Detection rules › Splunk

Linux Possible Privilege Escalation via PYTHONPATH

Status
production
Severity
medium
Group by
computer_name, file_name, process_guid, process_id, target_filename, user
Author
Raven Tait, Splunk
Source
github.com/splunk/security_content

The following analytic detects the creation of a malicious shared object at a Python importlib path outside the standard system library directories, a technique used to abuse PYTHONPATH for local privilege escalation. Attackers exploiting vulnerabilities such as the 2024 NeedRestart flaw (CVE-2024-48990) plant a crafted importlib/init.so in an attacker-controlled directory, then manipulate the PYTHONPATH environment variable so that a privileged process, such as NeedRestart running as root, loads the rogue module instead of the legitimate one, achieving code execution with elevated privileges. The detection monitors for file writes matching the importlib/init.so pattern that do not originate from expected system library paths.

Known false positives

  • Legitimate Python developers or system administrators may modify PYTHONPATH for testing or application configuration purposes. Filter based on known development environments and trusted user activity.

MITRE ATT&CK coverage

Telemetry coverage

PlatformRecord / event type
LinuxEvent ID 11: File created

Rule body

name: Linux Possible Privilege Escalation via PYTHONPATH
id: 5edda575-409a-4820-a048-5780796a181e
version: 1
creation_date: '2026-07-08'
modification_date: '2026-07-08'
author: Raven Tait, Splunk
status: production
type: TTP
description: |-
    The following analytic detects the creation of a malicious shared object at a Python importlib path outside the standard system library directories, a technique used to abuse PYTHONPATH for local privilege escalation.
    Attackers exploiting vulnerabilities such as the 2024 NeedRestart flaw (CVE-2024-48990) plant a crafted importlib/__init__.so in an attacker-controlled directory, then manipulate the PYTHONPATH environment variable so that a privileged process, such as NeedRestart running as root, loads the rogue module instead of the legitimate one, achieving code execution with elevated privileges.
    The detection monitors for file writes matching the importlib/__init__.so pattern that do not originate from expected system library paths.
data_source:
    - Sysmon for Linux EventID 11
search: |-
    | tstats `security_content_summariesonly`
      count min(_time) as firstTime
            max(_time) as lastTime
    
    from datamodel=Endpoint.Filesystem where
    
    Filesystem.file_path="*/importlib/__init__.so*"
    NOT Filesystem.file_path IN (
        "/lib/*",
        "/lib64/*",
        "/usr/lib/*",
        "/usr/lib64/*",
        "/usr/local/lib/*",
        "/usr/local/lib64/*"
    )
    
    by Filesystem.file_path Filesystem.file_name Filesystem.user Filesystem.dest
       Filesystem.process_guid Filesystem.process_id
    
    | `drop_dm_object_name(Filesystem)`
    | `security_content_ctime(firstTime)`
    | `security_content_ctime(lastTime)`
    | `linux_possible_privilege_escalation_via_pythonpath_filter`
how_to_implement: |-
    The detection is based on data that originates from Endpoint Detection and Response (EDR) agents. These agents are designed to provide security-related telemetry from the endpoints where the agent is installed. To implement this search, you must ingest logs that contain file creation events including the file path and user context. These logs must be processed using the appropriate Splunk Technology Add-ons that are specific to the EDR product. The logs must also be mapped to the `Filesystem` node of the `Endpoint` data model. Use the Splunk Common Information Model (CIM) to normalize the field names and speed up the data modeling process.
known_false_positives: |-
    Legitimate Python developers or system administrators may modify PYTHONPATH for testing or application configuration purposes. Filter based on known development environments and trusted user activity.
references:
    - https://www.bleepingcomputer.com/news/security/ubuntu-linux-impacted-by-decade-old-needrestart-flaw-that-gives-root/
    - https://www.qualys.com/2024/11/19/needrestart/needrestart.txt
finding:
    title: Possible Privilege Escalation via PYTHONPATH via [$file_name$] on [$dest$]
    entity:
        field: dest
        type: system
        score: 50
threat_objects:
    - field: file_name
      type: file_name
analytic_story:
    - Linux Privilege Escalation
    - Linux Post-Exploitation
    - Linux Persistence Techniques
asset_type: Endpoint
cve:
    - CVE-2024-48990
mitre_attack_id:
    - T1068
    - T1574.007
product:
    - Splunk Enterprise
    - Splunk Enterprise Security
    - Splunk Cloud
category: endpoint
security_domain: endpoint

Stages and Predicates

Stage 1: tstats

| tstats `security_content_summariesonly`
  count min(_time) as firstTime
        max(_time) as lastTime

from datamodel=Endpoint.Filesystem where

Filesystem.file_path="*/importlib/__init__.so*"
NOT Filesystem.file_path IN (
    "/lib/*",
    "/lib64/*",
    "/usr/lib/*",
    "/usr/lib64/*",
    "/usr/local/lib/*",
    "/usr/local/lib64/*"
)

by Filesystem.file_path Filesystem.file_name Filesystem.user Filesystem.dest
   Filesystem.process_guid Filesystem.process_id

Stage 2: search

| `drop_dm_object_name(Filesystem)`

Stage 3: search

| `security_content_ctime(firstTime)`

Stage 4: search

| `security_content_ctime(lastTime)`

Stage 5: search

| `linux_possible_privilege_escalation_via_pythonpath_filter`

Exclusions

The rule actively suppresses these predicates.

FieldKindExcluded valuesSearch
Filesystem.file_pathin"/lib/*", "/lib64/*", "/usr/lib/*", "/usr/lib64/*", "/usr/local/lib/*", "/usr/local/lib64/*"excludes:Filesystem.file_path

Indicators

These rows show field, operator, and value matches.

FieldKindValuesSearch
Filesystem.file_patheq
  • "*/importlib/__init__.so*"
field:"TargetFilename" kind:eq