hw3 & hw4 - CodeProcessor, CI/CD

.env.example

@@ -1,3 +1,51 @@
+## docker compose
+
+POSTGRES_USER=postgres
+POSTGRES_PASSWORD=postgres
+POSTGRES_DB=remote_code
+POSTGRES_PORT=5432
+
+REDIS_PORT=6379
+
+RABBITMQ_USER=guest
+RABBITMQ_PASSWORD=guest
+RABBITMQ_PORT=5672
+
+HTTP_PORT=8000
+METRICS_PORT=9100 # server / producer
+PROCESSOR_METRICS_PORT=9101 # consumer
+
+## runtime - not injected into containers, set via compose
+
 HTTP_ADDR=:8000
-TASK_PROCESSING_TIME=2s
 SHUTDOWN_TIMEOUT=10s
+
+METRICS_ADDR=:9100
+METRICS_SHUTDOWN_TIMEOUT=5s
+
+DATABASE_URL=postgres://${POSTGRES_USER}:${POSTGRES_PASSWORD}@localhost:${POSTGRES_PORT}/${POSTGRES_DB}?sslmode=disable
+MIGRATIONS_DIR=migrations
+
+REDIS_ADDR=localhost:${REDIS_PORT}
+REDIS_PASSWORD=
+REDIS_DB=0
+SESSION_TTL=168h
+
+RABBITMQ_URL=amqp://${RABBITMQ_USER}:${RABBITMQ_PASSWORD}@localhost:${RABBITMQ_PORT}/
+TASK_QUEUE_NAME=tasks
+TASK_QUEUE_PREFETCH=1
+TASK_QUEUE_RECONNECT_DELAY=1s
+
+PHILHARMONIC_URL=http://localhost:5555
+# empty means manager runs without auth
+PHILHARMONIC_TOKEN=
+PHILHARMONIC_IMAGE=sandbox:latest
+PHILHARMONIC_TASK_TIMEOUT=30s
+PHILHARMONIC_POLL_INTERVAL=1s
+# unset means PHILHARMONIC_TASK_TIMEOUT + 15s
+PHILHARMONIC_POLL_TIMEOUT=
+PHILHARMONIC_CPU=0.5
+# bytes
+PHILHARMONIC_MEMORY=268435456
+# CI publishes the sandbox to GHCR on every push to main
+PHILHARMONIC_IMAGE=ghcr.io/belyaevedu/backend-final/sandbox:latest

