:warning: THIS IS A EXPERIMENTAL DETECTION

This detection has been marked experimental by the Splunk Threat Research team. This means we have not been able to test, simulate, or build datasets for this detection. Use at your own risk. This analytic is NOT supported.

Try in Splunk Security Cloud

Description

The following analytic detects long URLs containing multiple SQL commands, indicating a potential SQL injection attack. This detection leverages web traffic data, specifically targeting web server destinations with URLs longer than 1024 characters or HTTP user agents longer than 200 characters. SQL injection is significant as it allows attackers to manipulate a web application's database, potentially leading to unauthorized data access or modification. If confirmed malicious, this activity could result in data breaches, unauthorized access, and complete system compromise. Immediate investigation and validation of alerts are crucial to mitigate these risks.

  • Type: TTP
  • Product: Splunk Enterprise, Splunk Enterprise Security, Splunk Cloud
  • Datamodel: Web
  • Last Updated: 2024-05-12
  • Author: Bhavin Patel, Splunk
  • ID: e0aad4cf-0790-423b-8328-7564d0d938f9

Annotations

ATT&CK

ATT&CK

ID Technique Tactic
T1190 Exploit Public-Facing Application Initial Access
Kill Chain Phase
  • Delivery
NIST
  • DE.CM
CIS20
  • CIS 13
CVE 


1
2
3
4
5
6
7



| tstats `security_content_summariesonly` count from datamodel=Web where Web.dest_category=web_server AND (Web.url_length > 1024 OR Web.http_user_agent_length > 200) by Web.src Web.dest Web.url Web.url_length Web.http_user_agent 
| `drop_dm_object_name("Web")` 
| eval url=lower(url) 
| eval num_sql_cmds=mvcount(split(url, "alter%20table")) + mvcount(split(url, "between")) + mvcount(split(url, "create%20table")) + mvcount(split(url, "create%20database")) + mvcount(split(url, "create%20index")) + mvcount(split(url, "create%20view")) + mvcount(split(url, "delete")) + mvcount(split(url, "drop%20database")) + mvcount(split(url, "drop%20index")) + mvcount(split(url, "drop%20table")) + mvcount(split(url, "exists")) + mvcount(split(url, "exec")) + mvcount(split(url, "group%20by")) + mvcount(split(url, "having")) + mvcount(split(url, "insert%20into")) + mvcount(split(url, "inner%20join")) + mvcount(split(url, "left%20join")) + mvcount(split(url, "right%20join")) + mvcount(split(url, "full%20join")) + mvcount(split(url, "select")) + mvcount(split(url, "distinct")) + mvcount(split(url, "select%20top")) + mvcount(split(url, "union")) + mvcount(split(url, "xp_cmdshell")) - 24 
| where num_sql_cmds > 3 
| `sql_injection_with_long_urls_filter`

Macros

The SPL above uses the following Macros:

:information_source: sql_injection_with_long_urls_filter is a empty macro by default. It allows the user to filter out any results (false positives) without editing the SPL.

Required fields

List of fields required to use this analytic.

  • _time
  • Web.dest_category
  • Web.url_length
  • Web.http_user_agent_length
  • Web.src
  • Web.dest
  • Web.url
  • Web.http_user_agent

How To Implement

To successfully implement this search, you need to be monitoring network communications to your web servers or ingesting your HTTP logs and populating the Web data model. You must also identify your web servers in the Enterprise Security assets table.

Known False Positives

It's possible that legitimate traffic will have long URLs or long user agent strings and that common SQL commands may be found within the URL. Please investigate as appropriate.

Associated Analytic Story

RBA

Risk Score Impact Confidence Message
25.0 50 50 SQL injection attempt with url $url$ detected on $dest$

:information_source: The Risk Score is calculated by the following formula: Risk Score = (Impact * Confidence/100). Initial Confidence and Impact is set by the analytic author.

Reference

Test Dataset

Replay any dataset to Splunk Enterprise by using our replay.py tool or the UI. Alternatively you can replay a dataset into a Splunk Attack Range

source | version: 4