNING:superset.views.error_handling:HTTPException
Traceback (most recent call last):
File "/app/.venv/lib/python3.11/site-packages/flask/app.py", line 1482, in full_dispatch_request
rv = self.preprocess_request()
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/.venv/lib/python3.11/site-packages/flask/app.py", line 1974, in preprocess_request
rv = self.ensure_sync(before_func)()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/.venv/lib/python3.11/site-packages/flask_limiter/extension.py", line 1160, in _check_request_limit
raise e
File "/app/.venv/lib/python3.11/site-packages/flask_limiter/extension.py", line 1157, in _check_request_limit
self.__evaluate_limits(endpoint, all_limits)
File "/app/.venv/lib/python3.11/site-packages/flask_limiter/extension.py", line 1141, in __evaluate_limits
raise RateLimitExceeded(
flask_limiter.errors.RateLimitExceeded: 429 Too Many Requests: 50 per 1 second
2025-10-14 05:38:03,829:WARNING:superset.views.error_handling:HTTPException" What is the default limit set for requests per minute/second, and how many requests are expected for a medium-sized workload? How can I fix the above error? How many requests will be generated if a user accesses a dashboard?

Screenshots/recordings

No response

Superset version

5.0.0

Python version

3.11

Node version

18 or greater

Browser

Chrome

Additional context

No response

Checklist

• I have searched Superset docs and Slack and didn't find a solution to my problem.
• I have searched the GitHub issue tracker and didn't find a similar bug report.
• I have checked Superset's logs for errors and if I found a relevant Python stacktrace, I included it here as text in the "additional context" section.

[dosubot[bot]]

The default rate limit in Superset 5.0.0 is 50 requests per second for the application and 5 per second for authentication endpoints. These are set in the config file as RATELIMIT_APPLICATION = "50 per second" and AUTH_RATE_LIMIT = "5 per second", and are enabled by default in production environments. You can adjust these limits by creating or editing a superset_config.py file and setting higher values for these variables if your workload requires it. See the config.py source for details.

When a user accesses a dashboard, Superset makes separate API requests for each chart on the dashboard, plus additional requests for dashboard metadata and datasets. For example, a dashboard with 10 charts will typically generate at least 10 chart data requests, plus a few more for dashboard and dataset info. The actual number can be higher if there are filters or interactive elements, and each user session will trigger its own set of requests. For dashboards with many charts or high concurrency, the number of requests can quickly exceed the default rate limit, especially during load testing or with multiple users see discussion.

To fix the 429 error, increase the RATELIMIT_APPLICATION value in your superset_config.py to match your expected peak request rate during load testing. For a medium-sized workload (for example, 10–50 concurrent users and dashboards with 5–20 charts), you may need to set the limit to several hundred requests per second, depending on your actual usage patterns and dashboard complexity. Monitor your logs to fine-tune this value and ensure your infrastructure can handle the increased load.

If you need distributed rate limiting (across multiple Superset instances), configure RATELIMIT_STORAGE_URI to use a backend like Redis. For more advanced options, refer to the Flask-Limiter documentation.

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}    

[dosubot[bot]]

The default global rate limit in Superset 5.0.0 is set to 50 requests per second, controlled by the RATELIMIT_APPLICATION setting in config.py. This is enabled by default in production environments and is intended as a security measure to prevent DDOS attacks. You can adjust this limit by overriding the value in a custom superset_config.py file or by setting the appropriate environment variable. For example, increasing RATELIMIT_APPLICATION to a higher value will allow more requests per second and help avoid 429 errors during load testing or with higher user concurrency. See the relevant config and comments here.

When a user accesses a dashboard, Superset typically sends one API request per chart (slice) on the dashboard, plus additional requests for metadata and annotations if present. For a dashboard with 10 charts, expect at least 10 concurrent requests, and more if there are filters or annotation layers. Complex dashboards can quickly approach or exceed the default rate limit, especially with multiple users or automated load testing. See the chart loading logic here and here.

For a medium-sized workload (for example, 10–50 users and dashboards with 10–20 charts), you may need to set the rate limit significantly higher than the default, depending on your expected concurrency and dashboard complexity. It's best to monitor your actual request volume during peak usage and tune the rate limit accordingly. Always test changes in a staging environment before applying them to production.

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}