Google Cloud Run Deployment

1. Multi-Stage Dockerfile

The included Dockerfile builds a minimal, static Go binary packaged inside an Alpine/scratch base image (~15MB):

FROM golang:1.25-alpine AS builder
WORKDIR /app
COPY go.mod go.sum ./
RUN go mod download
COPY . .
RUN CGO_ENABLED=0 GOOS=linux go build -ldflags="-s -w" -o /mcp-server .

FROM alpine:3.21
RUN apk --no-cache add ca-certificates
COPY --from=builder /mcp-server /mcp-server
EXPOSE 8080
ENV PORT=8080
ENTRYPOINT ["/mcp-server"]

2. Deploying via gcloud

export GCP_PROJECT="your-project-id"

gcloud run deploy mcp-server \
    --source . \
    --project $GCP_PROJECT \
    --region us-central1 \
    --allow-unauthenticated \
    --port 8080 \
    --memory 256Mi \
    --cpu 1 \
    --session-affinity

Why --session-affinity?

While Streamable HTTP is completely stateless, --session-affinity ensures legacy clients relying on long-lived SSE connections route subsequent requests to the same container instance.