rbenv | Ansible role for installing rbenv

I'm trying to install dependencies by running bundle install.

My Gemfile specifies Ruby version as follows:

I updated from ruby 2.7.1 to 3.1.1, then removed Gemfile.lock and ran bundle update (it's on a dev branch, so I can throw it away if this is a bad idea, I just wanted to see if it would work).

bundle update succeeds, but when I start the server:

For reasons to do with my CI setup I've needed to change the location where I install my Ruby gems from the default location to: bundle config set path 'vendor/bundle'

However, as soon as I do so, Sorbet loses it's cool and throws out over 6,000 errors. Why would Sorbet be so sensitively dependent on where the gem is installed? (I clear all of the gems before switching location so it's very unlikely to be due to a clash of gem versions).

Summary

When gems are installed to:

I'm freshly using rvm for running a legacy project.

I installed rvm fine. Running which ruby gives the correct rvm version:

I'm running bundle install and it failed in the middle with the error:

Installing Ruby 3.0.x works fine on M1 MacBooks using rbenv or asdf. But older versions like 2.7.x and 2.6.x are having various issues. How do I fix them, without installing both x86 and ARM versions of homebrew at the same time?

Trying to install openssl on homebrew using:

In a bid to run a Rails App on my local Ubuntu, I attempted to install the gems with the command gem install bunder / bundle install. However, I got the error:

rbenv: bundler: command not found . The `bundler' command exists in these Ruby versions:

2.4
2.4.1
2.4.2
2.4.5
2.5
2.5.0
2.5.1
2.5.3
2.5.7
2.5.8
2.6
2.6.0
2.6.1
2.6.3
2.6.4
2.6.5
2.6.6
2.6.7
2.6.8

I have an app that runs on Ruby 2.7.2. However, I detected a memory leak in Puma in cluster mode which doesn't go away with preload_app! and other configurations. I suspected the main culprit is ruby's memory management. The requests combines a huge JSON with nested arrays nested hashes etc.

So my idea was to use jemalloc ruby instead of the glibc malloc. Jemalloc reduces the problem in the local environment a little bit.

The current ruby AWS ELB keeps is in /opt/elasticbeanstalk/.rbenv/shims/ruby.

I want to use RVM instead, the ruby will be compiled with clang, with O3 optimization, march and mtune set to native, and use jemalloc instead: