+1 . Scheduler is a great tool because its feature packed and exploits what is "at hand" in a normal deployment environment (and it's the best shot at replacing cron & likes). The minute I had a redis-backed scheduler at hand (its there, sitting on my disk) I was kinda sad, because what makes the scheduler great is the ease of interaction with the database.... All users needed to switch to a complete different toolset to queue tasks, at that point... better to rely on something battle-tested (always a problem in web2py environment, too many newbies and little or few testers).
For a realtime offloading tool Celery it's probably the best solution out there (kinda harder to config, but hey....), pyres it's my favourite and rq and huey are littler in comparison but much easier to config and solve the 80% of the problems. I used them only as a mere task dispatchers/processors, i.e. outside the web world. PS: Bruno, I tried to follow your code when you originally posted it, but it had 2 "problems" : - 1: every request ends up building a connection towards Redis (I think a singleton is needed) - 2: I remember some problems about scheduling functions defined in models (i.e. adapting to the "crazy" execution model of web2py) As soon as I'll get home I'll try again, but is there a chance that in the meantime those problems have been already fixed ? -- --- 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 web2py+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
