Łukasz Niemier
Elixir/Erlang Developer
Senior Backend Engineer that got bored with Ruby circa 2016 and migrated to Elixir for some change in life. Since then it went south pretty fast - EEF Observability WG membership, pandemic, wedding, kid. Now works remotely for Remote (pun intended) and tries to manage the life. Author of new Elixir's Logger implementation that landed in 1.10
Past Activities
Code BEAM V America
11.05 - 11.45
Who supervises supervisors?
OTP supervisors allow programmers to write reliable software in occurrence of errors in our code, but what happens when there is a bug in the OTP itself? Or in cases when our system becomes unresponsive despite the fact that all processes are up and running. That is where the system supervisor, like systemd, comes to play to help operations to keep the application up and running while being able to reliably observe the state of application.
OBJECTIVE:
Present the community that the supervisor approach is not distinctive to the OTP and that it is very handy solution to publish and manage BEAM applications within the existing systems without using additional tooling like k8s or Nomad. How to run applications in constrained and safe environments without usage of Docker/Podman while making it easy to run and manage on host OSes.
AUDIENCE:
Intermediate/advanced developers who want to manage their own applications and DevOps that are managing BEAM applications.