I gave talk about this topic on CODE Beam V Americas, but I wasn't really satisfied with it. In this post I will try to describe what my presentation was meant to be about.

If you are wondering about the presentation, the slides are on SpeakerDeck.

Abstract§

Most of the operating systems are multi-process and multi-user operating systems. This has a lot of positive aspects, like to be able to do more than one thing at the time at our devices, but it introduces a lot of complexities that in most cases are hidden from the users and developers. These things still need to be handled in one or another way. The most basic problems are:

• some processes need to be started before user can interact with the OS in meaningful (for them) way (for example mounting filesystems, logging, etc.)
• some processes require strict startup ordering, for example you may need logging to be started before starting HTTP server
• system operator somehow need to know when the process is ready to do their work, which is often some time after process start
• system operator should be able to check process state in case when debugging is needed, most commonly via logs
• shutdown of the processes should be handled in a way, that will allow other processes to be shut down cleanly (for example application that uses DB should be down before DB itself)

Why we need system supervisor?§

System supervisor is a process started early in the OS boot, that should handle starting and managing all other processes that will be run on our system. It is often the init process (first process started by the OS that is running with PID 1) or it is first (and sometimes only) process started by the init process. Popular examples of such supervisors (often integrated with init systems):

• SysV which is "traditional" implementation that originates at UNIX System V (hence the name)
• BSD init that with some variations is used in BSD-based OSes (NetBSD, FreeBSD), it shares some similarities to SysV init and services description is provided by shell scripts
• OpenRC that also uses shell-based scripts for service description, used by Linux distributions like Gentoo or Alpine
• launchd that is used on Darwin (macOS, iPadOS, iOS, watchOS) systems that uses XML-based plists for services description
• runit which is small init and supervisor, but quite capable, for example used by Void Linux
• Upstart created by Canonical Ltd. as a replacement for SysV-like init system in Ubuntu (no longer in use in Ubuntu), still used in some distributions like ChromeOS or Synology NAS
• systemd (this is the name, not "SystemD") that was created by Red Hat employee, (in)famous Lennart Poettering, and later was adopted by almost all major Linux distributions which spawned some heated discussion about it

In this article I will focus on systemd, and its approach to "new-style system daemons".

DISCLAIMER