BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//cfp.all-systems-go.io//all-systems-go-2026//8HRQQN
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-all-systems-go-2026-E7MHZJ@cfp.all-systems-go.io
DTSTART;TZID=CET:20260930T142000
DTEND;TZID=CET:20260930T150000
DESCRIPTION:sdme is a single static binary that boots systemd-nspawn contai
 ners using\noverlayfs copy-on-write layers\, with OCI registry pulling and
  Kubernetes Pod\nYAML support. No daemon\, no runtime dependency beyond th
 e systemd you already\nrun.\n\n**Why I built it.** Every container runtime
  reinvents what systemd already\ndoes well: process supervision\, logging\
 , cgroups\, service ordering\, socket\nactivation. sdme makes the opposite
  bet. `systemd-nspawn` already boots a full\nsystem in isolation\; overlay
 fs already gives it a cheap copy-on-write root. Put\na thin\, daemonless C
 LI over the two and you get container workflows (clone\,\nimport\, pull\, 
 run) without a new runtime to install or a daemon to keep alive.\nIt began
  as a one-command clone of the running host\, but the bet is bigger than\n
 that: lean on the init system instead of rebuilding it.\n\n**What you get.
 ** Proper machine-like containers: each one runs a full systemd\ninit with
  journald\, D-Bus\, and `systemctl`\, not a single foreground process.\nEa
 ch container is highly configurable through native systemd surfaces\, with
 \nper-container resource limits\, networking\, and drop-in config. Clone y
 our\nrunning host into a throwaway container in one command\; run a Docker
  Hub image\n(nginx\, redis\, postgres) as a managed systemd service instea
 d of a foreground\nprocess\; deploy a multi-container pod from Kubernetes 
 Pod YAML with health\nprobes\, secrets\, and configmaps. A rootfs can also
  come from any distro\, an OCI\nregistry\, a tarball\, or a QCOW2 image\, 
 and you can build your own from a\nDockerfile-like config.\n\n**How it wor
 ks.** One overlayfs mount (host `/` as the lower layer\, a fresh\nupper) a
 nd one `systemd-nspawn` invocation give you a fully booted system. The\nbi
 nary does two jobs: it manages the overlayfs layering that makes each\ncon
 tainer a cheap copy-on-write root\, and it drives systemd over D-Bus to\ns
 tart\, stop\, and supervise containers. Everything else (init\, journald\,
 \ncgroups\, service ordering) is systemd doing what it already does\, with
  no\ndaemon in the middle.
DTSTAMP:20260710T220248Z
LOCATION:Galerie
SUMMARY:Containers without a new runtime: sdme on systemd-nspawn - Alexandr
 e Fiori
URL:https://cfp.all-systems-go.io/all-systems-go-2026/talk/E7MHZJ/
END:VEVENT
END:VCALENDAR