GHSA-F346-8RP3-4H9H
Vulnerability from github – Published: 2026-03-27 15:42 – Updated: 2026-03-27 15:42Summary
A flaw in TSPortal allowed attackers to create arbitrary user records in the database by abusing validation logic. While validation correctly rejected invalid usernames, a side effect within a validation rule caused user records to be created regardless of whether the request succeeded. This could be exploited to cause uncontrolled database growth, leading to a potential denial of service (DoS).
Details
When submitting a Data Processing Agreement (DPA) request in TSPortal, the DPAAlreadyLive validation rule previously called User::findOrCreate().
This method created a user record if one did not already exist.
Although username validation (via MirahezeUsernameRule) correctly rejected invalid usernames, the DPAAlreadyLive rule was still executed during validation. Because it performed a state-changing operation, it created user records even when the overall validation failed and no DPA was created.
As a result: - Validation correctly rejected invalid input - However, user records were still inserted into the database as a side effect
These records were created: - Without a successful DPA request - Without audit logging tied to a completed action - Without visibility into their origin
Impact
An attacker could exploit this behavior by automating requests with invalid usernames, resulting in:
- Mass creation of arbitrary user records
- Unbounded database growth
- Increased storage and indexing overhead
- Potential degradation of application performance
At scale, this could lead to a denial of service condition due to resource exhaustion.
Proof of Concept
- Submit a DPA request using an invalid username
- Ensure the request fails validation due to
MirahezeUsernameRule - Observe that a corresponding user record is still created in the database
This behavior was confirmed prior to remediation.
Root Cause
The issue stemmed from:
- Performing state-changing operations (findOrCreate) inside validation logic
- Validation rules executing regardless of overall validation success
- Lack of separation between validation and persistence layers
Mitigation
The issue has been fixed by removing database write operations from validation logic.
Specifically:
- Replaced User::findOrCreate() with a non-mutating lookup (User::firstWhere(...))
- Ensured validation rules only perform read operations
- Prevented user creation unless all validation passes
{
"affected": [
{
"database_specific": {
"last_known_affected_version_range": "\u003c= 33"
},
"package": {
"ecosystem": "Packagist",
"name": "miraheze/ts-portal"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "34"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-33541"
],
"database_specific": {
"cwe_ids": [
"CWE-400",
"CWE-770"
],
"github_reviewed": true,
"github_reviewed_at": "2026-03-27T15:42:20Z",
"nvd_published_at": "2026-03-26T21:17:05Z",
"severity": "MODERATE"
},
"details": "### Summary\nA flaw in TSPortal allowed attackers to create arbitrary user records in the database by abusing validation logic. While validation correctly rejected invalid usernames, a side effect within a validation rule caused user records to be created regardless of whether the request succeeded. This could be exploited to cause uncontrolled database growth, leading to a potential denial of service (DoS).\n\n### Details\nWhen submitting a Data Processing Agreement (DPA) request in TSPortal, the `DPAAlreadyLive` validation rule previously called `User::findOrCreate()`.\n\nThis method created a user record if one did not already exist.\n\nAlthough username validation (via `MirahezeUsernameRule`) correctly rejected invalid usernames, the `DPAAlreadyLive` rule was still executed during validation. Because it performed a state-changing operation, it created user records even when the overall validation failed and no DPA was created.\n\nAs a result:\n- Validation correctly rejected invalid input\n- However, user records were still inserted into the database as a side effect\n\nThese records were created:\n- Without a successful DPA request\n- Without audit logging tied to a completed action\n- Without visibility into their origin\n\n### Impact\nAn attacker could exploit this behavior by automating requests with invalid usernames, resulting in:\n\n- Mass creation of arbitrary user records\n- Unbounded database growth\n- Increased storage and indexing overhead\n- Potential degradation of application performance\n\nAt scale, this could lead to a denial of service condition due to resource exhaustion.\n\n### Proof of Concept\n1. Submit a DPA request using an invalid username\n2. Ensure the request fails validation due to `MirahezeUsernameRule`\n3. Observe that a corresponding user record is still created in the database\n\nThis behavior was confirmed prior to remediation.\n\n### Root Cause\nThe issue stemmed from:\n- Performing state-changing operations (`findOrCreate`) inside validation logic\n- Validation rules executing regardless of overall validation success\n- Lack of separation between validation and persistence layers\n\n### Mitigation\nThe issue has been fixed by removing database write operations from validation logic.\n\nSpecifically:\n- Replaced `User::findOrCreate()` with a non-mutating lookup (`User::firstWhere(...)`)\n- Ensured validation rules only perform read operations\n- Prevented user creation unless all validation passes",
"id": "GHSA-f346-8rp3-4h9h",
"modified": "2026-03-27T15:42:20Z",
"published": "2026-03-27T15:42:20Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/miraheze/TSPortal/security/advisories/GHSA-f346-8rp3-4h9h"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-33541"
},
{
"type": "PACKAGE",
"url": "https://github.com/miraheze/TSPortal"
},
{
"type": "WEB",
"url": "https://issue-tracker.miraheze.org/T15115"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
],
"summary": "TSPortal\u0027s Uncontrolled User Creation via Validation Side Effects Leads to Potential Denial of Service"
}
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.