Deploying Sidekiq To Heroku Discussion
Chris, another great episode! At the moment I have this in my Procfile: worker: bundle exec sidekiq -c 5 -q default -q mailers, since the only type of worker in my app is the mailer worker. At the moment it's not totally clear for me when/how I have to define the type of the worker. Let's say I would add another type (not mailer) of job to my app, but I don't want to scale my heroku dyno up since my one worker dyno has enough capacity to do both the mailer type and the other type of bg job. Could you tell me how you deal with that? I hope you will publish more videos about getting apps production ready on heroku :)
Great question. I would actually remove the queues you listed and then Sidekiq will process everything by default. You don't really need to separate out queues until you have so much background work happening that it makes sense to separate and prioritize them.
In the video you switched capitalization for the config option - for puma it should be `-C` while sidekiq expects `-c`.
Maybe they both support -C? Sidekiq's docs still show the capital one. https://github.com/mperham/...
W. Ama
Hi,
I have a problem with heroku.
It seems like every time I make changes to the migrations and run git push heroku master, heroku run rake db:migrate, I get an error in development when opening the current user's profile. All existing users seem to have this problem, while the new users after the migration update work well. This means that I would have to delete the existing users and recreate them, but that would not be possible if actual users use the app.
Is there a way to fix this without having to delete the existing users after every migration update on heroku?
Thank you.