I think this question is more about PyDAL than web2py. What I'm doing is using the web2py shell to call functions and do stuff within a web2py app, but without actually running the web server. Everything happens within a Python shell. In this environment, database transactions are not automatically committed. That is fine as far as it goes, but the problem is that, as far as I can tell, if the interpreter exits with open DB transactions, those transactions remain hanging in a "zombie" state. This can cause future DB operations to hang (if, for instance, an zombie transaction was trying to update a row).
Is there any way to guarantee that at *interpreter exit* all DB transactions will be closed (presumably aborted) and nothing will be left hanging? Is there a reason that PyDAL does not include, for instance, an atexit handler to ensure this? It is very annoying to be doing stuff in the shell and then have the DB frozen because I ended the interpreter session and forgot to close the connection. It seems like a natural safety maneuver to automatically close down all connections at interpreter exit, but it appears PyDAL doesn't do this. -- Resources: - http://web2py.com - http://web2py.com/book (Documentation) - http://github.com/web2py/web2py (Source code) - https://code.google.com/p/web2py/issues/list (Report Issues) --- You received this message because you are subscribed to the Google Groups "web2py-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
