dnf config-manager --enable crb
dnf install -y epel-release
dnf install -y debootstrap

apt install -y debootstrap

¹⁾

sudo debootstrap --variant=minbase trixie ./debian http://debian/debian

²⁾

sudo tar cC debian/ . | docker image import - ingo/debian:trixie

Tag latest hinzufügen:

docker image ls ingo/debian:trixie

→ id rauskopieren (z.B. dc9f1edde160)

docker image tag dc9f1edde160 ingo/debian:latest

docker container run ingo/debian echo hello world

→ hello world

mkdir nginx
cd nginx

++++ podman | bei Podman darf die Datei auch Containerfile heißen ++++

Dockerfile

FROM ingo/debian:trixie
 
ENV DEBIAN_FRONTEND=noninteractive
RUN set -eux; \
  apt-get -qq update; \
  apt-get install -y --no-install-recommends nginx
RUN echo 'A warm welcome from your Dockerfile' > /var/www/html/index.html
EXPOSE 80
ENTRYPOINT ["/usr/sbin/nginx", "-g", "daemon off;"]
 
ARG REFRESHED_AT=2026-01-07
LABEL version="0.0.1"
LABEL maintainer="me@example.com"
LABEL org.opencontainers.image.authors="me@example.com"
LABEL org.opencontainers.image.created=$REFRESHED_AT

³⁾ ++++ ENTRYPOINT / CMD / run-Command |

["script.sh"]

script.sh

["script.sh"]

/bin/dash

script.sh /bin/dash

["script.sh"]

["httpd"]

script.sh httpd

["script.sh"]

["httpd"]

/bin/dash

script.sh /bin/dash

["/bin/sh"]

/bin/sh

["/bin/sh"]

/bin/dash

/bin/dash

/bin/bash

++++ Image mit den Erweiterungen aus dem Dockerfile bauen:

docker build -t='ingo/nginx:0.0.1' .

Image anzeigen:

docker image ls ingo/nginx:0.0.1

→ id rauskopieren (z.B. 5879d7773761) Image taggen:

docker image tag 5879d7773761 ingo/nginx:latest

Container starten:

docker container run -d -p 80:80 --name my_nginx ingo/nginx

Zugriff auf nginx testen:

curl -s http://localhost:80

→ im Browser http://localhost öffnen

Dockerfile

…
RUN ln -sf /dev/stdout /var/log/nginx/access.log \
    && ln -sf /dev/stderr /var/log/nginx/error.log
…

docker build --no-cache --pull -t='ingo/nginx:0.0.2' .

++++ podman |

podman build --no-cache -t='ingo/nginx:0.0.2' .

→ allerdings werden dann alle 12 STEPs neu gebaut.

TODO: liegt das an dem fehlenden –pull? AI Slop Vermutung: Docker built your image without errors because it keeps un-namespaced local images as-is. In contrast, Podman automatically prefixes locally built or untagged registry images with localhost/ to prevent collisions with official registries. When you use the –pull flag, Podman is forced to try downloading a fresh version of that base image from an external source. Because it sees localhost/ingo/debian:trixie, Podman literally looks for a web-facing registry running on your actual machine (https://localhost/v2/). Since you don't have a container registry service actively running on port 443 of your host machine, the network request fails with „connection refused“

Unter Debian mit Docker version 26.1.5+dfsg1, build a72d7cd klappt –pull auch nicht. ERROR: failed to solve: ingo/debian:trixie: failed to resolve source metadata for docker.io/ingo/debian:trixie: pull access denied, repository does not exist or may require authorization: server message: insufficient_scope: authorization failed ++++

Jetzt noch mal Container starten und

docker logs

ansehen.

Bauen mit einem Release Zeitstempel:

REFRESHED_AT=$(date '+%Y-%m-%d')
docker build --no-cache --pull \
  --build-arg REFRESHED_AT=$REFRESHED_AT \
  -t='ingo/nginx:0.0.2' \
  -t=ingo/nginx:$REFRESHED_AT .

Nginx as Systemd:

mkdir ~/ubi9-nginx && cd ~/ubi9-nginx
cat >Dockerfile <<EOR
FROM registry.access.redhat.com/ubi9/ubi-init
RUN cat <<'EOF' >/etc/yum.repos.d/nginx.repo
[nginx-stable]
name=nginx stable repo
baseurl=http://nginx.org/packages/rhel/9/\$basearch/
gpgcheck=1
enabled=1
gpgkey=https://nginx.org/keys/nginx_signing.key
module_hotfixes=true
EOF
 
RUN dnf -y install nginx \
 && dnf clean all \
 && systemctl enable nginx
STOPSIGNAL SIGRTMIN+3
CMD ["/sbin/init"]
EOR
 
docker build -t ubi9-nginx-systemd:0.0.1 .
docker run --name ubi9-nginx --privileged -d -p 8090:80 ubi9-nginx-systemd:0.0.1
docker exec -it ubi9-nginx nginx -v
docker exec -it ubi9-nginx systemctl status
docker exec -it ubi9-nginx journalctl -u nginx
# ups..
docker logs ubi9-nginx

# build with compose
cat >compose.yml <<EOR
services:
  nginx:
    build: .
    privileged: true
    ports:
      - "8090:80"    
    tmpfs:
      - /run
      - /run/lock
      - /tmp
    stop_signal: SIGRTMIN+3
EOR
docker compose build
docker compose up -d
docker compose exec nginx ps -ef
docker compose down

++++ this doesn't solve the console logging problem |

cat >Dockerfile <<EOR
FROM registry.access.redhat.com/ubi9/ubi-init
RUN cat <<'EOF' >/etc/yum.repos.d/nginx.repo
[nginx-stable]
name=nginx stable repo
baseurl=http://nginx.org/packages/rhel/9/\$basearch/
gpgcheck=1
enabled=1
gpgkey=https://nginx.org/keys/nginx_signing.key
module_hotfixes=true
EOF
 
RUN dnf -y install nginx \
 && dnf clean all
 
# nginx logs to files (important for tail)
RUN mkdir -p /var/log/nginx
 
# systemd override for nginx
RUN mkdir -p /etc/systemd/system/nginx.service.d && \
    cat <<'EOF' >/etc/systemd/system/nginx.service.d/override.conf
[Service]
StandardOutput=journal
StandardError=journal
EOF
 
# log forwarder service
RUN cat <<'EOF' >/etc/systemd/system/nginx-log-forwarder.service
[Unit]
Description=Nginx log forwarder
After=nginx.service
Requires=nginx.service
[Service]
Type=simple
ExecStart=/bin/sh -c '/usr/bin/tail -F /var/log/nginx/access.log /var/log/nginx/error.log'
Restart=always
[Install]
WantedBy=multi-user.target
EOF
 
RUN systemctl enable nginx \
 && systemctl enable nginx-log-forwarder
STOPSIGNAL SIGRTMIN+3
CMD ["/sbin/init"]
EOR
podman compose build
podman compose up -d
curl 127.0.0.1:8090
podman compose exec nginx systemctl status nginx-log-forwarder
podman compose exec nginx journalctl -u nginx-log-forwarder
# see access logs
podman logs nginx_nginx_1
# no logging output!

++++