I've used HireFire on heroku for a couple of projects to handle workers (not dynos). It's worked great for occasional (user triggered) jobs. When using it for larger batches of jobs (rake triggered), I've had it choke on me a couple of times when going from zero to N workers. Result: that app has a minimum_worker of 1.
-N On Jul 14, 8:43 am, James Miller <[email protected]> wrote: > My main use case so far is on small apps where running a worker for the > whole month isn't necessary -- it fires up a worker, processes a job or two > and shuts it down. If you read through the code/specs, it's pretty straight > forward -- it checks for the count() of jobs in Delayed::Job or Resque and > acts accordingly -- I haven't had any trouble with stability. > > On Thu, Jul 14, 2011 at 8:21 AM, Patrick Crowley <[email protected]>wrote: > > > > > > > > > > Yes, I'm using it but only using the OSS version as a gem in the app -- > > haven't looked at the SaaS stuff. > > > Cool. Have you run into any issues with it? > > > He mentions that the web service is more reliable, stable, etc. Just > > wondering if you've had any problems or needed to work around any issues. > > > -- Patrick > > > -- > > SD Ruby mailing list > > [email protected] > >http://groups.google.com/group/sdruby -- SD Ruby mailing list [email protected] http://groups.google.com/group/sdruby
