GKE Forbidden Request from Unusual User Agent

edit
IMPORTANT: This documentation is no longer updated. Refer to Elastic's version policy and the latest documentation.

GKE Forbidden Request from Unusual User Agent

edit

Detects the first occurrence of a failed GKE API request from a previously unseen user agent. Adversary tooling often uses non-standard clients; combined with authorization failures this can indicate RBAC probing or exploitation attempts.

Rule type: new_terms

Rule indices:

  • logs-gcp.audit-*

Severity: medium

Risk score: 47

Runs every: 5m

Searches indices from: now-6m (Date Math format, see also Additional look-back time)

Maximum alerts per execution: 100

References:

Tags:

  • Domain: Cloud
  • Domain: Kubernetes
  • Data Source: GCP
  • Data Source: Google Cloud Platform
  • Use Case: Threat Detection
  • Tactic: Execution
  • Tactic: Discovery
  • Resources: Investigation Guide

Version: 1

Rule authors:

  • Elastic

Rule license: Elastic License v2

Investigation guide

edit

Triage and analysis

Investigating GKE Forbidden Request from Unusual User Agent

A novel user agent with failed API calls may be scanner or post-compromise tooling probing RBAC.

Investigation steps

  • Review user_agent.original, client.user.email, event.action, gcp.audit.resource_name, and source.ip.
  • Determine whether the client is expected (new SDK, CI image) or external reconnaissance.
  • Hunt for successful requests from the same UA or source after the failures.

False positives

  • First use of a legitimate new client library; add a scoped UA exception after validation.

Setup

edit

The GCP Fleet integration with GKE audit logs enabled is required to be compatible with this rule.

Rule query

edit
data_stream.dataset:gcp.audit and service.name:k8s.io and event.outcome:failure and
user_agent.original:(* and not (*kubernetes/$Format or kube-probe* or gke-exec-auth-plugin*)) and
not client.user.email:(
  "system:addon-manager" or system\:*controller* or system\:gke-* or
  "system:apiserver" or "system:kube-scheduler" or "system:metrics-server-nanny" or
  "system:kube-proxy" or "system:clustermetrics" or "system:vpa-recommender" or
  "system:cluster-autoscaler" or "system:kubestore-collector" or
  "system:konnectivity-server" or
  "system:serviceaccount:kube-system:pod-garbage-collector" or
  "system:serviceaccount:kube-system:generic-garbage-collector" or system\:node\:* or
  "gcp:kube-bootstrap" or *container-engine-robot*
)

Framework: MITRE ATT&CKTM