Allocation of Resources Without Limits or Throttling in ueberauth guardian allows denial of service via unbounded atom creation from attacker-controlled binary input.

Guardian.Permissions.AtomEncoding encodes permission scopes by passing arbitrary binaries to String.toatom/1. When encode/3 in lib/guardian/permissions/atomencoding.ex is called with a list, each binary entry is handled by the encodevalue/3 binary clause, which calls String.toatom(value) with no allow-list check. The permset argument (the application's small, finite set of legitimate permission names) is discarded, so any external string flows straight into atom creation. This encoder is selected with use Guardian.Permissions, encoding: Guardian.Permissions.AtomEncoding and reached through the imported encode/3 entry point.

String.toatom/1 creates a brand-new atom for every previously unseen binary, atoms are never garbage collected, and the BEAM atom table is fixed at roughly 1,048,576 entries by default. An application that funnels attacker-influenced permission scopes (from a request body, a JWT claim, or other external input) into encode/3 therefore mints one permanent atom per distinct value. A modest stream of varied, unauthenticated input permanently consumes the atom table and crashes the BEAM node with systemlimit, taking down every application running on it.

The default encoder is Guardian.Permissions.BitwiseEncoding, which is not affected.

This issue affects guardian: from 2.0.0 before 2.4.1.

A denial-of-service vulnerability caused by unbounded resource allocation was discovered in the audit logging functionality, due to a missing size limit on input recorded into audit entries. An unauthenticated attacker can submit requests containing excessively large input that is recorded into audit entries, possibly exhausting the available disk space and rendering the system inoperable.

Guardian language-system passes the id GET parameter directly into a PHP exec() call in transcribe.php (line 15) without sanitization: exec(\"php jobs/transcribe.php \".$loginsession.\" \".$GET['id'].\" ...\"). No authentication is required. An unauthenticated remote attacker can append shell metacharacters to execute arbitrary OS commands on the server.

Guardian language-system passes the id GET parameter directly into a PHP exec() call in text.php (line 15) without sanitization: exec(\"php jobs/text.php \".$loginsession.\" \".$GET['id'].\" ...\"). No authentication is required. An unauthenticated remote attacker can append shell metacharacters to execute arbitrary OS commands on the server.

Guardian language-system passes the id GET parameter directly into a PHP exec() call in translate.php (line 14) without sanitization: exec(\"php jobs/translate.php \".$loginsession.\" \".$GET['id'].\" ...\"). No authentication is required. An unauthenticated remote attacker can append shell metacharacters to execute arbitrary OS commands on the server.

Guardian language-system passes the name GET parameter directly into an unsanitized SQL query in designer.php (line 124): SELECT FROM complex WHERE name='\".$GET['name'].\"'. An authenticated attacker can perform error-based SQL injection to extract database contents.

Guardian language-system passes the id GET parameter directly into an unsanitized SQL query in subtitles.php (line 16): SELECT id, filename, extension, type FROM files where id = '\".$GET['id'].\"'. An authenticated attacker can perform error-based SQL injection to extract database contents.

Guardian language-system passes the id GET parameter directly into an unsanitized SQL query in textfile.php (line 17): SELECT id, filename, extension, type, duration, owner, private FROM files where id = '\".$GET['id'].\"'. An authenticated attacker can perform error-based SQL injection to extract database contents.

A Stored HTML Injection vulnerability was discovered in the Schedule Restore Archive functionality due to improper validation of an input parameter. An authenticated user with administrative privileges can define a malicious restore schedule containing HTML tags. When a victim views the affected schedule, the injected HTML renders in their browser, enabling phishing and possibly open redirect attacks. Full XSS exploitation and direct information disclosure are prevented by the existing input validation and Content Security Policy configuration.

An Angular template injection vulnerability was discovered in the Reports functionality due to improper validation of an input parameter. An authenticated user with report privileges can define a malicious report containing an Angular template payload, or a victim can be socially engineered to import a malicious report template. When the victim views or imports the report, the Angular template executes in their browser context, allowing the attacker to modify application data, or disrupt application availability. Full XSS exploitation and direct information disclosure are prevented by the existing input validation and Content Security Policy configuration.

