GitHub user dosubot[bot] added a comment to the discussion: Need help setting 
up Superset chart in AWS EKS behind ALB

<!-- Answer -->
That error means Superset’s SQLAlchemy connection pool is exhausted—by default, 
each pod allows 5 concurrent DB connections with up to 10 overflow, and new 
requests wait up to 30 seconds before failing. This is common in Kubernetes/EKS 
when you have multiple pods, slow DB responses, or long-running queries. Each 
pod has its own pool, so total DB connections = (number of pods) × (pool_size + 
max_overflow) [details](https://github.com/apache/superset/discussions/35997).

To fix this, increase the pool size and overflow in your configOverrides. Add 
to your `superset_config.py` (or Helm `configOverrides`):

```python
SQLALCHEMY_ENGINE_OPTIONS = {
    "pool_size": 20,         # or higher, depending on DB capacity
    "max_overflow": 30,      # or higher
    "pool_timeout": 60,      # seconds to wait for a connection
}
```

Monitor your DB to ensure it can handle the total connections. Also, check for 
slow or stuck queries in your DB logs, and consider enabling async query 
execution with Celery to offload long-running queries 
[example](https://github.com/apache/superset/issues/33218).

If you have multiple pods, make sure your DB’s max_connections is high enough 
for all pods combined. After updating, restart your Superset pods to apply the 
changes.

For long queries, also increase these timeouts in your configOverrides:

```python
SUPERSET_WEBSERVER_TIMEOUT = 1800
SQLLAB_TIMEOUT = 1800
```

And set matching timeouts in your ALB/Ingress annotations.

If you still see pool exhaustion, check for session/transaction leaks or DB 
resource bottlenecks.


<!-- Dosu Comment Footer -->
*To reply, just mention [@dosu](https://go.dosu.dev/dosubot).*

---
<sup>How did I do? 
[Good](https://app.dosu.dev/response-feedback/8f4e22d9-511a-407a-8fdb-8dcaa5ebe6b7?feedback_type=great_response)
 | 
[Irrelevant](https://app.dosu.dev/response-feedback/8f4e22d9-511a-407a-8fdb-8dcaa5ebe6b7?feedback_type=irrelevant_answer)
 | 
[Incorrect](https://app.dosu.dev/response-feedback/8f4e22d9-511a-407a-8fdb-8dcaa5ebe6b7?feedback_type=incorrect_sources)
 | 
[Verbose](https://app.dosu.dev/response-feedback/8f4e22d9-511a-407a-8fdb-8dcaa5ebe6b7?feedback_type=too_verbose)
 | 
[Hallucination](https://app.dosu.dev/response-feedback/8f4e22d9-511a-407a-8fdb-8dcaa5ebe6b7?feedback_type=hallucination)
 | [Report 
🐛](https://app.dosu.dev/response-feedback/8f4e22d9-511a-407a-8fdb-8dcaa5ebe6b7?feedback_type=bug_report)
 | 
[Other](https://app.dosu.dev/response-feedback/8f4e22d9-511a-407a-8fdb-8dcaa5ebe6b7?feedback_type=other)</sup>

[![Chat with 
Dosu](https://dosu.dev/dosu-chat-badge.svg)](https://app.dosu.dev/a28d3c7e-a9d3-459e-9fb6-3a6f9ff4f357/ask?utm_source=github)&nbsp;[![Open
 in 
Cursor](https://dosu.dev/dosu-in-cursor.svg)](https://cursor.com/link/prompt?text=That%20error%20means%20Superset%E2%80%99s%20SQLAlchemy%20connection%20pool%20is%20exhausted%E2%80%94by%20default%2C%20each%20pod%20allows%205%20concurrent%20DB%20connections%20with%20up%20to%2010%20overflow%2C%20and%20new%20requests%20wait%20up%20to%2030%20seconds%20before%20failing.%20This%20is%20common%20in%20Kubernetes/EKS%20when%20you%20have%20multiple%20pods%2C%20slow%20DB%20responses%2C%20or%20long-running%20queries.%20Each%20pod%20has%20its%20own%20pool%2C%20so%20total%20DB%20connections%20%3D%20%28number%20of%20pods%29%20%C3%97%20%28pool_size%20%2B%20max_overflow%29%C2%A0%5Bdetails%5D%28https%3A//github.com/apache/superset/discussions/35997%29.%0A%0ATo%20fix%20this%2C%20increase%20the%20pool%20size%20and%20overflow%20in%20your%20configOverride
 
s.%20Add%20to%20your%20%60superset_config.py%60%20%28or%20Helm%20%60configOverrides%60%29%3A%0A%0A%60%60%60python%0ASQLALCHEMY_ENGINE_OPTIONS%20%3D%20%7B%0A%20%20%20%20%22pool_size%22%3A%2020%2C%20%20%20%20%20%20%20%20%20%23%20or%20higher%2C%20depending%20on%20DB%20capacity%0A%20%20%20%20%22max_overflow%22%3A%2030%2C%20%20%20%20%20%20%23%20or%20higher%0A%20%20%20%20%22pool_timeout%22%3A%2060%2C%20%20%20%20%20%20%23%20seconds%20to%20wait%20for%20a%20connection%0A%7D%0A%60%60%60%0A%0AMonitor%20your%20DB%20to%20ensure%20it%20can%20handle%20the%20total%20connections.%20Also%2C%20check%20for%20slow%20or%20stuck%20queries%20in%20your%20DB%20logs%2C%20and%20consider%20enabling%20async%20query%20execution%20with%20Celery%20to%20offload%20long-running%20queries%C2%A0%5Bexample%5D%28https%3A//github.com/apache/superset/issues/33218%29.%0A%0AIf%20you%20have%20multiple%20pods%2C%20make%20sure%20your%20DB%E2%80%99s%20max_connections%20is%20high%20enough%20for%20all%20pods%20combined.%20After%20u
 
pdating%2C%20restart%20your%20Superset%20pods%20to%20apply%20the%20changes.%0A%0AFor%20long%20queries%2C%20also%20increase%20these%20timeouts%20in%20your%20configOverrides%3A%0A%0A%60%60%60python%0ASUPERSET_WEBSERVER_TIMEOUT%20%3D%201800%0ASQLLAB_TIMEOUT%20%3D%201800%0A%60%60%60%0A%0AAnd%20set%20matching%20timeouts%20in%20your%20ALB/Ingress%20annotations.%0A%0AIf%20you%20still%20see%20pool%20exhaustion%2C%20check%20for%20session/transaction%20leaks%20or%20DB%20resource%20bottlenecks.)&nbsp;[![Join
 
Discord](https://img.shields.io/badge/join-5865F2?logo=discord&logoColor=white&label=)](https://go.dosu.dev/discord-bot)&nbsp;[![Share
 on 
X](https://img.shields.io/badge/X-share-black)](https://twitter.com/intent/tweet?text=%40dosu_ai%20helped%20me%20solve%20this%20issue!&url=https%3A//github.com/apache/superset/discussions/36296)

GitHub link: 
https://github.com/apache/superset/discussions/36296#discussioncomment-15128167

----
This is an automatically sent email for [email protected].
To unsubscribe, please send an email to: 
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to