AK 4.1 updates (#727)

Author: hvishwanath

Dockerfile.multiplatform

@@ -0,0 +1,76 @@
+FROM alpine:3.19
+
+# Install runtime dependencies
+RUN apk add --no-cache \
+    curl \
+    git \
+    nodejs \
+    npm \
+    bash \
+    ca-certificates \
+    libc6-compat
+
+# Install Go 1.22.2 (required by go.mod)
+ARG TARGETARCH
+RUN if [ "$TARGETARCH" = "arm64" ]; then \
+        GO_ARCH="arm64"; \
+    else \
+        GO_ARCH="amd64"; \
+    fi && \
+    curl -L "https://go.dev/dl/go1.22.2.linux-${GO_ARCH}.tar.gz" -o go.tar.gz && \
+    tar -C /usr/local -xzf go.tar.gz && \
+    rm go.tar.gz
+
+# Set Go environment variables
+ENV GOPATH="/go"
+ENV PATH="/usr/local/go/bin:${PATH}"
+ENV PATH="${GOPATH}/bin:${PATH}"
+
+# Download Hugo binary directly (much more space efficient than compiling)
+ARG TARGETARCH
+RUN if [ "$TARGETARCH" = "arm64" ]; then \
+        HUGO_ARCH="arm64"; \
+    else \
+        HUGO_ARCH="amd64"; \
+    fi && \
+    echo "Downloading Hugo for architecture: ${HUGO_ARCH}" && \
+    curl -L "https://github.com/gohugoio/hugo/releases/download/v0.123.7/hugo_extended_0.123.7_linux-${HUGO_ARCH}.tar.gz" -o hugo.tar.gz && \
+    echo "Extracting Hugo..." && \
+    tar -xzf hugo.tar.gz && \
+    echo "Contents after extraction:" && \
+    ls -la && \
+    echo "Hugo binary details:" && \
+    ls -la hugo && \
+    echo "Moving Hugo binary..." && \
+    cp hugo /usr/local/bin/hugo && \
+    chmod +x /usr/local/bin/hugo && \
+    echo "Hugo binary location and permissions:" && \
+    ls -la /usr/local/bin/hugo && \
+    echo "Testing Hugo binary:" && \
+    ldd /usr/local/bin/hugo && \
+    /usr/local/bin/hugo version && \
+    rm hugo.tar.gz hugo
+
+# Install global dependencies
+RUN npm install -g postcss postcss-cli autoprefixer
+
+# Copy entrypoint script
+COPY scripts/entrypoint.sh /usr/local/bin/
+RUN chmod +x /usr/local/bin/entrypoint.sh
+
+# Create working directory
+WORKDIR /src
+
+# Configure Git to trust the working directory
+RUN git config --global --add safe.directory /src
+
+# Verify installations
+RUN node --version && \
+    npm --version && \
+    npx --version && \
+    hugo version && \
+    go version
+
+EXPOSE 1313
+
+ENTRYPOINT ["/usr/local/bin/entrypoint.sh"]

Makefile

@@ -1,20 +1,31 @@
 # Hugo configuration
 OUTPUT_DIR := output
-DOCKER_IMAGE := hvishwanath/hugo:v0.123.7-ext
+HUGO_BASE_IMAGE := hvishwanath/hugo:v0.123.7-ext-multiplatform
+DOCKER_IMAGE := $(HUGO_BASE_IMAGE)
 #PROD_IMAGE := hvishwanath/kafka-site-md:1.2.0
 PROD_IMAGE := us-west1-docker.pkg.dev/play-394201/kafka-site-md/kafka-site-md:1.6.0
 
-.PHONY: build serve clean docker-image prod-image prod-run buildx-setup
+.PHONY: build serve clean docker-image hugo-base-multi-platform prod-image prod-run buildx-setup ghcr-prod-image
 
 # Setup buildx for multi-arch builds
 buildx-setup:
 	docker buildx create --name multiarch --driver docker-container --use || true
 	docker buildx inspect multiarch --bootstrap
 
-# Build the Docker image
+# Build the Docker image (single platform)
 docker-image:
 	docker build -t $(DOCKER_IMAGE) . --push
 
+# Build and push multi-platform Hugo base image
+hugo-base-multi-platform: buildx-setup
+	docker buildx build \
+		--platform linux/amd64,linux/arm64 \
+		--tag $(HUGO_BASE_IMAGE) \
+		--file Dockerfile.multiplatform \
+		--build-arg BUILDKIT_INLINE_CACHE=1 \
+		--push \
+		.
+
 # Build the static site using Docker
 build: 
 	docker pull $(DOCKER_IMAGE)
@@ -48,8 +59,19 @@ prod-run: prod-image
 	docker pull $(PROD_IMAGE)
 	docker run --rm -p 8080:80 $(PROD_IMAGE)
 
