handle empty agent config in Go client and fix build failures

.pipelines/azure_pipeline_mergedbranches.yaml

@@ -78,7 +78,7 @@ jobs:
 
       cd $(Build.SourcesDirectory)/deployment/arc-k8s-extension/ServiceGroupRoot/Scripts
       tar -czvf ../artifacts.tar.gz ../../../../charts/azuremonitor-containers/ pushChartToAcr.sh
-    
+
       cd $(Build.SourcesDirectory)/deployment/arc-k8s-extension-release-v2/ServiceGroupRoot/Scripts
       tar -czvf ../artifacts.tar.gz arcExtensionRelease.sh
 
@@ -95,7 +95,7 @@ jobs:
 
   - task: CredScan@3
     displayName: "SDL : Run credscan"
-  
+
   - task: CopyFiles@2
     displayName: "Copy ev2 deployment artifacts"
     inputs:
@@ -140,7 +140,7 @@ jobs:
     inputs:
       pathToPublish: '$(Build.ArtifactStagingDirectory)'
       artifactName: drop
-  
+
   - task: Armory@2
     displayName: 'Run ARMory'
     inputs:
@@ -175,6 +175,7 @@ jobs:
 
         sudo apt-get update && sudo apt-get -y install qemu binfmt-support qemu-user-static
         docker system prune --all -f
+        docker images -q --filter "dangling=true" | xargs docker rmi
         docker run --rm --privileged multiarch/qemu-user-static --reset -p yes
 
         docker buildx create --name testbuilder
@@ -185,6 +186,11 @@ jobs:
         az account set -s ${{ variables.subscription }}
         az acr login -n ${{ variables.containerRegistry }}
 
+        # NOTE: Using the prometheus-collector team's cached buildx image since moby/buildkit:buildx-stable-1 getting throttled
+        docker pull mcr.microsoft.com/azuremonitor/containerinsights/cidev/prometheus-collector/images:buildx-stable-1
+        docker buildx create --name dockerbuilder --driver docker-container --driver-opt image=mcr.microsoft.com/azuremonitor/containerinsights/cidev/prometheus-collector/images:buildx-stable-1 --use
+        docker buildx inspect --bootstrap
+
         if [ "$(Build.Reason)" != "PullRequest" ]; then
           docker buildx build --platform $(BUILD_PLATFORMS) --tag ${{ variables.repoImageName }}:$(linuxImagetag) -f kubernetes/linux/Dockerfile.multiarch --metadata-file $(Build.ArtifactStagingDirectory)/linux/metadata.json --build-arg IMAGE_TAG=$(linuxTelemetryTag) --build-arg GOLANG_BASE_IMAGE=$(GOLANG_BASE_IMAGE) --build-arg CI_BASE_IMAGE=$(CI_BASE_IMAGE) --push --provenance=false .
 
@@ -274,8 +280,54 @@ jobs:
       scriptLocation: inlineScript
       inlineScript: |
         curl -sfL https://raw.githubusercontent.com/aquasecurity/trivy/main/contrib/install.sh | sh -s -- -b /usr/local/bin
+        PRIMARY_TRIVY_DB_REPOSITORY="ghcr.io/aquasecurity/trivy-db"
+        SECONDARY_TRIVY_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-db"
+        PRIMARY_TRIVY_JAVA_DB_REPOSITORY="ghcr.io/aquasecurity/trivy-java-db"
+        SECONDARY_TRIVY_JAVA_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-java-db"
+
+        # Set initial repositories to primary
+        export TRIVY_DB_REPOSITORY=$PRIMARY_TRIVY_DB_REPOSITORY
+        export TRIVY_JAVA_DB_REPOSITORY=$PRIMARY_TRIVY_JAVA_DB_REPOSITORY
+
+        # Function to run Trivy scan and handle output
+        run_trivy_scan() {
+            trivy image --ignore-unfixed --no-progress --severity HIGH,CRITICAL,MEDIUM "${{ variables.repoImageName }}:$(linuxImagetag)" > trivy_output.log 2>&1
+            return $?
+        }
 
