Peter Zhu

I've [created a project to demo this](https://github.com/peterzhu2118/rails_url_helper_demo). The test that demonstrates this can be found [here](https://github.com/peterzhu2118/rails_url_helper_demo/blob/055af4a61aa3abbcefe81f8445c9aea6e58fe6d7/test/integration/sidekiq_url_test.rb#L6).

Hey @feliperenan, Thanks for looking into this! Since this isn't a devise issue and since that issue you referenced in Rails doesn't fix this issue, should I raise this issue...

@feliperenan I dug a little deeper into this, it seems the problem is devise's `authenticate` method in the routes. The problem disappears if I change: ``` authenticate :user, ->(_u) {...

@feliperenan That's a really interesting observation. Following the redirect seems to be a temporary workaround: ``` p "new_user_session_path before request: #{new_user_session_path}" p "root_path before request: #{root_path}" get '/sidekiq' follow_redirect! p...

@feliperenan Wow that's a really detailed report! Thanks for taking the time to look into this! I'll try to do more digging into this to determine the source of the...

I think that looks good. But we should still sync the `extconf.rb` between the ruby/openssl and ruby/ruby right?

I opened a ticket here: https://bugs.ruby-lang.org/issues/18978.

Thanks for working on this @kos31de! Feel free to reach out if you need any help!

AFAIK `IO.popen` runs a forked process? ruby_memcheck should correctly handle forks. It configures valgrind to follow all forked children. https://github.com/Shopify/ruby_memcheck/blob/810e5a41ddc7698d504f0ec9a2f1b19a1a26abf4/lib/ruby_memcheck/configuration.rb#L9

During RubyKaigi, I discussed the issue of long completion times in CI when running with ruby_memcheck with @nateberkopec. He agreed that a good solution is to only run the ruby_memcheck...