+# Build and push production image to GHCR
+ghcr-prod-image: build buildx-setup
+	docker buildx build \
+		--platform linux/amd64,linux/arm64 \
+		--tag ghcr.io/$(shell basename $(shell git rev-parse --show-toplevel))/kafka-site-md:prod-$(shell git rev-parse --abbrev-ref HEAD) \
+		--tag ghcr.io/$(shell basename $(shell git rev-parse --show-toplevel))/kafka-site-md:prod-$(shell git rev-parse --short HEAD) \
+		--tag ghcr.io/$(shell basename $(shell git rev-parse --show-toplevel))/kafka-site-md:prod-$(shell date +%Y%m%d-%H%M%S) \
+		--file Dockerfile.prod \
+		--push \
+		.
+
 # Clean the output directory and remove Docker images
 clean:
 	rm -rf $(OUTPUT_DIR)
-	docker rmi $(DOCKER_IMAGE) $(PROD_IMAGE)
+	docker rmi $(DOCKER_IMAGE) $(HUGO_BASE_IMAGE) $(PROD_IMAGE)
 	docker buildx rm multiarch || true

README.md

@@ -167,4 +167,4 @@ make clean
 4. Test locally using `make serve`
 5. Submit a pull request
 
-For more details about the migration to Markdown and the overall architecture, see [KIP-1133](https://cwiki.apache.org/confluence/display/KAFKA/KIP-1133%3A+AK+Documentation+and+Website+in+Markdown).
+For more details about the migration to Markdown and the overall architecture, see [KIP-1133](https://cwiki.apache.org/confluence/display/KAFKA/KIP-1133%3A+AK+Documentation+and+Website+in+Markdown).

content/en/0110/streams/core-concepts.md

@@ -80,7 +80,7 @@ Kafka Streams allows direct read-only queries of the state stores by methods, th
 
 # Processing Guarantees
 
-In stream processing, one of the most frequently asked question is "does my stream processing system guarantee that each record is processed once and only once, even if some failures are encountered in the middle of processing?" Failing to guarantee exactly-once stream processing is a deal-breaker for many applications that cannot tolerate any data-loss or data duplicates, and in that case a batch-oriented framework is usually used in addition to the stream processing pipeline, known as the [Lambda Architecture](http://lambda-architecture.net/). Prior to 0.11.0.0, Kafka only provides at-least-once delivery guarantees and hence any stream processing systems that leverage it as the backend storage could not guarantee end-to-end exactly-once semantics. In fact, even for those stream processing systems that claim to support exactly-once processing, as long as they are reading from / writing to Kafka as the source / sink, their applications cannot actually guarantee that no duplicates will be generated throughout the pipeline. Since the 0.11.0.0 release, Kafka has added support to allow its producers to send messages to different topic partitions in a [transactional and idempotent manner](/#semantics), and Kafka Streams has hence added the end-to-end exactly-once processing semantics by leveraging these features. More specifically, it guarantees that for any record read from the source Kafka topics, its processing results will be reflected exactly once in the output Kafka topic as well as in the state stores for stateful operations. Note the key difference between Kafka Streams end-to-end exactly-once guarantee with other stream processing frameworks' claimed guarantees is that Kafka Streams tightly integrates with the underlying Kafka storage system and ensure that commits on the input topic offsets, updates on the state stores, and writes to the output topics will be completed atomically instead of treating Kafka as an external system that may have side-effects. To read more details on how this is done inside Kafka Streams, readers are recommended to read [KIP-129](https://cwiki.apache.org/confluence/display/KAFKA/KIP-129%3A+Streams+Exactly-Once+Semantics). In order to achieve exactly-once semantics when running Kafka Streams applications, users can simply set the `processing.guarantee` config value to **exactly_once** (default value is **at_least_once**). More details can be found in the [**Kafka Streams Configs**](/0110/documentation#streamsconfigs) section. 
+In stream processing, one of the most frequently asked question is "does my stream processing system guarantee that each record is processed once and only once, even if some failures are encountered in the middle of processing?" Failing to guarantee exactly-once stream processing is a deal-breaker for many applications that cannot tolerate any data-loss or data duplicates, and in that case a batch-oriented framework is usually used in addition to the stream processing pipeline, known as the [Lambda Architecture](https://en.wikipedia.org/wiki/Lambda_architecture). Prior to 0.11.0.0, Kafka only provides at-least-once delivery guarantees and hence any stream processing systems that leverage it as the backend storage could not guarantee end-to-end exactly-once semantics. In fact, even for those stream processing systems that claim to support exactly-once processing, as long as they are reading from / writing to Kafka as the source / sink, their applications cannot actually guarantee that no duplicates will be generated throughout the pipeline. Since the 0.11.0.0 release, Kafka has added support to allow its producers to send messages to different topic partitions in a [transactional and idempotent manner](/#semantics), and Kafka Streams has hence added the end-to-end exactly-once processing semantics by leveraging these features. More specifically, it guarantees that for any record read from the source Kafka topics, its processing results will be reflected exactly once in the output Kafka topic as well as in the state stores for stateful operations. Note the key difference between Kafka Streams end-to-end exactly-once guarantee with other stream processing frameworks' claimed guarantees is that Kafka Streams tightly integrates with the underlying Kafka storage system and ensure that commits on the input topic offsets, updates on the state stores, and writes to the output topics will be completed atomically instead of treating Kafka as an external system that may have side-effects. To read more details on how this is done inside Kafka Streams, readers are recommended to read [KIP-129](https://cwiki.apache.org/confluence/display/KAFKA/KIP-129%3A+Streams+Exactly-Once+Semantics). In order to achieve exactly-once semantics when running Kafka Streams applications, users can simply set the `processing.guarantee` config value to **exactly_once** (default value is **at_least_once**). More details can be found in the [**Kafka Streams Configs**](/0110/documentation#streamsconfigs) section. 
 
