Apache Linkis: Arbitrary File Read via Double URL Encoding Bypass
Description
A vulnerability in Apache Linkis.
Problem Description When using the JDBC engine and da When using the JDBC engine and data source functionality, if the URL parameter configured on the frontend has undergone multiple rounds of URL encoding, it may bypass the system's checks. This bypass can trigger a vulnerability that allows unauthorized access to system files via JDBC parameters.
Scope of Impact
This issue affects Apache Linkis: from 1.3.0 through 1.7.0.
Severity level
moderate Solution Continuously check if the connection information contains the "%" character; if it does, perform URL decoding.
Users are recommended to upgrade to version 1.8.0, which fixes the issue.
More questions about this vulnerability can be discussed here: https://lists.apache.org/list?dev@linkis.apache.org:2025-9:cve
AI Insight
LLM-synthesized narrative grounded in this CVE's description and references.
Apache Linkis JDBC engine and data source functionality allow arbitrary file read via double URL encoding bypass, affecting versions 1.3.0 through 1.7.0.
A vulnerability in Apache Linkis (CVE-2025-29847) affects the JDBC engine and data source functionality. The issue arises when a URL parameter configured on the frontend undergoes multiple rounds of URL encoding, which can bypass the system's input validation checks [2][3]. This encoding discrepancy allows an attacker to inject malicious JDBC parameters that evade the intended filtering mechanisms.
An attacker can exploit this vulnerability by crafting a specially encoded JDBC URL. The bypass relies on the fact that the system does not properly decode the URL after multiple encoding iterations, leading to the acceptance of malicious parameters that would otherwise be blocked [3]. No authentication is required if the attacker can supply the JDBC connection string through the frontend interface
Successful exploitation allows unauthorized access to system files via JDBC parameters, potentially leading to sensitive data exposure [2][3]. The vulnerability is rated as moderate severity and affects Apache Linkis versions from 1.3.0 through 1.7.0
The recommended mitigation is to upgrade to Apache Linkis version 1.8.0, which fixes the issue. As a temporary workaround, administrators should continuously check if connection information contains the '%' character and perform URL decoding where necessary [2][3].
AI Insight generated on May 19, 2026. Synthesized from this CVE's description and the cited reference URLs; citations are validated against the source bundle.
Affected packages
Versions sourced from the GitHub Security Advisory.
| Package | Affected versions | Patched versions |
|---|---|---|
org.apache.linkis:linkisMaven | >= 1.3.0, < 1.8.0 | 1.8.0 |
Affected products
2Patches
0No patches discovered yet.
Vulnerability mechanics
AI mechanics synthesis has not run for this CVE yet.
References
5- github.com/advisories/GHSA-c399-q49h-qwc8ghsaADVISORY
- lists.apache.org/thread/03l5rfkgdt022o75jp8x4tzpqxz8g057ghsavendor-advisoryWEB
- nvd.nist.gov/vuln/detail/CVE-2025-29847ghsaADVISORY
- www.openwall.com/lists/oss-security/2025/09/19/2ghsaWEB
- lists.apache.org/listghsaWEB
News mentions
0No linked articles in our index yet.