Docker build for multi-architecture amd/arm

.github/workflows/ci-build-manual-crf.yml

@@ -32,17 +32,21 @@ jobs:
       - name: Create more disk space
         run: sudo rm -rf /usr/share/dotnet && sudo rm -rf /opt/ghc && sudo rm -rf "/usr/local/share/boost" && sudo rm -rf "$AGENT_TOOLSDIRECTORY"
       - uses: actions/checkout@v4
-      - name: Build and push
-        id: docker_build
-        uses: mr-smithers-excellent/docker-build-push@v6
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@v3
+      - name: Set up Docker Buildx
+        id: buildx
+        uses: docker/setup-buildx-action@v3
+      - name: Login to Docker Hub
+        uses: docker/login-action@v1
         with:
-          dockerfile: Dockerfile.crf
           username: ${{ secrets.DOCKERHUB_USERNAME_LFOPPIANO }}
           password: ${{ secrets.DOCKERHUB_TOKEN_LFOPPIANO }}
-          image: lfoppiano/grobid
-          registry: docker.io
-          pushImage: true
-          tags:  |
-            latest-develop, ${{ github.event.inputs.custom_tag}}
-      - name: Image digest
-        run: echo ${{ steps.docker_build.outputs.digest }}
+      - name: Build and push
+        uses: docker/build-push-action@v3
+        with:
+          file: Dockerfile.crf
+          platforms: linux/amd64,linux/arm64
+          push: true
+          tags: |
+            lfoppiano/grobid:latest-develop, lfoppiano/grobid:${{ github.event.inputs.custom_tag}}

Dockerfile.crf

@@ -13,7 +13,7 @@
 # -------------------
 # build builder image
 # -------------------
-FROM openjdk:17-jdk-slim as builder
+FROM eclipse-temurin:17-jdk-focal as builder
 
 USER root
 
@@ -63,7 +63,7 @@ RUN rm -rf grobid-source
 # -------------------
 # build runtime image
 # -------------------
-FROM openjdk:17-slim
+FROM eclipse-temurin:17-jre-focal
 
 RUN apt-get update && \
     apt-get -y upgrade && \
@@ -80,7 +80,18 @@ WORKDIR /opt/grobid
 
 COPY --from=builder /opt/grobid .
 
-ENV GROBID_SERVICE_OPTS "-Djava.library.path=grobid-home/lib/lin-64:grobid-home/lib/lin-64/jep --add-opens java.base/java.lang=ALL-UNNAMED --add-opens java.base/sun.nio.ch=ALL-UNNAMED --add-opens java.base/java.io=ALL-UNNAMED"
+ARG TARGETARCH
+ARG PROCESS_MECHANISM
+
+RUN if [ "$TARGETARCH" = "arm64" ]; then \
+        PROCESS_MECHANISM="-Djdk.lang.Process.launchMechanism=vfork "; \
+    else \
+        PROCESS_MECHANISM=""; \
+    fi
+
+RUN echo "Selected process spawning mechanism: $PROCESS_MECHANISM"
+
+ENV GROBID_SERVICE_OPTS "$PROCESS_MECHANISM-Djava.library.path=grobid-home/lib/lin-64:grobid-home/lib/lin-64/jep --add-opens java.base/java.lang=ALL-UNNAMED --add-opens java.base/sun.nio.ch=ALL-UNNAMED --add-opens java.base/java.io=ALL-UNNAMED"
 
 CMD ["./grobid-service/bin/grobid-service"]
 

doc/Grobid-docker.md

@@ -77,7 +77,7 @@ It is then possible to select other Deep Learning models for other processing st
 
 ## CRF-only image
 
-The process for retrieving and running the image is as follow:
+The process for retrieving and running the image is as follows:
 
 - Pull the image from docker HUB (check the [latest version number](https://hub.docker.com/r/lfoppiano/grobid/tags)):
 
@@ -115,6 +115,24 @@ Access the service:
 
 Grobid web services are then available as described in the [service documentation](https://grobid.readthedocs.io/en/latest/Grobid-service/).
 
+### ARM64
+
+Since version 0.8.1 the CRF-only image is available for both x86_64 and arm64. 
+The `lfoppiano/grobid:{latest_grobid_version}` support both architectures by default. 
+
+To run it on arm64, use the following command:
+
+```bash
+docker run --ulimit core=0 --platform linux/amd64 --init -p 8070:8070 --name grobid lfoppiano/grobid:{latest_grobid_version}
+```
+
+As discussed [here](https://github.com/kermitt2/grobid/issues/1119#issuecomment-2362204879), it is mandatory to provide the parameter `--platform linux/amd64` to force the usage of amd64 architecture and avoid the raising of the rosetta error: 
+
+```
+rosetta error: failed to open elf at /lib64/ld-linux-x86-64.so.2
+```
+
+**NOTE**: The arm64 image has been tested only on `macOs` (not linux `arm64`).
 
 ## Configure using the yaml config file