A Stored Cross-Site Scripting vulnerability was discovered in the Assets and Nodes functionality due to improper validation of an input parameter. An authenticated user with custom fields privileges can define a malicious custom field containing a JavaScript payload. When the victim views the Assets or Nodes pages, the XSS executes in their browser context, allowing the attacker to perform unauthorized actions as the victim, such as modify application data, disrupt application availability, and access limited sensitive information.

A Stored HTML Injection vulnerability was discovered in the Alerted Nodes Dashboard functionality due to improper validation on an input parameter.

A malicious authenticated user with the required privileges could edit a node label to inject HTML tags. If the system is configured to use the Alerted Nodes Dashboard, and alerts are reported for the affected node, then the injected HTML may render in the browser of a victim user interacting with it, enabling phishing and possibly open redirect attacks. Full XSS exploitation and direct information disclosure are prevented by the existing input validation and Content Security Policy configuration.

An issue in the TLS certification mechanism of Guardian Gryphon v01.06.0006.22 allows attackers to execute commands as root.

A path traversal vulnerability was discovered in the Import Arc data archive functionality due to insufficient validation of the input file. An authenticated user with limited privileges, by uploading a specifically-crafted Arc data archive, can potentially write arbitrary files in arbitrary paths, altering the device configuration and/or affecting its availability.

A Stored HTML Injection vulnerability was discovered in the Time Machine Snapshot Diff functionality due to improper validation of network traffic data. An unauthenticated attacker can send specially crafted network packets at two different times to inject HTML tags into asset attributes across two snapshots. Exploitation requires a victim to use the Time Machine Snapshot Diff feature on those specific snapshots and perform specific GUI actions, at which point the injected HTML renders in their browser, enabling phishing and open redirect attacks. Full XSS exploitation is prevented by input validation and Content Security Policy. Attack complexity is high due to multiple required conditions.

A Stored Cross-Site Scripting vulnerability was discovered in the Dashboards functionality due to improper validation of an input parameter.

An authenticated low-privilege user can craft a malicious dashboard containing a JavaScript payload and share it with victim users, or a victim can be socially engineered to import a malicious dashboard template. When the victim views or imports the dashboard, the XSS executes in their browser context, allowing the attacker to perform unauthorized actions as the victim, such as modify application data, disrupt application availability, and access limited sensitive information.

A SQL Injection vulnerability was discovered in the CLI functionality due to improper validation of an input parameter. An authenticated user with limited privileges can execute arbitrary SELECT SQL statements on the DBMS used by the web application, potentially exposing unauthorized data.

A path traversal vulnerability was discovered in the Time Machine functionality due to missing validation of two input parameters. An authenticated user with limited privileges, by issuing a specifically-crafted request, can potentially alter the structure and content of files in the /data folder, and/or affect their availability.

A SQL Injection vulnerability was discovered in the Alert functionality due to improper validation of an input parameter. An authenticated user with limited privileges can execute arbitrary SELECT SQL statements on the DBMS used by the web application, potentially exposing unauthorized data.

A SQL Injection vulnerability was discovered in the Alert functionality due to improper validation of an input parameter. An authenticated user with limited privileges can execute arbitrary SQL statements on the DBMS used by the web application, potentially exposing unauthorized data, altering their structure and content, and/or affecting their availability.

A SQL Injection vulnerability was discovered in the Smart Polling functionality due to improper validation of an input parameter. An authenticated user with limited privileges can execute arbitrary SELECT SQL statements on the DBMS used by the web application, potentially exposing unauthorized data.

An access control vulnerability was discovered in the CLI functionality due to a specific access restriction not being properly enforced for users with limited privileges. An authenticated user with limited privileges can issue administrative CLI commands, altering the device configuration, and/or affecting its availability.

A client-side path traversal vulnerability was discovered in the web management interface front-end due to missing validation of an input parameter. An authenticated user with limited privileges can craft a malicious URL which, if visited by an authenticated victim, leads to a Cross-Site Scripting (XSS) attack.

A privilege escalation vulnerability may enable a service account to elevate its privileges.

The sudo rules configured for a local service account were excessively permissive, potentially allowing administrative access if a malicious actor could execute arbitrary commands as that account.

It is important to note that no such vector has been identified in this instance.

Audit records for OpenAPI requests may include sensitive information.

This could lead to unauthorized accesses and privilege escalation.