.github/workflows/ci.yml

@@ -0,0 +1,127 @@
+name: CI
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+
+jobs:
+  lint:
+    name: vet & lint
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: actions/setup-go@v5
+        with:
+          go-version-file: go.mod
+
+      - name: vet
+        run: go vet ./...
+
+      - name: golangci-lint
+        uses: golangci/golangci-lint-action@v8
+        with:
+          version: v2.11.3
+
+  publish-sandbox:
+    name: publish sandbox image to ghcr
+    # push to main only basically
+    if: github.event_name == 'push'
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      packages: write
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: docker/setup-buildx-action@v3
+
+      - id: repo
+        name: lowercase repo name by ghcr requirement
+        run: echo "name=${GITHUB_REPOSITORY,,}" >> "$GITHUB_OUTPUT"
+
+      - name: login to GHCR
+        uses: docker/login-action@v3
+        with:
+          registry: ghcr.io
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: build and push
+        uses: docker/build-push-action@v6
+        with:
+          context: .
+          file: sandbox/Dockerfile
+          push: true
+          tags: |
+            ghcr.io/${{ steps.repo.outputs.name }}/sandbox:latest
+            ghcr.io/${{ steps.repo.outputs.name }}/sandbox:${{ github.sha }}
+          cache-from: type=gha
+          cache-to: type=gha,mode=max
+
+  e2e:
+    name: e2e (compose + philharmonic + pytest)
+    runs-on: ubuntu-latest
+    needs: lint
+    env:
+      DATABASE_URL: postgres://postgres:postgres@localhost:5432/remote_code?sslmode=disable
+      REDIS_ADDR: localhost:6379
+      RABBITMQ_URL: amqp://guest:guest@localhost:5672/
+      PHILHARMONIC_URL: http://localhost:5555
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: actions/setup-go@v5
+        with:
+          go-version-file: go.mod
+
+      - name: infra
+        run: docker compose up -d --wait postgres redis rabbitmq
+
+      - name: build sandbox image
+        run: docker build -t sandbox:latest -f sandbox/Dockerfile .
+
+      - name: start philharmonic manager + worker on the runner
+        run: |
+          git clone --depth 1 --branch v0.1.1 https://github.com/belyaevedu/philharmonic /tmp/philharmonic
+          (cd /tmp/philharmonic && CGO_ENABLED=0 go build -o /tmp/phrm .)
+
+          /tmp/phrm manager --host 127.0.0.1 --port 5555 --workers 127.0.0.1:5556 > /tmp/phrm-manager.log 2>&1 &
+          /tmp/phrm worker --host 127.0.0.1 --port 5556 --name dev-worker > /tmp/phrm-worker.log 2>&1 &
+
+          # wait until the manager reads the worker's stats
+          for i in $(seq 1 30); do
+            curl -sf localhost:5555/nodes | grep -q '"Cores":[1-9]' && break
+            sleep 1
+          done
+          curl -sf localhost:5555/nodes | grep -q '"Cores":[1-9]'
+
+      - name: build & run server and processor
+        run: |
+          go build -o /tmp/server ./cmd/server
+          go build -o /tmp/processor ./cmd/processor
+          /tmp/server > /tmp/server.log 2>&1 &
+          METRICS_ADDR=:9101 /tmp/processor > /tmp/processor.log 2>&1 &
+          # any HTTP answer (even 401/404) means the server is serving.
+          # it only serves after the migrations are done
+          curl -s -o /dev/null --retry 30 --retry-delay 1 --retry-all-errors \
+            localhost:8000/status/00000000-0000-0000-0000-000000000000
+
+      - name: pytest
+        run: |
+          python -m pip install pytest requests
+          python -m pytest tests/hw2.py tests/hw3.py -v
+
+      - name: debug logs
+        if: failure()
+        run: |
+          docker compose ps
+          docker compose logs --tail 30 postgres redis rabbitmq || true
+          for f in /tmp/server.log /tmp/processor.log /tmp/phrm-manager.log /tmp/phrm-worker.log; do
+            if [ -f "$f" ]; then echo "== $f =="; tail -n 50 "$f"; fi
+          done
+
+      - name: teardown
+        if: always()
+        run: docker compose down -v

Dockerfile

@@ -1,24 +1,28 @@
-FROM golang:1.26-alpine AS builder
+FROM golang:1.26-alpine@sha256:ce864e7223ac17b1775e6fd0b4c0db580c2eb50e7953a427916379e4b92a1628 AS builder
 
 WORKDIR /src
 
-RUN apk add --no-cache git ca-certificates
+RUN apk add --no-cache git~=2.54.0-r0 ca-certificates~=20260611-r0
 
 COPY go.mod go.sum ./
 RUN go mod download
 
-COPY . .
+COPY cmd ./cmd
+COPY internal ./internal
+COPY migrations ./migrations
 
 RUN CGO_ENABLED=0 GOOS=linux \
     go build -trimpath -ldflags="-s -w" -o /out/server ./cmd/server
 
-FROM gcr.io/distroless/static-debian12:nonroot AS runtime
+FROM gcr.io/distroless/static-debian12:nonroot@sha256:afa5c872c891853ca7fcf1f12c3edb23f7eeef36189728842dd51042ff57f7ab AS runtime
 
-WORKDIR /
+WORKDIR /app
 
 COPY --from=builder /out/server /app/server
+COPY --from=builder /src/migrations /app/migrations
 