 [Previous](/0110/streams/developer-guide) [Next](/0110/streams/architecture)
 

content/en/10/streams/core-concepts.md

@@ -82,7 +82,7 @@ Kafka Streams allows direct read-only queries of the state stores by methods, th
 
 # Processing Guarantees
 
-In stream processing, one of the most frequently asked question is "does my stream processing system guarantee that each record is processed once and only once, even if some failures are encountered in the middle of processing?" Failing to guarantee exactly-once stream processing is a deal-breaker for many applications that cannot tolerate any data-loss or data duplicates, and in that case a batch-oriented framework is usually used in addition to the stream processing pipeline, known as the [Lambda Architecture](http://lambda-architecture.net/). Prior to 0.11.0.0, Kafka only provides at-least-once delivery guarantees and hence any stream processing systems that leverage it as the backend storage could not guarantee end-to-end exactly-once semantics. In fact, even for those stream processing systems that claim to support exactly-once processing, as long as they are reading from / writing to Kafka as the source / sink, their applications cannot actually guarantee that no duplicates will be generated throughout the pipeline. Since the 0.11.0.0 release, Kafka has added support to allow its producers to send messages to different topic partitions in a [transactional and idempotent manner](/#semantics), and Kafka Streams has hence added the end-to-end exactly-once processing semantics by leveraging these features. More specifically, it guarantees that for any record read from the source Kafka topics, its processing results will be reflected exactly once in the output Kafka topic as well as in the state stores for stateful operations. Note the key difference between Kafka Streams end-to-end exactly-once guarantee with other stream processing frameworks' claimed guarantees is that Kafka Streams tightly integrates with the underlying Kafka storage system and ensure that commits on the input topic offsets, updates on the state stores, and writes to the output topics will be completed atomically instead of treating Kafka as an external system that may have side-effects. To read more details on how this is done inside Kafka Streams, readers are recommended to read [KIP-129](https://cwiki.apache.org/confluence/display/KAFKA/KIP-129%3A+Streams+Exactly-Once+Semantics). In order to achieve exactly-once semantics when running Kafka Streams applications, users can simply set the `processing.guarantee` config value to **exactly_once** (default value is **at_least_once**). More details can be found in the [**Kafka Streams Configs**](/10/documentation#streamsconfigs) section. 
+In stream processing, one of the most frequently asked question is "does my stream processing system guarantee that each record is processed once and only once, even if some failures are encountered in the middle of processing?" Failing to guarantee exactly-once stream processing is a deal-breaker for many applications that cannot tolerate any data-loss or data duplicates, and in that case a batch-oriented framework is usually used in addition to the stream processing pipeline, known as the [Lambda Architecture](https://en.wikipedia.org/wiki/Lambda_architecture). Prior to 0.11.0.0, Kafka only provides at-least-once delivery guarantees and hence any stream processing systems that leverage it as the backend storage could not guarantee end-to-end exactly-once semantics. In fact, even for those stream processing systems that claim to support exactly-once processing, as long as they are reading from / writing to Kafka as the source / sink, their applications cannot actually guarantee that no duplicates will be generated throughout the pipeline. Since the 0.11.0.0 release, Kafka has added support to allow its producers to send messages to different topic partitions in a [transactional and idempotent manner](/#semantics), and Kafka Streams has hence added the end-to-end exactly-once processing semantics by leveraging these features. More specifically, it guarantees that for any record read from the source Kafka topics, its processing results will be reflected exactly once in the output Kafka topic as well as in the state stores for stateful operations. Note the key difference between Kafka Streams end-to-end exactly-once guarantee with other stream processing frameworks' claimed guarantees is that Kafka Streams tightly integrates with the underlying Kafka storage system and ensure that commits on the input topic offsets, updates on the state stores, and writes to the output topics will be completed atomically instead of treating Kafka as an external system that may have side-effects. To read more details on how this is done inside Kafka Streams, readers are recommended to read [KIP-129](https://cwiki.apache.org/confluence/display/KAFKA/KIP-129%3A+Streams+Exactly-Once+Semantics). In order to achieve exactly-once semantics when running Kafka Streams applications, users can simply set the `processing.guarantee` config value to **exactly_once** (default value is **at_least_once**). More details can be found in the [**Kafka Streams Configs**](/10/documentation#streamsconfigs) section. 
 
 [Previous](/10/streams/tutorial) [Next](/10/streams/architecture)