Improved Dockerfile by adding multi-stage build and increasing image size to allow faster restart.

Author: AlexanderSchultheiss

Dockerfile

@@ -1,18 +1,29 @@
 # syntax=docker/dockerfile:1
+FROM alpine:3.14
+
+# Prepare the environment
+RUN apk add maven git
+
+# Build the jar files
+WORKDIR /home/user
+COPY local-maven-repo ./local-maven-repo
+COPY src ./src
+COPY pom.xml .
+RUN mvn package || exit
+
+RUN git clone --progress https://git.busybox.net/busybox/
+RUN git clone --progress https://github.com/torvalds/linux.git
+
 FROM ubuntu:20.04
 
 # Create a user
-ARG USER_ID
-ARG GROUP_ID
-RUN addgroup --gid $GROUP_ID user
-RUN adduser --disabled-password  --home /home/user --gecos '' --uid $USER_ID --ingroup user user
+RUN adduser --disabled-password  --home /home/user --gecos '' user
 
 # Prepare the environment
 RUN apt-get update \
     && apt-get install -y --no-install-recommends tzdata
 RUN apt-get install -y --no-install-recommends build-essential libelf-dev libssl-dev flex bison libselinux1-dev git
 
-
 # Setup working directory
 WORKDIR /resources
 COPY ./docker-resources/ubuntu-repos.txt .
@@ -26,17 +37,33 @@ RUN apt-get update \
     && apt-get install -y --no-install-recommends gcc-4.4
 RUN ln -s /bin/gcc-4.4 /bin/gcc
 RUN rm -rf VariabilityExtraction
-RUN ls -al
 RUN gcc --version
 RUN java -version
-RUN apt-get install -y --no-install-recommends maven bc
+RUN apt-get install -y --no-install-recommends bc
 
+
+WORKDIR /home/user
+# Copy JAR from previous stage
+COPY --from=0 /home/user/target /home/user/target
+RUN cp target/VariabilityExtraction-*-jar-with* .
+
+# Copy required scripts and properties
 COPY docker-resources/extract.sh /home/user/
+COPY docker-resources/entrypoint.sh /home/user/
+COPY docker-resources/fix-perms.sh /home/user/
+COPY docker-resources/KernelHaven.jar /home/user/
 COPY docker-resources/variability_analysis_BusyBox.properties /home/user/
 COPY docker-resources/variability_analysis_Linux.properties /home/user/
+
 RUN mkdir -p /home/user/extraction-results/output
 RUN chown user:user /home/user -R
-WORKDIR /home/user
+RUN chmod +x entrypoint.sh
+RUN chmod +x fix-perms.sh
 RUN chmod +x extract.sh
 
-ENTRYPOINT ["./extract.sh"]
+# Copy repositories from previous stage
+COPY --from=0 /home/user/linux /home/user/linux
+COPY --from=0 /home/user/busybox /home/user/busybox
+
+ENTRYPOINT ["./entrypoint.sh", "./extract.sh"]
+USER user

docker-resources/extract.sh

@@ -5,32 +5,12 @@ java -version
 
 cd /home/user || exit
 ls -l
-git clone --progress https://oauth2:[email protected]/mse/VariantSync/VariabilityExtraction.git
-cd VariabilityExtraction || exit
-git status
-echo "Listing files in VariabilityExtraction"
-ls -l
-echo ""
-
-echo "Building with Maven"
-mvn package || exit
-echo ""
-
-echo "Copying resources"
-cp target/VariabilityExtraction-*-jar-with* docker-resources/KernelHaven.jar ..
-cd ..
-echo ""
-
-echo "Files in WORKDIR"
-ls .
-echo ""
 
 echo "Files in extraction-results"
 ls -l extraction-results
 
 if [ "$1" == 'busybox' ]
 then
-    git clone --progress https://git.busybox.net/busybox/
     echo "Executing variability extraction of BusyBox."
     if [ $# == 1 ]
     then
@@ -44,8 +24,6 @@ then
     fi
 elif [ "$1" == 'linux' ]
 then
-    echo "Cloning Linux, this will take quite some time."
-    git clone --progress https://github.com/torvalds/linux.git
     echo "Executing variability extraction of Linux."
     if [ $# == 1 ]
     then

docker-resources/maven-setup.sh

@@ -27,7 +27,7 @@ extraction.timeout = 600
 ###  Analysis Parameters ###
 # How many commits should be analyzed in parallel? Warning: each task requires a considerable amount
 # of resources
-analysis.number_of_tasks = 3
+analysis.number_of_tasks = 1
 
 # The folders which are considered by KbuildMiner can be specified manually with the following property. Normally,
 # they are determined automatically, which we consider to be the best option in most cases

docker-resources/variability_analysis_Linux.properties

@@ -1,4 +1,5 @@
 @echo "Starting extraction"
+if not exist "extraction-results" mkdir extraction-results
 
 @if "%1"=="busybox" GOTO BUSYBOX
 @if "%1"=="linux" GOTO LINUX
@@ -7,18 +8,12 @@
 @GOTO AFTER
 
 :BUSYBOX
-@if not exist "extraction-results" mkdir extraction-results
-@if not exist "extraction-results/busybox" mkdir extraction-results/busybox
-
 @echo "Starting the extraction"
 @docker run --rm -v "%cd%/extraction-results/busybox":"/home/user/extraction-results/output" variability-extraction %*
 
 @GOTO AFTER
 
 :LINUX
-@if not exist "extraction-results" mkdir extraction-results
-@if not exist "extraction-results/linux" mkdir extraction-results/linux
-
 @echo "Starting the extraction"
 @docker run --rm -v "%cd%/extraction-results/linux":"/home/user/extraction-results/output" variability-extraction %*