 EXPOSE 8000
+EXPOSE 9100
 
 USER 65532:65532
 

Dockerfile.processor

@@ -0,0 +1,24 @@
+FROM golang:1.26-alpine@sha256:ce864e7223ac17b1775e6fd0b4c0db580c2eb50e7953a427916379e4b92a1628 AS builder
+
+WORKDIR /src
+
+RUN apk add --no-cache git~=2.54.0-r0 ca-certificates~=20260611-r0
+
+COPY go.mod go.sum ./
+RUN go mod download
+
+COPY cmd ./cmd
+COPY internal ./internal
+
+RUN CGO_ENABLED=0 GOOS=linux \
+    go build -trimpath -ldflags="-s -w" -o /out/processor ./cmd/processor
+
+FROM gcr.io/distroless/static-debian12:nonroot@sha256:afa5c872c891853ca7fcf1f12c3edb23f7eeef36189728842dd51042ff57f7ab AS runtime
+
+WORKDIR /app
+
+EXPOSE 9100
+
+COPY --from=builder /out/processor /app/processor
+
+ENTRYPOINT ["/app/processor"]

cmd/processor/main.go

@@ -0,0 +1,86 @@
+package main
+
+import (
+	"context"
+	"fmt"
+	"log"
+	"os/signal"
+	"syscall"
+	"time"
+
+	"github.com/belyaevedu/remote-code-service/internal/config"
+	"github.com/belyaevedu/remote-code-service/internal/controller"
+	"github.com/belyaevedu/remote-code-service/internal/domain"
+	"github.com/belyaevedu/remote-code-service/internal/repository/postgres"
+	"github.com/belyaevedu/remote-code-service/internal/repository/queue"
+	"github.com/belyaevedu/remote-code-service/internal/service"
+)
+
+func main() {
+	dbCfg, err := config.LoadDBConfig()
+	if err != nil {
+		log.Fatalf("invalid db config: %v", err)
+	}
+	queueCfg, err := config.LoadQueueConfig()
+	if err != nil {
+		log.Fatalf("invalid queue config: %v", err)
+	}
+	philCfg, err := config.LoadPhilharmonicConfig()
+	if err != nil {
+		log.Fatalf("invalid philharmonic config: %v", err)
+	}
+	metricsCfg, err := config.LoadMetricsConfig()
+	if err != nil {
+		log.Fatalf("invalid metrics config: %v", err)
+	}
+
+	ctx, stop := signal.NotifyContext(context.Background(), syscall.SIGINT, syscall.SIGTERM)
+	defer stop()
+
+	db, err := postgres.New(ctx, dbCfg)
+	if err != nil {
+		log.Fatalf("postgres: %v", err)
+	}
+	defer db.Close()
+
+	if err := db.WaitReady(ctx); err != nil {
+		log.Fatalf("waiting for schema: %v", err)
+	}
+
+	phrmExecutor := service.NewPhilharmonicExecutor(philCfg)
+
+	// pulling the sandbox image on every worker before starting to process
+	prewarmCtx, cancel := context.WithTimeout(ctx, 2*time.Minute)
+	if err := phrmExecutor.PreWarm(prewarmCtx); err != nil {
+		log.Printf("processor: sandbox pre-warm failed: %v (continuing)", err)
+	} else {
+		log.Printf("processor: sandbox image %q pre-warmed on workers", philCfg.SandboxImage)
+	}
+	cancel()
+
+	executor := service.NewInstrumentedExecutor(phrmExecutor)
+
+	consumer := queue.NewConsumer(queueCfg)
+
+	metricsServer := controller.NewMetricsServer(metricsCfg.Addr, metricsCfg.ShutdownTimeout)
+	go func() {
+		if err := metricsServer.Start(ctx); err != nil {
+			log.Fatalf("metrics server exited with error: %v", err)
+		}
+	}()
+
+	handle := func(ctx context.Context, msg domain.TaskMessage) error {
+		result, err := executor.Execute(ctx, msg)
+		if err != nil {
+			return fmt.Errorf("execute task %s: %w", msg.TaskID, err)
+		}
+
+		return db.SaveTaskResult(ctx, msg.TaskID, &domain.Result{Output: result.Output})
+	}
+
+	log.Printf("processor: consuming queue %q", queueCfg.Queue)
+	if err := consumer.Consume(ctx, handle); err != nil {
+		log.Fatalf("consumer exited with error: %v", err)
+	}
+	log.Printf("processor stopped gracefully")
+}