Reduce size of Khoj Docker images by removing layers and caches

- Align Dockerfile and prod.Dockerfile code
- Reduce Docker image size by 25% by reducing Docker layers and
  removing package caches
This commit is contained in:
Debanjum Singh Solanky 2024-09-29 03:52:11 -07:00
parent 9b10b3e7a1
commit 1e8ce52d98
2 changed files with 38 additions and 26 deletions

View file

@ -3,19 +3,23 @@ FROM ubuntu:jammy
LABEL homepage="https://khoj.dev" LABEL homepage="https://khoj.dev"
LABEL repository="https://github.com/khoj-ai/khoj" LABEL repository="https://github.com/khoj-ai/khoj"
LABEL org.opencontainers.image.source="https://github.com/khoj-ai/khoj" LABEL org.opencontainers.image.source="https://github.com/khoj-ai/khoj"
LABEL org.opencontainers.image.description="Your second brain, containerized for personal, local deployment."
# Install System Dependencies # Install System Dependencies
RUN apt update -y && apt -y install python3-pip swig curl RUN apt update -y && apt -y install \
python3-pip \
# Install Node.js and Yarn swig \
RUN curl -sL https://deb.nodesource.com/setup_20.x | bash - curl \
RUN apt -y install nodejs # Required by RapidOCR
RUN curl -sS https://dl.yarnpkg.com/debian/pubkey.gpg | apt-key add - libgl1 \
RUN echo "deb https://dl.yarnpkg.com/debian/ stable main" | tee /etc/apt/sources.list.d/yarn.list libglx-mesa0 \
RUN apt update && apt -y install yarn libglib2.0-0 && \
# Required by Next.js Web app
# Install RapidOCR dependencies curl -sL https://deb.nodesource.com/setup_20.x | bash - && \
RUN apt -y install libgl1 libgl1-mesa-glx libglib2.0-0 curl -sS https://dl.yarnpkg.com/debian/pubkey.gpg | apt-key add - && \
echo "deb https://dl.yarnpkg.com/debian/ stable main" | tee /etc/apt/sources.list.d/yarn.list && \
apt update -y && apt -y --no-install-recommends install nodejs yarn && \
apt clean && rm -rf /var/lib/apt/lists/*
# Install Application # Install Application
WORKDIR /app WORKDIR /app
@ -33,7 +37,7 @@ ENV PYTHONPATH=/app/src:$PYTHONPATH
# Go to the directory src/interface/web and export the built Next.js assets # Go to the directory src/interface/web and export the built Next.js assets
WORKDIR /app/src/interface/web WORKDIR /app/src/interface/web
RUN bash -c "yarn cache clean && yarn install --verbose && yarn ciexport" RUN bash -c "yarn install --frozen-lockfile --verbose && yarn ciexport && yarn cache clean"
WORKDIR /app WORKDIR /app
# Run the Application # Run the Application

View file

@ -1,25 +1,33 @@
# syntax=docker/dockerfile:1
FROM ubuntu:jammy FROM ubuntu:jammy
LABEL homepage="https://khoj.dev"
LABEL repository="https://github.com/khoj-ai/khoj"
LABEL org.opencontainers.image.source="https://github.com/khoj-ai/khoj" LABEL org.opencontainers.image.source="https://github.com/khoj-ai/khoj"
LABEL org.opencontainers.image.description="Your second brain, containerized for multi-user, cloud deployment"
# Install System Dependencies # Install System Dependencies
RUN apt update -y && apt -y install python3-pip libsqlite3-0 ffmpeg libsm6 libxext6 swig curl RUN apt update -y && apt -y install \
python3-pip \
# Install Node.js and Yarn libsqlite3-0 \
RUN curl -sL https://deb.nodesource.com/setup_20.x | bash - ffmpeg \
RUN apt -y install nodejs libsm6 \
RUN curl -sS https://dl.yarnpkg.com/debian/pubkey.gpg | apt-key add - libxext6 \
RUN echo "deb https://dl.yarnpkg.com/debian/ stable main" | tee /etc/apt/sources.list.d/yarn.list swig \
RUN apt update && apt -y install yarn curl && \
# Required by Next.js Web app
WORKDIR /app curl -sL https://deb.nodesource.com/setup_20.x | bash - && \
curl -sS https://dl.yarnpkg.com/debian/pubkey.gpg | apt-key add - && \
echo "deb https://dl.yarnpkg.com/debian/ stable main" | tee /etc/apt/sources.list.d/yarn.list && \
apt update -y && apt -y --no-install-recommends install nodejs yarn && \
apt clean && rm -rf /var/lib/apt/lists/*
# Install Application # Install Application
WORKDIR /app
COPY pyproject.toml . COPY pyproject.toml .
COPY README.md . COPY README.md .
ARG VERSION=0.0.0 ARG VERSION=0.0.0
RUN sed -i "s/dynamic = \\[\"version\"\\]/version = \"$VERSION\"/" pyproject.toml && \ RUN sed -i "s/dynamic = \\[\"version\"\\]/version = \"$VERSION\"/" pyproject.toml && \
TMPDIR=/home/cache/ pip install --cache-dir=/home/cache/ -e .[prod] pip install --no-cache-dir -e .[prod]
# Copy Source Code # Copy Source Code
COPY . . COPY . .
@ -29,7 +37,7 @@ ENV PYTHONPATH=/app/src:$PYTHONPATH
# Go to the directory src/interface/web and export the built Next.js assets # Go to the directory src/interface/web and export the built Next.js assets
WORKDIR /app/src/interface/web WORKDIR /app/src/interface/web
RUN bash -c "yarn cache clean && yarn install --verbose && yarn ciexport" RUN bash -c "yarn install --frozen-lockfile --verbose && yarn ciexport && yarn cache clean"
WORKDIR /app WORKDIR /app
# Run the Application # Run the Application
@ -37,4 +45,4 @@ WORKDIR /app
# but these should be passed in through the docker-compose.yml file. # but these should be passed in through the docker-compose.yml file.
ARG PORT ARG PORT
EXPOSE ${PORT} EXPOSE ${PORT}
ENTRYPOINT [ "gunicorn", "-c", "gunicorn-config.py", "src.khoj.main:app" ] ENTRYPOINT ["gunicorn", "-c", "gunicorn-config.py", "src.khoj.main:app"]