LOGC-40: Migrate to workbench ClickHouse for tests

.github/dependabot.yaml

@@ -16,3 +16,17 @@ updates:
     directory: "/images/svc-base/"
     schedule:
       interval: "weekly"
+
+  - package-ecosystem: "docker"
+    directories:
+      - /env/default
+    schedule:
+      interval: "weekly"
+    ignore:
+      - dependency-name: "clickhouse/clickhouse-server"
+    groups:
+      workbench:
+        patterns:
+          - "*"
+        update-types:
+          - patch

.github/workflows/tests.yaml

@@ -35,22 +35,34 @@ jobs:
           username: ${{ github.repository_owner }}
           password: ${{ github.token }}
 
-      - name: Start Workbench
-        uses: scality/workbench@v0.8.0
-
-      - name: Start ClickHouse
-        run: docker compose up -d
-
-      - name: Wait for ClickHouse to be ready
-        run: |
-          timeout 30 bash -c 'until docker compose exec -T clickhouse-shard-1 clickhouse-client --query "SELECT 1" > /dev/null 2>&1; do sleep 1; done'
-          timeout 30 bash -c 'until docker compose exec -T clickhouse-shard-2 clickhouse-client --query "SELECT 1" > /dev/null 2>&1; do sleep 1; done'
-
       - name: Run GolangCI linter
         uses: golangci/golangci-lint-action@v9
         with:
           version: v2.1.6
 
+      - name: Start Workbench
+        uses: scality/workbench@v0.9.0
+
+      - name: Wait for ClickHouse setup and fix permissions
+        run: |
+          # Wait for workbench setup container to complete
+          timeout 60 bash -c 'until [ "$(docker inspect -f {{.State.Status}} workbench-setup-clickhouse 2>/dev/null)" = "exited" ]; do sleep 1; done'
+
+          # Wait for ClickHouse shards to be ready (workbench uses port 9002 internally)
+          for i in {1..30}; do
+            if docker exec workbench-clickhouse-shard-1 clickhouse-client --host localhost --port 9002 --query "SELECT 1" >/dev/null 2>&1; then
+              break
+            fi
+            sleep 2
+          done
+
+          for i in {1..30}; do
+            if docker exec workbench-clickhouse-shard-2 clickhouse-client --host localhost --port 9002 --query "SELECT 1" >/dev/null 2>&1; then
+              break
+            fi
+            sleep 2
+          done
+
       - name: Run unit tests
         run: make test-coverage
 
@@ -75,9 +87,3 @@ jobs:
         run: |
           workbench logs
           workbench down
-
-      - name: Stop ClickHouse
-        if: always()
-        run: |
-          docker compose logs
-          docker compose down

Makefile

@@ -28,7 +28,7 @@ clean:
 
 test: all
 	mkdir -p $(COVDATA_DIR)
-	COVDATA_DIR=$(COVDATA_DIR) go test $(COVER_TESTFLAGS) -v ./...
+	COVDATA_DIR=$(COVDATA_DIR) go test $(COVER_TESTFLAGS) -v ./cmd/... ./pkg/...
 
 test-coverage: COVER_BUILDFLAGS=$(COVER_COMMONFLAGS)
 test-coverage: COVER_TESTFLAGS=$(COVER_COMMONFLAGS) -coverprofile cover.out

docker-compose.yml

@@ -1,44 +1,4 @@
 services:
-  clickhouse-shard-1:
-    image: docker.io/clickhouse/clickhouse-server:24.3.2.23-alpine
-    container_name: log-courier-clickhouse-shard-1
-    hostname: clickhouse-shard-1
-    ports:
-      - "9002:9000"  # Native protocol
-      - "8123:8123"  # HTTP interface
-    environment:
-      CLICKHOUSE_DB: logs
-      CLICKHOUSE_USER: default
-      CLICKHOUSE_PASSWORD: ""
-    volumes:
-      - clickhouse-shard-1-data:/var/lib/clickhouse
-      - ./deployment/clickhouse-config.xml:/etc/clickhouse-server/config.d/cluster.xml:ro
-    healthcheck:
-      test: ["CMD", "clickhouse-client", "--query", "SELECT 1"]
-      interval: 10s
-      timeout: 5s
-      retries: 5
-
-  clickhouse-shard-2:
-    image: docker.io/clickhouse/clickhouse-server:24.3.2.23-alpine
-    container_name: log-courier-clickhouse-shard-2
-    hostname: clickhouse-shard-2
-    ports:
-      - "9003:9000"  # Native protocol
-      - "8124:8123"  # HTTP interface
-    environment:
-      CLICKHOUSE_DB: logs
-      CLICKHOUSE_USER: default
-      CLICKHOUSE_PASSWORD: ""
-    volumes:
-      - clickhouse-shard-2-data:/var/lib/clickhouse
-      - ./deployment/clickhouse-config.xml:/etc/clickhouse-server/config.d/cluster.xml:ro
-    healthcheck:
-      test: ["CMD", "clickhouse-client", "--query", "SELECT 1"]
-      interval: 10s
-      timeout: 5s
-      retries: 5
-
   log-courier:
     build:
       context: .
@@ -48,19 +8,9 @@ services:
     network_mode: host
     profiles:
       - full
-    depends_on:
-      clickhouse-shard-1:
-        condition: service_healthy
-      clickhouse-shard-2:
-        condition: service_healthy
     environment:
       SUPERVISORD_CONF: supervisord.conf
     volumes:
       - ./deployment/conf/supervisord.conf:/conf/supervisord.conf:ro
       - ./deployment/conf/env:/conf/env:ro
 
