---
id: "CVE-2026-121819"
page: "https://infrastack.ai/live/CVE-2026-121819"
kind: "vulnerability"
aliases: ["CVE-2026-121819"]
title: "armeria-xds: symlink check lets control plane read files outside allowed directories"
severity: "medium"
cvss: 5.9
epss: null
exploited: false
status: "open"
category: "package"
products: []
ecosystems: []
tags: ["armeria", "grpc", "java", "package", "xds"]
not_affected_if: ["default-config", "unaffected-version"]
published_at: "2026-11-06T08:17:20+00:00"
first_seen_at: "2026-11-06T08:40:33+00:00"
updated_at: "2026-11-06T08:42:38+00:00"
sources: [{"source": "nvd", "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-121819", "first_seen_at": "2026-11-06T08:40:33+00:00"}]
exploit_sources: []
---

# armeria-xds: symlink check lets control plane read files outside allowed directories

Link for people: https://infrastack.ai/live/CVE-2026-121819 (this .md is for agents)

armeria-xds 1.40.0 through 1.42.0 checks file paths without resolving symbolic links, so a compromised or semi-trusted xDS control plane can make the client read files outside its allowed root directories. In deployments that use a credential injector, text secrets may be sent to an upstream an attacker can observe.

_Summary, check and fix written by a model from the sources; labelled as such._

## How to check

- Check the armeria-xds version in your dependency file (for example `build.gradle` or `pom.xml`) against the affected range `1.40.0` through `1.42.0`.
- Check whether your service uses DataSourcePolicy with allowed root directories and a credential injector.

## Fix

- armeria-xds: unknown (the sources give no fixed version)

## Timeline

- 2026-11-06T08:40:33+00:00 nvd https://nvd.nist.gov/vuln/detail/CVE-2026-121819

## Documents

- [advisory] Vendor advisory: github.com https://github.com/line/armeria/security/advisories/GHSA-m64h-m6r5-6jrx

## From the sources

A vulnerability has been identified in armeria-xds versions 1.40.0 through 1.42.0, where DataSourcePolicy checks filename containment without resolving symbolic links, allowing a compromised or semi-trusted xDS control plane to cause the xDS client to read files outside configured allowed root directories when a symbolic link inside an allowed directory points outside it. In deployments using a credential injector, header-compatible text secrets may be exposed to an attacker-observable upstream.

---

_Free to read, with fair use of about 60 pages a minute. For higher limits or API access: sales@infrastack.ai_