You're missing the third, and more likely option -- there's something
wrong with your code. I've been running MySQL and Postgres installs
for many years, and I've never encountered the problem you describe.
This isn't a case of "oh, that just happens and we live with it".
It's also possible that you have found a bug, but it's an edge case
particular to your particular usage. Looking at your stack trace,
you're getting the exception at the end of request processing when
Django tries to 'clean up' and close the connection to the database.
However, at that point, the database has already been closed. This
means that either:
* You have already (implicitly or explicitly) closed the connection
* The database connection has dropped out for some reason
* The 'end of request' signal has been fired twice.
Absent of any specific details, I can't narrow down the problem any
more than that. We would need to see some specific code to give any
more advice. If you can reduce your code to a minimal test case that
demonstrates the problem, that might help.
Yours,
Russ Magee %-)