> Hi. > > I originally deployed a Pyramid app in production like this: > > nginx -> uwsgi -> pryamid > > uwsgi is for the app started with this `uwsgi --options` command > > That was great, but my needs changed as I've had to launch new > applications on the same server. I'm a bit confused with all the options. > > my (current) needs are starting to look like this: > > WebApplication – 8-16 processes > Graphite (stats) – 1-2 processes > Sentry (stats) – 1-2 processes > > > I'm not sure what makes more sense: > > 1. managing a separate uwsgi master process + processes for each of the > 3+ apps > 2. using an emperor process handle the 3+ apps > > the broodloord/zerg stuff looks really interesting, and like it might > benefit us a lot. but that would mean switching from the command-line > version to the file based config -- and that will be a chunk of devops > time. > > can someone weigh in with that they'd do ? >
I do not think zerg/broodlord is the right pattern to follow in early stage of analysis (they are pretty complex). I would go with Emperor and i would attach the sentry daemon using attach-daemon facilities: http://uwsgi-docs.readthedocs.org/en/latest/AttachingDaemons.html -- Roberto De Ioris http://unbit.it _______________________________________________ uWSGI mailing list [email protected] http://lists.unbit.it/cgi-bin/mailman/listinfo/uwsgi
