Detection rules › Elastic

GKE Ephemeral Container Added to Pod

Status
production
Severity
medium
Time window
6m
Author
Elastic
Source
github.com/elastic/detection-rules

Detects allowed updates or patches to the pods/ephemeralcontainers subresource on GKE by a non-system identity. Ephemeral containers are commonly used for debugging (kubectl debug) but can also be abused to inject tooling into a running pod, access mounted secrets, and execute commands in the target pod context. Attackers with sufficient RBAC may use ephemeral containers to escalate privileges, move laterally, or establish persistence without deploying a new workload.

Known false positives

  • Cluster operators or SREs may legitimately use ephemeral containers for debugging production workloads. Baseline approved admin identities and tune exclusions for known automation.

MITRE ATT&CK coverage

TacticTechniques
Execution
Privilege Escalation

Telemetry coverage

Rules detecting the same action

These rules filter on the same operation.

Rule body

[metadata]
creation_date = "2026/07/13"
integration = ["gcp"]
maturity = "production"
updated_date = "2026/07/13"

[rule]
author = ["Elastic"]
description = """
Detects allowed updates or patches to the pods/ephemeralcontainers subresource on GKE by a non-system identity.
Ephemeral containers are commonly used for debugging (kubectl debug) but can also be abused to inject tooling into a
running pod, access mounted secrets, and execute commands in the target pod context. Attackers with sufficient RBAC may
use ephemeral containers to escalate privileges, move laterally, or establish persistence without deploying a new
workload.
"""
false_positives = [
    """
    Cluster operators or SREs may legitimately use ephemeral containers for debugging production workloads. Baseline
    approved admin identities and tune exclusions for known automation.
    """,
]
from = "now-6m"
index = ["logs-gcp.audit-*"]
language = "kuery"
license = "Elastic License v2"
name = "GKE Ephemeral Container Added to Pod"
note = """## Triage and analysis

### Investigating GKE Ephemeral Container Added to Pod

Ephemeral containers allow adding a container to an existing pod for troubleshooting. When abused, they can gain
interactive access to a workload, read sensitive files, and run tools that were not present in the original image.

### Possible investigation steps

- Review `client.user.email`, `source.ip`, and `user_agent.original` and confirm the identity is authorized to use
  ephemeral containers.
- Inspect `gcp.audit.resource_name` to identify the targeted pod and owning workload.
- If request bodies are captured, review the ephemeral container image, command, and securityContext for privilege
  indicators.
- Correlate with follow-on audit activity such as pod exec, secret reads, TokenRequest, or RBAC modifications.

### False positive analysis

- Approved on-call debugging with `kubectl debug` may match. Allowlist known admin identities after review.

### Response and remediation

- If unauthorized, remove excessive RBAC that grants update or patch on pods/ephemeralcontainers and rotate exposed
  credentials.
- Quarantine or redeploy impacted workloads and hunt for additional compromised pods or identities.

"""
setup = """
The GCP Fleet integration with GKE audit logs enabled is required to be compatible with this rule.
"""
references = [
    "https://kubernetes.io/docs/concepts/workloads/pods/ephemeral-containers/",
]
risk_score = 47
rule_id = "d52e2418-978c-4184-9eab-e084af1f76b7"
severity = "medium"
tags = [
    "Domain: Cloud",
    "Domain: Kubernetes",
    "Data Source: GCP",
    "Data Source: Google Cloud Platform",
    "Use Case: Threat Detection",
    "Tactic: Privilege Escalation",
    "Tactic: Execution",
    "Resources: Investigation Guide",
]
timestamp_override = "event.ingested"
type = "query"

query = '''
data_stream.dataset:gcp.audit and service.name:"k8s.io" and event.outcome:success and
event.action:(
  "io.k8s.core.v1.pods.ephemeralcontainers.update" or
  "io.k8s.core.v1.pods.ephemeralcontainers.patch"
) and not client.user.email:(
  system\:node\:* or
  system\:serviceaccount\:kube-system\:*
)
'''

[[rule.threat]]
framework = "MITRE ATT&CK"

[[rule.threat.technique]]
id = "T1611"
name = "Escape to Host"
reference = "https://attack.mitre.org/techniques/T1611/"

[rule.threat.tactic]
id = "TA0004"
name = "Privilege Escalation"
reference = "https://attack.mitre.org/tactics/TA0004/"

[[rule.threat]]
framework = "MITRE ATT&CK"

[[rule.threat.technique]]
id = "T1609"
name = "Container Administration Command"
reference = "https://attack.mitre.org/techniques/T1609/"

[rule.threat.tactic]
id = "TA0002"
name = "Execution"
reference = "https://attack.mitre.org/tactics/TA0002/"

[rule.investigation_fields]
field_names = [
    "@timestamp",
    "client.user.email",
    "source.ip",
    "user_agent.original",
    "event.action",
    "event.outcome",
    "gcp.audit.resource_name",
    "gcp.audit.request",
    "data_stream.namespace",
]

Stages and Predicates

Stage 1: query

data_stream.dataset:gcp.audit and service.name:"k8s.io" and event.outcome:success and
event.action:(
  "io.k8s.core.v1.pods.ephemeralcontainers.update" or
  "io.k8s.core.v1.pods.ephemeralcontainers.patch"
) and not client.user.email:(
  system\:node\:* or
  system\:serviceaccount\:kube-system\:*
)

Exclusions

The rule actively suppresses these predicates.

FieldKindExcluded valuesSearch
client.user.emailstarts_withsystem:node:excludes:client.user.email field:"client.user.email" value:"system:node:"
client.user.emailstarts_withsystem:serviceaccount:kube-system:excludes:client.user.email field:"client.user.email" value:"system:serviceaccount:kube-system:"

Indicators

These rows show field, operator, and value matches.