-volumes:
-  clickhouse-shard-1-data:
-    driver: local
-  clickhouse-shard-2-data:
-    driver: local

env/default/values.yaml

@@ -2,16 +2,17 @@ global:
   log_level: info
 
 features:
-  scuba:
-    enabled: false
-    enable_service_user: false
-
+  access_logging:
+    enabled: true
 
 cloudserver:
-  image: ghcr.io/scality/cloudserver:7.70.77
+  image: ghcr.io/scality/cloudserver:9.2.14
 
 vault:
-  image: ghcr.io/scality/vault:7.81.0
+  image: ghcr.io/scality/vault:7.83.0
 
 s3_metadata:
-  image: ghcr.io/scality/metadata:8.19.0-standalone
+  image: ghcr.io/scality/metadata:9.0.0-standalone
+
+clickhouse:
+  image: clickhouse/clickhouse-server:24.3.2.23-alpine

pkg/logcourier/processor_test.go

@@ -423,21 +423,21 @@ var _ = Describe("Processor", func() {
 						_ = testProcessor.Run(testCtx)
 					}()
 
-					// Wait for upload attempt
+					// Wait for upload to fail
 					Eventually(func() int64 {
-						return countingUploader.GetUploadCount()
+						return countingUploader.GetFailureCount()
 					}).WithTimeout(3 * time.Second).WithPolling(100 * time.Millisecond).Should(
 						BeNumerically(">=", 1),
 					)
 
 					uploadCount := countingUploader.GetUploadCount()
+					failureCount := countingUploader.GetFailureCount()
 					cancel()
 
 					// Verify only one upload attempt was made
 					Expect(uploadCount).To(Equal(int64(1)), "Should attempt upload exactly once")
 
 					// Verify the upload failed
-					failureCount := countingUploader.GetFailureCount()
 					Expect(failureCount).To(Equal(int64(1)), "Upload should have failed")
 
 					// Verify offset was not committed
@@ -505,21 +505,21 @@ var _ = Describe("Processor", func() {
 						_ = testProcessor.Run(testCtx)
 					}()
 
-					// Wait for upload attempt
+					// Wait for upload to fail
 					Eventually(func() int64 {
-						return countingUploader.GetUploadCount()
+						return countingUploader.GetFailureCount()
 					}).WithTimeout(3 * time.Second).WithPolling(100 * time.Millisecond).Should(
 						BeNumerically(">=", 1),
 					)
 
 					uploadCount := countingUploader.GetUploadCount()
+					failureCount := countingUploader.GetFailureCount()
 					cancel()
 
 					// Verify only one upload attempt was made
 					Expect(uploadCount).To(Equal(int64(1)), "Should attempt upload exactly once")
 
 					// Verify the upload failed
-					failureCount := countingUploader.GetFailureCount()
 					Expect(failureCount).To(Equal(int64(1)), "Upload should have failed")
 
 					// Verify offset was not committed
@@ -672,21 +672,21 @@ var _ = Describe("Processor", func() {
 						_ = testProcessor.Run(testCtx)
 					}()
 
-					// Wait for upload attempt
+					// Wait for upload to fail
 					Eventually(func() int64 {
-						return countingUploader.GetUploadCount()
+						return countingUploader.GetFailureCount()
 					}).WithTimeout(3 * time.Second).WithPolling(100 * time.Millisecond).Should(
 						BeNumerically(">=", 1),
 					)
 
 					uploadCount := countingUploader.GetUploadCount()
+					failureCount := countingUploader.GetFailureCount()
 					cancel()
 
 					// Verify only one upload attempt was made
 					Expect(uploadCount).To(Equal(int64(1)), "Should attempt upload exactly once")
 
 					// Verify the upload failed
-					failureCount := countingUploader.GetFailureCount()
 					Expect(failureCount).To(Equal(int64(1)), "Upload should have failed")
 
 					// Verify offset was not committed

pkg/testutil/clickhouse.go

@@ -68,7 +68,7 @@ ORDER BY (raftSessionID, bucketName, insertedAt, timestamp, req_id)
 
 const schemaAccessLogsFederated = `
 CREATE TABLE IF NOT EXISTS %s.access_logs_federated AS %s.access_logs
-ENGINE = Distributed(test_cluster, %s, access_logs, raftSessionID)
+ENGINE = Distributed(workbench_cluster, %s, access_logs, raftSessionID)
 `
 
 const schemaOffsetsLocal = `
@@ -86,7 +86,7 @@ ORDER BY (bucketName, raftSessionID)
 
 const schemaOffsetsFederated = `
 CREATE TABLE IF NOT EXISTS %s.offsets_federated AS %s.offsets
-ENGINE = Distributed(test_cluster, %s, offsets, raftSessionID)
+ENGINE = Distributed(workbench_cluster, %s, offsets, raftSessionID)
 `
 
 // ==========================================