Hi,

I'm not entirely sure whether this is technically a Mezzanine issue or a 
Django issue. I noticed recently after a deploy (Python 2.7, Mezzanine 
3.0.9, Django 1.6, Postgres 9.3 on a Debian 7 box) that I was generating 
big Postgres log files, filled with this event:

https://gist.github.com/ciaranbradley/11207009

I tracked it down to the poll_twitter command running in cron. The command 
was working as expected, but it would generate a new entry each time it 
ran. After finding this similar issue:

http://stackoverflow.com/questions/16841505/django-resetting-postgres-connection

I applied the technique to the poll_twitter command last night and the 
problem seems to be gone. My poll_twitter now looks like this:

https://gist.github.com/ciaranbradley/11207319

The command still seems to be up and running as normal and my log file 
loves me again.

I don't see a specific instruction on Django docs to ensure you close the 
database connection when you create a command, so it might be fair to 
assume that the engine should close the connection gracefully after a 
command is run. Also I note that my deploy is low traffic (just a sandbox 
at the minute) so I suspect that higher traffic sites might be immune if 
the command uses an already open persistent connection. But that's just 
pure speculation on my part.

Is anyone else experiencing similar?

Thanks

Ciaran
  

-- 
You received this message because you are subscribed to the Google Groups 
"Mezzanine 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.

Reply via email to