Re: Crash reporting for TDF builds

Il 31/05/2016 12:49, Markus Mohrhard ha scritto:
> Hey Riccardo,
>
> On Mon, May 30, 2016 at 9:47 AM, Riccardo Magliocchetti
> <riccardo.magliocchetti@gmail.com <mailto:riccardo.magliocchetti@gmail.com>> wrote:
>
>     Yep, since you are using uwsgi (and if you are willing to use it also in
>     development) you can leverage its spooler or cron functionalities [1] to
>     handle these tasks. Or they can be implemented as django management commands
>     so you don't need another script to call them via http.
>
>     [1] http://uwsgi-docs.readthedocs.io/en/latest/PythonDecorators.html
>
> What would be the advantage to writing a python management script that works
> through http?

None, what i am suggesting is to avoid http for management commands :)

> My plan was to write a python management script and add it as a
> cron job (about one minute for the processing of uploaded crashes and once a day
> for the crash count).

Calling a cron every minute seems a bit silly though, better use a queue in my 
opinion (like uwsgi spooler).

> The advantage that I currently see for an independent script is that I can call
> it from wherever I am. Not only if I'm logged-in. What would be the benefit of
> the other solutions?

It's self contained and you don't need to setup anything outside your app.

Hope to have some time to poke at this the next weekend.

--
Riccardo Magliocchetti
@rmistaken

http://menodizero.it