Ansible role for installing Erlang/OTP via the RabbitMQ team's apt packages. Manages EPMD lifecycle (socket or service mode with systemd drop-ins), optional rebar3 installation, and GPG key / apt pinning for the RabbitMQ Erlang repository.

• Ansible 2.15 or higher (ansible-core 2.15+ / Ansible 8+).
• Ubuntu 20.04+ or Debian 11+.

A partial listing of commonly adjusted variables. See defaults/main.yml for the full set.

• erlang_version_major (default "27") — OTP major series for the package repository.
• erlang_rebar (default false) — install rebar3 build tool.
• erlang_rebar_version (default "3.27.0") — rebar3 version to install.

By default the role manages the host's EPMD lifecycle: it detects available systemd units (preferring epmd.socket, falling back to epmd.service), drops in an override pinning the listen address and port, enables and starts the unit, and verifies reachability via epmd -names.

• erlang_epmd_managed (default true) — opt out by setting to false.
• erlang_epmd_unit_kind (default "socket") — preferred unit kind. "service" for always-on daemon.
• erlang_epmd_listen_address (default "127.0.0.1") — bind address. Non-loopback requires the safety latch below.
• erlang_epmd_port (default 4369) — listener port.
• erlang_epmd_allow_remote (default false) — safety latch. Must be true to permit a non-loopback bind address.

Service mode note: when erlang_epmd_unit_kind: service, the role masks epmd.socket to prevent port competition. Switching back to socket mode requires systemctl unmask epmd.socket.

Tests use Molecule with Docker and Testinfra. Run the full suite with make test, or target a specific platform (e.g. make test-ubuntu2404).

The test suite verifies Erlang package installation, OTP version, apt repository and pinning configuration, EPMD systemd drop-in and reachability, and rebar3 installation. Tests run across all supported Linux distributions.

This project has been developed with AI assistance. Contributions making use of AI generated content are welcome, however they must be human reviewed prior to submission as pull requests, or issues. All contributors must be able to fully explain and defend any AI generated code, documentation, issues, or tests they submit. Contributions making use of AI must have this explicitly declared in the pull request or issue. This also applies to utilization of AI for reviewing of pull requests.

Are always welcome!