.github,docker/unified: improve caching and fix bugs (#598)

- set up a GHA scheduled job to build the container nightly 
- enabling pushing a llama-swap:unified and a llama-swap:unified-Y-M-D
image to ghcr.io
- tidy up Dockerfile to use a non-root user and llama-swap as an entry
point
This commit is contained in:
Benson Wong
2026-03-23 22:24:40 +09:00
committed by GitHub
parent 916d13f5bd
commit 2c282dccad
5 changed files with 100 additions and 34 deletions
+9 -12
View File
@@ -74,6 +74,11 @@ COPY --from=builder-base /usr/local/cuda/lib64/stubs/libcuda.so /usr/local/cuda/
RUN pip3 install --no-cache-dir numpy sentencepiece
# Create llama-swap user and config directory
RUN useradd --system --no-create-home --shell /sbin/nologin llama-swap && \
mkdir -p /etc/llama-swap/config && \
chown -R llama-swap:llama-swap /etc/llama-swap
WORKDIR /app
# Copy whisper.cpp binaries and libraries
@@ -97,17 +102,7 @@ COPY --from=llama-swap-download /install/llama-swap-version /tmp/
RUN ldconfig
# Convenience symlinks
RUN ln -sf /usr/local/bin/llama-cli /usr/local/bin/llama && \
ln -sf /usr/local/bin/whisper-cli /usr/local/bin/whisper && \
ln -sf /usr/local/bin/sd-cli /usr/local/bin/stable-diffusion
# Validate all binaries exist
RUN set -e && \
for bin in llama-server llama-cli whisper-server whisper-cli sd-server sd-cli llama-swap; do \
test -x /usr/local/bin/$bin || { echo "FATAL: $bin missing from /usr/local/bin"; exit 1; }; \
done && \
echo "All binaries validated successfully"
COPY config.example.yaml /etc/llama-swap/config/config.yaml
# Version tracking
RUN echo "llama.cpp: ${LLAMA_COMMIT_HASH}" > /versions.txt && \
@@ -118,4 +113,6 @@ RUN echo "llama.cpp: ${LLAMA_COMMIT_HASH}" > /versions.txt && \
echo "build_timestamp: $(date -u +%Y-%m-%dT%H:%M:%SZ)" >> /versions.txt
WORKDIR /models
CMD ["bash"]
USER llama-swap
ENTRYPOINT ["llama-swap"]
CMD ["-config", "/etc/llama-swap/config/config.yaml", "-listen", "0.0.0.0:8080"]