Closed tobyych closed 2 years ago
At a Ctrl-C, I can imagine some internal state is in flux, but maybe we can improve something there.
But what do you mean 'occasionally otherwise'? That doesn't sound nice. Would you be able to enable some tracing? See https://python-oracledb.readthedocs.io/en/latest/user_guide/tracing.html#low-level-python-oracledb-driver-tracing
I did some searching and it looks like this issue is due to interpreter shutdown and the fact that in some cases, module globals are set to the value None
which prevents this code from running properly. This issue has been resolved in most situations from Python 3.4 onwards, but you must have a case where that is not true! Since this is not something specific to python-oracledb only occurs during interpreter shutdown I will go ahead and close this. If you have more details on the environment and a test case that demonstrates the problem, feel free to re-open with that information and I'll take another look.
platform.platform: Linux-5.10.60.1-microsoft-standard-WSL2-x86_64-with-glibc2.29 sys.maxsize > 2**32: True platform.python_version: 3.8.10 oracledb.version: 1.0.3
There is a message in console reading "Exception ignored".
No, running Thin mode.
I'm running other's code and there is a little wrapper class as follows. I run a Flask app that connects to Oracle DB, and the message always shows when I do Ctrl+C to stop the process. I see this message occasionally otherwise, but not sure about the root cause.