Last updated:

Receive and Customize Auditing Logs

KubeSphere Auditing Logs provide a security-relevant chronological set of records documenting the sequence of activities that have affected the system by individual users, administrators, or other components of the system. Each request to KubeSphere generates an event that is then written to a webhook and processed according to a certain rule. The event will be ignored, stored, or generate an alert based on different rules.

Enable KubeSphere Auditing Logs

To enable auditing logs, see KubeSphere Auditing Logs.

Receive Auditing Logs from KubeSphere

The KubeSphere Auditing Log system receives auditing logs only from KubeSphere by default, while it can also receive auditing logs from Kubernetes.

Users can stop receiving auditing logs from KubeSphere by changing the value of auditing.enable in ConfigMap kubesphere-config in the namespace kubesphere-system using the following command:

kubectl edit cm -n kubesphere-system kubesphere-config

Change the value of auditing.enabled as false to stop receiving auditing logs from KubeSphere.

  spec:
    auditing:
      enabled: false

You need to restart the KubeSphere apiserver to make the changes effective.

Receive Auditing Logs from Kubernetes

To make the KubeSphere Auditing Log system receive auditing logs from Kubernetes, you need to add a Kubernetes audit policy file and Kubernetes audit webhook config file to /etc/kubernetes/manifests/kube-apiserver.yaml as follows.

Audit policy

apiVersion: v1
kind: Pod
metadata:
  name: kube-apiserver
  namespace: kube-system
spec:
  containers:
  - command:
    - kube-apiserver
    - --audit-policy-file=/etc/kubernetes/audit/audit-policy.yaml
    - --audit-webhook-config-file=/etc/kubernetes/audit/audit-webhook.yaml
    volumeMounts:
    - mountPath: /etc/kubernetes/audit
      name: k8s-audit
      readOnly: true
  volumes:
  - hostPath:
      path: /etc/kubernetes/audit
      type: DirectoryOrCreate
    name: k8s-audit

Note

This operation will restart the Kubernetes apiserver.

The file audit-policy.yaml defines rules about what events should be recorded and what data they should include. You can use a minimal audit policy file to log all requests at the Metadata level:

# Log all requests at the Metadata level.
apiVersion: audit.k8s.io/v1
kind: Policy
rules:
- level: Metadata

For more information about the audit policy, see Audit Policy.

Audit webhook

The file audit-webhook.yaml defines the webhook which the Kubernetes auditing logs will be sent to. Here is an example configuration of the Kube-Auditing webhook.

apiVersion: v1
kind: Config
clusters:
- name: kube-auditing
  cluster:
    server: https://{ip}:6443/audit/webhook/event
    insecure-skip-tls-verify: true
contexts:
- context:
    cluster: kube-auditing
    user: ""
  name: default-context
current-context: default-context
preferences: {}
users: []

The ip is the CLUSTER-IP of Service kube-auditing-webhook-svc in the namespace kubesphere-logging-system. You can get it using this command.

kubectl get svc -n kubesphere-logging-system

Note

You need to restart the Kubernetes apiserver to make the changes effective after you modified these two files.

Edit the CRD Webhook kube-auditing-webhook, and change the value of k8sAuditingEnabled to true through the following commands.

kubectl edit webhooks.auditing.kubesphere.io kube-auditing-webhook

spec:
  auditing:
    k8sAuditingEnabled: true

Tip

You can also use a user of platform-admin role to log in to the console, search Webhook in CRDs on the Cluster Management page, and edit kube-auditing-webhook directly.

To stop receiving auditing logs from Kubernetes, remove the configuration of auditing webhook backend, then change the value of k8sAuditingEnabled to false.

Customize Auditing Logs

KubeSphere Auditing Log system provides a CRD Webhook kube-auditing-webhook to customize auditing logs. Here is an example yaml file:

apiVersion: auditing.kubesphere.io/v1alpha1
kind: Webhook
metadata:
  name: kube-auditing-webhook
spec:
  auditLevel: RequestResponse
  auditSinkPolicy:
    alertingRuleSelector:
      matchLabels:
        type: alerting
    archivingRuleSelector:
      matchLabels: 
        type: persistence
  image: kubesphere/kube-auditing-webhook:v0.1.0
  archivingPriority: DEBUG
  alertingPriority: WARNING
  replicas: 2
  receivers:
    - name: alert
      type: alertmanager
      config:
        service:
          namespace: kubesphere-monitoring-system
          name: alertmanager-main
          port: 9093

Parameter	Description	Default
`replicas`	The replica number of the Kube-Auditing webhook.	2
`archivingPriority`	The priority of the archiving rule. The known audit types are `DEBUG`, `INFO`, and `WARNING`.	`DEBUG`
`alertingPriority`	The priority of the alerting rule. The known audit types are `DEBUG`, `INFO`, and `WARNING`.	`WARNING`
`auditLevel`	The level of auditing logs. The known levels are: - `None`: don't log events. - `Metadata`: log request metadata (requesting user, timestamp, resource, verb, etc.) but not requests or response bodies. - `Request`: log event metadata and request bodies but no response body. This does not apply to non-resource requests. - `RequestResponse`: log event metadata, requests, and response bodies. This does not apply to non-resource requests.	`Metadata`
`k8sAuditingEnabled`	Whether to receive Kubernetes auditing logs.	`false`
`receivers`	The receivers to receive alerts.

Note

You can change the level of Kubernetes auditing logs by modifying the file audit-policy.yaml, then restart the Kubernetes apiserver.

Feedback

Was this page Helpful?

Receive the latest news, articles and updates from KubeSphere

Thanks for the feedback. If you have a specific question about how to use KubeSphere, ask it on Slack. Open an issue in the GitHub repo if you want to report a problem or suggest an improvement.

Previous : Event Query Next : Auditing Rules

What’s on this Page