-        trivy image --ignore-unfixed --no-progress --severity HIGH,CRITICAL,MEDIUM --exit-code 1 ${{ variables.repoImageName }}:$(linuxImagetag)
+        # Attempt scan up to 5 times with repository fallback
+        for i in {1..5}; do
+            echo "Running Trivy scan attempt $i..."
+
+            # Run the Trivy scan and capture exit code
+            run_trivy_scan
+            TRIVY_EXIT_CODE=$?
+
+            # Check if scan was successful
+            if [ $TRIVY_EXIT_CODE -eq 0 ]; then
+                echo "Trivy scan succeeded."
+                cat trivy_output.log
+                break
+            fi
+
+            # If the first attempt fails, switch to secondary repositories
+            if [ $i -eq 1 ]; then
+                echo "Primary repositories failed with an error. Switching to secondary repositories."
+                export TRIVY_DB_REPOSITORY=$SECONDARY_TRIVY_DB_REPOSITORY
+                export TRIVY_JAVA_DB_REPOSITORY=$SECONDARY_TRIVY_JAVA_DB_REPOSITORY
+            fi
+
+            # Log and wait before retrying if an error occurred
+            echo "Error: Trivy scan attempt $i failed. Retrying... ($i/5)"
+            cat trivy_output.log
+            sleep 5 # Wait 5 seconds before retrying
+        done
+
+        # Final check: if still failing after 5 attempts, exit with error
+        if [ $TRIVY_EXIT_CODE -ne 0 ]; then
+            echo "Error: Trivy scan failed after 5 retries."
+            exit 1
+        fi
 
     # Find in cache or download a specific version of Go and add it to the PATH.
   - task: GoTool@0
@@ -287,7 +339,7 @@ jobs:
         ls
         make
     displayName: 'Execute Makefile for Linux Build'
-  
+
   - task: CodeQL3000Finalize@0
     condition: eq(variables.IS_MAIN_BRANCH, true)
 
@@ -355,6 +407,7 @@ jobs:
       azureSubscription: ${{ variables.armServiceConnectionName }}
       scriptType: ps
       scriptLocation: inlineScript
+      retryCountOnTaskFailure: 2
       inlineScript: |
         mkdir -p $(Build.ArtifactStagingDirectory)/windows
         cd kubernetes/windows
@@ -398,7 +451,7 @@ jobs:
 
         echo "Extract fluent-bit"
         docker cp signingContainer:C:\opt\fluent-bit .
-        
+
         echo "Extract Ruby"
         docker cp signingContainer:C:\ruby31 .
 
@@ -525,7 +578,7 @@ jobs:
       targetType: 'inline'
       script: |
         docker create --name pushContainer ${{ variables.repoImageName }}:$(windowsImageTag)-$(windows2019BaseImageVersion)-unsigned
-        
+
         echo "Copy Signed binaries/folders back to docker image..."
         docker cp $(Build.ArtifactStagingDirectory)/fpSigning/CertificateGenerator.exe pushContainer:C:\opt\amalogswindows\certgenerator\CertificateGenerator.exe
         docker cp $(Build.ArtifactStagingDirectory)/fpSigning/CertificateGenerator.dll pushContainer:C:\opt\amalogswindows\certgenerator\CertificateGenerator.dll
@@ -588,6 +641,7 @@ jobs:
       azureSubscription: ${{ variables.armServiceConnectionName }}
       scriptType: ps
       scriptLocation: inlineScript
+      retryCountOnTaskFailure: 2
       inlineScript: |
         mkdir -p $(Build.ArtifactStagingDirectory)/windows
         cd kubernetes/windows
@@ -631,7 +685,7 @@ jobs:
 
         echo "Extract fluent-bit"
         docker cp signingContainer:C:\opt\fluent-bit .
-        
+
         echo "Extract Ruby"
         docker cp signingContainer:C:\ruby31 .
 
@@ -758,7 +812,7 @@ jobs:
       targetType: 'inline'
       script: |
         docker create --name pushContainer ${{ variables.repoImageName }}:$(windowsImageTag)-$(windows2022BaseImageVersion)-unsigned
-        
+
         echo "Copy Signed binaries/folders back to docker image..."
         docker cp $(Build.ArtifactStagingDirectory)/fpSigning/CertificateGenerator.exe pushContainer:C:\opt\amalogswindows\certgenerator\CertificateGenerator.exe
         docker cp $(Build.ArtifactStagingDirectory)/fpSigning/CertificateGenerator.dll pushContainer:C:\opt\amalogswindows\certgenerator\CertificateGenerator.dll
@@ -907,7 +961,7 @@ jobs:
     inputs:
       pathToPublish: '$(Build.ArtifactStagingDirectory)'
       artifactName: drop
-  
+
   - task: AntiMalware@4
     displayName: 'Run MpCmdRun.exe'
     inputs:

