feat(ingestion-base): convert to ubuntu image (datahub-project#11263)

G-Research · Aug 28, 2024 · 2066f1f · 2066f1f
1 parent 15c1cfc
commit 2066f1f
Show file tree

Hide file tree

Showing 5 changed files with 23 additions and 9 deletions.
diff --git a/.github/workflows/docker-unified.yml b/.github/workflows/docker-unified.yml
@@ -615,7 +615,7 @@ jobs:
           target: full-install
           images: |
             ${{ env.DATAHUB_INGESTION_BASE_IMAGE }}
-          tags: ${{ needs.setup.outputs.full_tag }}
+          image_tag: ${{ needs.setup.outputs.full_tag }}
           username: ${{ secrets.ACRYL_DOCKER_USERNAME }}
           password: ${{ secrets.ACRYL_DOCKER_PASSWORD }}
           build-args: |

diff --git a/docker/datahub-ingestion-base/Dockerfile b/docker/datahub-ingestion-base/Dockerfile
@@ -5,41 +5,52 @@ ARG BASE_IMAGE=base
 ARG ALPINE_REPO_URL=http://dl-cdn.alpinelinux.org/alpine
 ARG GITHUB_REPO_URL=https://github.com
 ARG DEBIAN_REPO_URL=https://deb.debian.org/debian
+ARG UBUNTU_REPO_URL=http://ports.ubuntu.com/ubuntu-ports
 ARG PIP_MIRROR_URL=https://pypi.python.org/simple
 
-FROM powerman/dockerize:0.19 as dockerize-binary
+FROM powerman/dockerize:0.19 AS dockerize-binary
 
-FROM python:3.10 AS base
+FROM ubuntu:22.04 AS base
 
 ARG GITHUB_REPO_URL
 
 ENV DEBIAN_FRONTEND=noninteractive
 
 # Optionally set corporate mirror for deb
 ARG DEBIAN_REPO_URL
-RUN if [ "${DEBIAN_REPO_URL}" != "http://deb.debian.org/debian" ] ; then sed -i "s#http.*://deb.debian.org/debian#${DEBIAN_REPO_URL}#g" /etc/apt/sources.list.d/debian.sources ; fi
+ARG UBUNTU_REPO_URL
+RUN if [ "${DEBIAN_REPO_URL}" != "http://deb.debian.org/debian" ] ; then sed -i "s#http.*://deb.debian.org/debian#${DEBIAN_REPO_URL}#g" /etc/apt/sources.list ; fi
+RUN if [ "${UBUNTU_REPO_URL}" != "http://ports.ubuntu.com/ubuntu-ports" ] ; then sed -i "s#http.*://ports.ubuntu.com/ubuntu-ports#${UBUNTU_REPO_URL}#g" /etc/apt/sources.list ; fi
 
 # Optionally set corporate mirror for pip
 ARG PIP_MIRROR_URL
 RUN if [ "${PIP_MIRROR_URL}" != "https://pypi.python.org/simple" ] ; then pip config set global.index-url ${PIP_MIRROR_URL} ; fi
 ENV UV_INDEX_URL=${PIP_MIRROR_URL}
 
-RUN apt-get update && apt-get install -y -qq \
+RUN apt-get update && apt-get upgrade -y \
+    && apt-get install -y -qq \
+    python3 \
+    python3-pip \
+    python3-venv \
+    python-is-python3 \
     python3-ldap \
     libldap2-dev \
     libsasl2-dev \
     libsasl2-modules \
     libaio1 \
     libsasl2-modules-gssapi-mit \
     krb5-user \
+    krb5-config \
+    libkrb5-dev \
     wget \
     zip \
     unzip \
     ldap-utils \
     unixodbc \
     libodbc2 \
     && python -m pip install --no-cache --upgrade pip uv>=0.1.10 wheel setuptools \
-    && rm -rf /var/lib/apt/lists/* /var/cache/apk/*
+    && apt-get clean \
+    && rm -rf /var/lib/{apt,dpkg,cache,log}/
 
 COPY --from=dockerize-binary /usr/local/bin/dockerize /usr/local/bin
 

diff --git a/docker/datahub-ingestion-base/build.gradle b/docker/datahub-ingestion-base/build.gradle
@@ -12,7 +12,7 @@ ext {
     docker_target = project.getProperties().getOrDefault("dockerTarget", "slim")
     docker_version = "${version}${docker_target == 'slim' ? '-slim' : ''}"
 
-    revision = 4 // increment to trigger rebuild
+    revision = 5 // increment to trigger rebuild
 }
 
 docker {

diff --git a/docker/datahub-ingestion/Dockerfile b/docker/datahub-ingestion/Dockerfile
@@ -3,14 +3,17 @@ ARG APP_ENV=full
 ARG BASE_IMAGE=acryldata/datahub-ingestion-base
 ARG DOCKER_VERSION=head-full
 ARG DEBIAN_REPO_URL=https://deb.debian.org/debian
+ARG UBUNTU_REPO_URL=http://ports.ubuntu.com/ubuntu-ports
 ARG PIP_MIRROR_URL=https://pypi.python.org/simple
 
 FROM $BASE_IMAGE:$DOCKER_VERSION AS base
 
 # Optionally set corporate mirror for deb
 USER 0
 ARG DEBIAN_REPO_URL
-RUN if [ "${DEBIAN_REPO_URL}" != "http://deb.debian.org/debian" ] ; then sed -i "s#http.*://deb.debian.org/debian#${DEBIAN_REPO_URL}#g" /etc/apt/sources.list.d/debian.sources ; fi
+ARG UBUNTU_REPO_URL
+RUN if [ "${DEBIAN_REPO_URL}" != "http://deb.debian.org/debian" ] ; then sed -i "s#http.*://deb.debian.org/debian#${DEBIAN_REPO_URL}#g" /etc/apt/sources.list ; fi
+RUN if [ "${UBUNTU_REPO_URL}" != "http://ports.ubuntu.com/ubuntu-ports" ] ; then sed -i "s#http.*://ports.ubuntu.com/ubuntu-ports#${UBUNTU_REPO_URL}#g" /etc/apt/sources.list ; fi
 USER datahub
 
 # Optionally set corporate mirror for pip

diff --git a/docker/datahub-ingestion/build.gradle b/docker/datahub-ingestion/build.gradle
@@ -12,7 +12,7 @@ ext {
     docker_target = project.getProperties().getOrDefault("dockerTarget", "slim")
     docker_version = "${version}${docker_target == 'slim' ? '-slim' : ''}"
 
-    revision = 4 // increment to trigger rebuild
+    revision = 5 // increment to trigger rebuild
 }
 
 dependencies {