Skip to main content
Vulnerability Database/CVE-2026-114058

CVE-2026-114058: ArcadeDB Information Disclosure Vulnerability

CVE-2026-114058 is an information disclosure flaw in ArcadeDB that allows authenticated users to access unauthorized database schema metadata. This post explains its impact, affected versions, and mitigation steps.

Published:

Disclaimer: This content was generated using AI. While we strive for accuracy, please verify critical information with official sources.

Sources: National Vulnerability Database, CISA KEV, Exploit-DB

CVE-2026-114058 Overview

CVE-2026-114058 is a missing authorization vulnerability [CWE-862] in ArcadeDB's Bolt protocol handler. The flaw affects ArcadeDB versions from 26.2.1 before 26.10.1. Authenticated users holding credentials for one database can read schema metadata from other databases on the same instance. Attackers set the db field in RUN or BEGIN messages and invoke db.labels() or SHOW INDEXES to enumerate schemas they are not authorized to access. The vulnerability enables cross-tenant schema disclosure within a shared ArcadeDB deployment.

Critical Impact

Authenticated users can enumerate schema metadata, labels, and indexes belonging to databases they do not own, breaking tenant isolation in shared ArcadeDB instances.

Affected Products

  • ArcadeDB versions 26.2.1 through 26.10.0
  • Bolt protocol handler component
  • Multi-database ArcadeDB deployments sharing authentication

Discovery Timeline

  • 2026-10-21 - CVE-2026-114058 published to NVD
  • 2026-10-21 - Last updated in NVD database

Technical Details for CVE-2026-114058

Vulnerability Analysis

The vulnerability stems from the ArcadeDB Bolt protocol handler failing to validate that the authenticated user has permission to access the database named in protocol-level messages. Bolt messages such as RUN and BEGIN include a db field that selects the target database for the transaction. ArcadeDB accepts this field without verifying the user's authorization against the requested database. An attacker who authenticates to any database on the instance can redirect queries toward other databases and retrieve schema information.

The exposure is limited to schema metadata rather than row-level data. Queries such as db.labels() return the list of node labels defined in the target database. SHOW INDEXES returns index definitions, including the properties they cover. This information discloses structural details that can inform follow-on attacks or reveal sensitive naming conventions.

Root Cause

The root cause is missing authorization enforcement at the Bolt protocol layer. Authentication confirms the user's identity against one database, but the handler does not re-check authorization when a subsequent message specifies a different db target. Access control is applied inconsistently between the authentication step and the per-message routing step.

Attack Vector

Exploitation requires network reachability to the Bolt endpoint and valid credentials for at least one database on the ArcadeDB server. The attacker opens a Bolt session, authenticates normally, and then issues a RUN or BEGIN message with the db field pointing to an unauthorized database. The server executes schema-metadata procedures such as db.labels() or SHOW INDEXES against that database and returns the results. No user interaction is required, and the attack can be scripted against each database name the attacker wishes to enumerate.

See the GitHub Security Advisory GHSA-hw9x-xx38-rg28 and the VulnCheck Advisory for ArcadeDB for additional technical context.

Detection Methods for CVE-2026-114058

Indicators of Compromise

  • Bolt protocol sessions in which the authenticated user issues RUN or BEGIN messages targeting a db value that differs from the database used during authentication.
  • Execution of db.labels() or SHOW INDEXES against databases the session user is not provisioned for.
  • Repeated Bolt requests cycling through multiple database names from a single authenticated session, consistent with enumeration behavior.

Detection Strategies

  • Enable Bolt protocol logging on the ArcadeDB server and correlate the authenticated principal with the db field in each RUN and BEGIN message.
  • Build alerting on mismatches between the authentication database and the per-message target database.
  • Monitor for high-frequency execution of schema introspection procedures from accounts that do not normally perform administrative tasks.

Monitoring Recommendations

  • Capture ArcadeDB access logs centrally and retain Bolt session metadata for forensic review.
  • Baseline normal query patterns per service account so cross-database schema queries stand out.
  • Alert on any authenticated user invoking SHOW INDEXES or db.labels() against more than one database within a short interval.

How to Mitigate CVE-2026-114058

Immediate Actions Required

  • Upgrade ArcadeDB to version 26.10.1 or later, which contains the authorization fix for the Bolt protocol handler.
  • Rotate credentials for any account that may have been shared across databases while the vulnerable version was deployed.
  • Review Bolt access logs for cross-database RUN or BEGIN traffic and investigate any matches.

Patch Information

The vendor fixed the missing authorization check in ArcadeDB 26.10.1. Administrators should upgrade all nodes in a cluster to the patched release. Full technical details and patch references are available in the GitHub Security Advisory GHSA-hw9x-xx38-rg28.

Workarounds

  • Restrict network access to the Bolt port so only trusted application tiers can reach it.
  • Isolate sensitive databases onto separate ArcadeDB instances until patching is complete, removing the shared authentication boundary.
  • Disable the Bolt protocol interface if the deployment relies solely on HTTP or other protocols, eliminating the vulnerable code path.

Disclaimer: This content was generated using AI. While we strive for accuracy, please verify critical information with official sources.

Sources: National Vulnerability Database, CISA KEV, Exploit-DB

Experience the Most Advanced Cybersecurity Platform

See how the world’s most intelligent, autonomous cybersecurity platform can protect your organization today and into the future.