kubernetes/linux/Dockerfile.multiarch

@@ -141,6 +141,29 @@ RUN ln -s /lib/libnssckbi.so /lib/p11-kit-trust.so
 FROM distroless_image AS vulnscan
 COPY .trivyignore .trivyignore
 RUN ["/bin/bash", "-c", "curl -sfL https://raw.githubusercontent.com/aquasecurity/trivy/main/contrib/install.sh | sh -s -- -b /usr/local/bin v0.39.0"]
+
+# Set up primary and secondary repository URLs
+ENV PRIMARY_TRIVY_DB_REPOSITORY="ghcr.io/aquasecurity/trivy-db"
+ENV SECONDARY_TRIVY_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-db"
+ENV PRIMARY_TRIVY_JAVA_DB_REPOSITORY="ghcr.io/aquasecurity/trivy-java-db"
+ENV SECONDARY_TRIVY_JAVA_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-java-db"
+
+# Download Trivy main database with a fallback mechanism
+RUN export TRIVY_DB_REPOSITORY=$PRIMARY_TRIVY_DB_REPOSITORY && \
+    trivy image --download-db-only || \
+    (echo "Primary TRIVY_DB_REPOSITORY failed, trying secondary." && \
+     export TRIVY_DB_REPOSITORY=$SECONDARY_TRIVY_DB_REPOSITORY && \
+     trivy image --download-db-only) || \
+    (echo "Both TRIVY_DB_REPOSITORY sources failed." && exit 1)
+
+# Download Trivy Java database with a fallback mechanism
+RUN export TRIVY_JAVA_DB_REPOSITORY=$PRIMARY_TRIVY_JAVA_DB_REPOSITORY && \
+    trivy fs --scanners vuln --vuln-type library --download-java-db-only || \
+    (echo "Primary TRIVY_JAVA_DB_REPOSITORY failed, trying secondary." && \
+     export TRIVY_JAVA_DB_REPOSITORY=$SECONDARY_TRIVY_JAVA_DB_REPOSITORY && \
+     trivy fs --scanners vuln --vuln-type library --download-java-db-only) || \
+    (echo "Both TRIVY_JAVA_DB_REPOSITORY sources failed." && exit 1)
+
 RUN ["/bin/bash", "-c", "trivy rootfs --ignore-unfixed --no-progress --severity HIGH,CRITICAL,MEDIUM --skip-files \"/usr/local/bin/trivy\" /"]
 RUN ["/bin/bash", "-c", "trivy rootfs --ignore-unfixed --no-progress --severity HIGH,CRITICAL,MEDIUM /usr/lib"]
 RUN ["/bin/bash", "-c", "trivy rootfs --exit-code 1 --ignore-unfixed --no-progress --severity HIGH,CRITICAL,MEDIUM --skip-files \"/usr/local/bin/trivy\" / > /dev/null 2>&1 && trivy rootfs --exit-code 1 --ignore-unfixed --no-progress --severity HIGH,CRITICAL,MEDIUM /usr/lib > /dev/null 2>&1"]

source/plugins/go/src/ingestion_token_utils.go

@@ -444,17 +444,35 @@ func getAgentConfiguration(imdsAccessToken string) (configurationId string, chan
 		return configurationId, channelId, err
 	}
 
-	if len(agentConfiguration.Configurations[0].Content.Channels) == 0 {
-		message := "getAgentConfiguration: Received empty agentConfiguration.Configurations[0].Content.Channels"
-		Log(message)
-		SendException(message)
-		return configurationId, channelId, err
+	for _, config := range agentConfiguration.Configurations {
+		if len(config.Content.Channels) == 0 {
+			// this is expected because AMCS will return agent config based on OS Type. For example, syslog is not supported on windows hence config will not have channels and data sources
+			message := "getAgentConfiguration: Received empty config.Content.Channels"
+			Log(message)
+			continue
+		}
+
+		configurationId = config.Configurationid
+		for _, channel := range config.Content.Channels {
+			if channel.ID != "" {
+				channelId = channel.ID
+				break
+			}
+		}
+
+		if !ContainerLogV2ConfigMap && len(config.Content.Extensionconfigurations.Containerinsights) > 0 {
+			for _, ciExtensionInstance := range config.Content.Extensionconfigurations.Containerinsights {
+				ContainerLogSchemaV2 = ciExtensionInstance.Extensionsettings.DataCollectionSettings.EnableContainerLogV2
+			}
+		}
+		break
 	}
 
-	configurationId = agentConfiguration.Configurations[0].Configurationid
-	channelId = agentConfiguration.Configurations[0].Content.Channels[0].ID
-	if !ContainerLogV2ConfigMap && len(agentConfiguration.Configurations[0].Content.Extensionconfigurations.Containerinsights) > 0 {
-		ContainerLogSchemaV2 = agentConfiguration.Configurations[0].Content.Extensionconfigurations.Containerinsights[0].Extensionsettings.DataCollectionSettings.EnableContainerLogV2
+	if configurationId == "" || channelId == "" {
+		message := "getAgentConfiguration: Failed to obtain configurationId or channelId"
+		Log(message)
+		SendException(message)
+		return configurationId, channelId, errors.New(message)
 	}
 	Log("getAgentConfiguration: obtained configurationId: %s, channelId: %s", configurationId, channelId)
 	Log("Info getAgentConfiguration: end")