Error - sqlalchemy.exc.TimeoutError

[Former user]

Hello,

I just upgraded today to 0.5rc1 and after browsing for one or two minutes, my (Pylons) webapp hangs and terminates in an "internal server error". In the logs SQLAlchemy seems to be the problem, I have the following error : "Error - sqlalchemy.exc.TimeoutError: QueuePool limit of size 5 overflow 10 reached, connection timed out, timeout 30". I didn't had this problem with version 0.4.7 and I didn't change my way to connect to the database .. so maybe a bug .. ?

Thanks

[Mike Bayer]

sorry, would need some detail on this issue. QueuePool has not been significantly modified in 0.5 and tests for pool resizing pass.

[Mike Bayer]

I would add that QueuePool had a default setting of "pool_threadlocal=True" in 0.4 which has been turned off in 0.5. So if your Pylons app opens Connections many times in a request and never returns any of them, the threadlocal setting in 0.4 may have hidden this issue assuming a fixed thread pool. So make sure you are closing your Connections after use. Adding "pool_threadlocal=True" to create_engine() would revert to 0.4's behavior.

[Mike Bayer]

Removing milestone: 0.5.0 (automated comment)