Configure Audit Logging

To enable audit logging, configure audit log settings in the globalVariables section of your cluster custom resource (CR).

Add the audit log configuration to your sdb-cluster.yaml:

apiVersion: memsql.com/v1alpha1
kind: MemsqlCluster
metadata:
 name: sdb-cluster
spec:
 license: <license_key>
 adminHashedPassword: "<hashed_password>"

 globalVariables:
   auditlog_level: ADMIN-ONLY
   auditlog_disk_sync: OFF
   auditlog_rotation_size: 134217728
   auditlog_rotation_time: 3600

 nodeImage:
   repository: singlestore/node
   tag: alma-8.7.10-28804d3b1b

 redundancyLevel: 2

 aggregatorSpec:
   count: 2
   cores: 8
   memoryMB: 32768
   storageGB: 256

 leafSpec:
   count: 2
   cores: 8
   memoryMB: 32768
   storageGB: 512

Apply the configuration:

kubectl apply -f sdb-cluster.yaml

You can configure the following audit log variables:

SHOW GLOBAL VARIABLES LIKE 'audit%';

Connect to your cluster and verify the settings:

You can collect audit logs from your Kubernetes cluster using a Kubernetes Job.

To automate report collection and upload to your storage backend, create a Kubernetes Job. This approach works with on-premises storage (NFS, local persistent volumes) and object storage systems (MinIO, S3-compatible storage).

• Create storage credentials secret (one-time setup)

  For S3-compatible storage:

  Shell

  Copy

  kubectl create secret generic aws-credentials \
   --from-literal=access-key-id=YOUR_ACCESS_KEY \
   --from-literal=secret-access-key=YOUR_SECRET_KEY

  For other storage backends, create appropriate secrets for your authentication method.

• Create cluster-collection-job.yaml

  YAML

  Copy

  apiVersion: batch/v1
  kind: Job
  metadata:
    name: singlestore-report-collection
  spec:
    template:
      spec:
        serviceAccountName: tools
        containers:
        - name: report-collector
          image: singlestore/tools:latest
          command: ["/bin/bash", "-c"]
          args:
          - |
            # Collect the cluster report
            sdb-report collect-kube --cluster-name sdb-cluster --namespace default --output-path /tmp/report
            REPORT_FILE=$(ls -t /tmp/report/*.tar.gz | head -1)
  
            # Upload to S3-compatible object storage (e.g., MinIO)
            aws s3 cp $REPORT_FILE s3://${BUCKET_NAME}/cluster-reports/ --endpoint-url ${S3_ENDPOINT}
  
            # On-premises storage options:
            # - NFS: cp $REPORT_FILE /mnt/nfs/cluster-reports/
            # - Local PV: cp $REPORT_FILE /mnt/storage/cluster-reports/
  
          env:
          # S3-compatible storage configuration (e.g., MinIO)
          - name: AWS_ACCESS_KEY_ID
            valueFrom:
              secretKeyRef:
                name: storage-credentials
                key: access-key-id
          - name: AWS_SECRET_ACCESS_KEY
            valueFrom:
              secretKeyRef:
                name: storage-credentials
                key: secret-access-key
          - name: BUCKET_NAME
            value: "your-bucket-name"
          - name: S3_ENDPOINT
            value: "http://minio:9000"
  
          # Optional: mount on-premises storage
          # volumeMounts:
          # - name: nfs-storage
          #   mountPath: /mnt/nfs
          # - name: local-storage
          #   mountPath: /mnt/storage
  
        restartPolicy: Never
  
        # Optional: define on-premises volumes
        # volumes:
        # - name: nfs-storage
        #   nfs:
        #     server: your-nfs-server
        #     path: /path/to/storage
        # - name: local-storage
        #   hostPath:
        #     path: /path/to/local/storage
        #     type: Directory
  
    backoffLimit: 3

• Run the job

  Shell

  Copy

  kubectl apply -f cluster-collection-job.yaml

• Check progress and view logs

  Shell

  Copy

  # Check job status
  kubectl get jobs
  
  # View logs
  kubectl logs job/singlestore-report-collection
  
  # Verify upload to external storage
  
  # For S3-compatible object storage (for example, MinIO):
  aws s3 ls s3://<your-bucket>/cluster-reports/ --endpoint-url <your-endpoint>
  
  # For NFS or local persistent storage (from a mounted node or pod):
  ls /mnt/nfs/cluster-reports/

• Clean up (optional)

  Shell

  Copy

  kubectl delete job singlestore-report-collection