Skip to main content
COSMICBYTEZLABS
NewsSecurityHOWTOsToolsStudyTraining
ProjectsChecklistsAI RankingsNewsletterStatusTagsAbout
Subscribe

Press Enter to search or Esc to close

News
Security
HOWTOs
Tools
Study
Training
Projects
Checklists
AI Rankings
Newsletter
Status
Tags
About
RSS Feed
Reading List
Subscribe

Stay in the Loop

Get the latest security alerts, tutorials, and tech insights delivered to your inbox.

Subscribe NowFree forever. No spam.
COSMICBYTEZLABS

Your trusted source for IT intelligence, cybersecurity insights, and hands-on technical guides.

1154+ Articles
126+ Guides

CONTENT

  • Latest News
  • Security Alerts
  • HOWTOs
  • Projects
  • Exam Prep

RESOURCES

  • Search
  • Browse Tags
  • Newsletter Archive
  • Reading List
  • RSS Feed

COMPANY

  • About Us
  • Contact
  • Privacy Policy
  • Terms of Service

© 2026 CosmicBytez Labs. All rights reserved.

System Status: Operational
  1. Home
  2. Security
  3. GlassFish Gadget Handler Expression Language RCE (CVE-2026-2587)
GlassFish Gadget Handler Expression Language RCE (CVE-2026-2587)

Critical Security Alert

This vulnerability is actively being exploited. Immediate action is recommended.

SECURITYCRITICALCVE-2026-2587

GlassFish Gadget Handler Expression Language RCE (CVE-2026-2587)

A critical CVSS 9.6 Remote Code Execution vulnerability in GlassFish's server-side gadget handler allows attackers to inject Expression Language...

Dylan H.

Security Team

May 20, 2026
5 min read

Affected Products

  • GlassFish gadget handler (XML template rendering)

Executive Summary

A critical unauthenticated Remote Code Execution vulnerability (CVE-2026-2587) has been disclosed in GlassFish's server-side gadget handler. The vulnerability stems from unsafe evaluation of user-supplied Expression Language (EL) expressions within the server's XML template processing pipeline. An attacker who can supply a crafted .xml file to the vulnerable endpoint can achieve full remote code execution without prior authentication.

With a CVSS score of 9.6 (Critical), this vulnerability is rated more severely than its sibling CVE-2026-2586 due to the lack of an authentication requirement, significantly lowering the barrier for exploitation.

CVSS Score: 9.6 (Critical)


Vulnerability Overview

AttributeValue
CVE IDCVE-2026-2587
CVSS Score9.6 (Critical)
TypeServer-Side Template Injection (EL Injection)
Attack VectorNetwork
Attack ComplexityLow
Privileges RequiredNone
User InteractionNone
Affected ComponentGlassFish gadget handler / XML template renderer
Public ExploitNot confirmed at publication
Published2026-05-19

Affected Products

ProductStatus
GlassFish (gadget handler / XML template rendering)Vulnerable

Technical Details

Vulnerability Root Cause

The GlassFish gadget handler processes .xml configuration files and evaluates user-supplied values within a context where Java Expression Language (EL) expressions are resolved at runtime. When an attacker can control the contents of such a file — or submit input that reaches the template renderer — embedded EL expressions like ${...} are evaluated by the Java EL engine prior to output, giving the attacker the ability to call arbitrary Java methods.

Expression Language injection in Java application servers can lead to full code execution because the EL context has access to the Java runtime, class loaders, and process execution facilities.

Attack Chain

1. Attacker identifies a GlassFish endpoint that processes XML input
   or accepts file uploads handled by the gadget renderer
 
2. Attacker crafts a malicious .xml file containing an EL expression
   designed to invoke OS command execution via Java Runtime or
   similar class accessible from the EL context
 
3. The GlassFish gadget handler parses the XML and evaluates
   the embedded EL expression
 
4. The Java EL engine resolves the expression, invoking the
   attacker-controlled code path
 
5. Arbitrary commands execute with the privileges of the
   GlassFish service account
 
6. Attacker achieves full remote code execution — no credentials required

Why CVSS 9.6?

The score reflects the complete absence of an authentication barrier (unlike CVE-2026-2586), combined with network accessibility, low complexity, no user interaction, and High impact across all three pillars (Confidentiality, Integrity, Availability). EL injection reaching OS command execution is considered one of the highest-severity classes of server-side vulnerability.


Impact Assessment

Impact AreaDescription
Unauthenticated RCEAny network-reachable attacker can execute arbitrary code
Full Server CompromiseAccess to all data and processes on the host at service account privilege
Credential TheftApplication secrets, database passwords, keystores, and tokens accessible
PersistenceAttackers can deploy backdoors or modify deployed applications
Supply Chain RiskCompromised application servers can be used to tamper with application artifacts
Lateral MovementInternal network access enables pivoting to databases and downstream systems

Recommendations

Immediate Actions

  1. Isolate internet-facing GlassFish instances immediately — disable or firewall the vulnerable endpoint pending a vendor patch
  2. Apply available patches from Oracle or the Eclipse GlassFish project as soon as released
  3. Disable XML gadget rendering endpoints if not required by your deployment
  4. Review server logs for evidence of exploitation attempts — look for unusual request patterns to XML-handling endpoints

Network-Level Mitigations

- Block external access to GlassFish service ports at the perimeter
- Validate and restrict file upload functionality to trusted content types
- Deploy a Web Application Firewall (WAF) to detect EL injection patterns
  in HTTP request bodies and parameters

WAF Detection Pattern (Example)

# Detect EL expression injection attempts in HTTP parameters
\$\{[^}]*\}|\#\{[^}]*\}

Detection Indicators

IndicatorDescription
EL expression syntax in HTTP POST bodies or query paramsActive exploitation attempt
Unexpected Java process spawning child processesPossible successful EL injection
Outbound connections from GlassFish service userPossible post-exploitation C2
New or modified files in GlassFish deployment directoriesPossible backdoor deployment

Relationship to CVE-2026-2586

Both CVE-2026-2586 and CVE-2026-2587 affect the GlassFish platform and were published on the same date, suggesting they were discovered as part of the same security research engagement. The key distinction:

CVE-2026-2586CVE-2026-2587
AuthenticationRequired (low privilege)Not required
CVSS9.19.6
VectorAdmin Console HTTP requestXML template EL injection

Organizations should treat both vulnerabilities as requiring immediate attention, but CVE-2026-2587 represents the higher priority due to its unauthenticated exploitability.


Post-Remediation Checklist

  1. Apply official patch — verify the GlassFish version post-update via asadmin version
  2. Audit XML processing endpoints — disable or sanitize all endpoints that accept user-controlled XML input
  3. Rotate all credentials on the affected server: database passwords, API keys, SSL certificates
  4. Scan for web shells or unauthorized files in GlassFish domain directories
  5. Review network egress logs for suspicious outbound connections during the vulnerability exposure window
  6. Update WAF rules to detect EL injection patterns in application traffic

References

  • NIST NVD — CVE-2026-2587
  • NIST NVD — CVE-2026-2586
  • Eclipse GlassFish Project
  • CWE-917: Server-Side Template Injection
  • OWASP EL Injection
#GlassFish#CVE-2026-2587#RCE#Remote Code Execution#Expression Language#SSTI#Java#NVD

Related Articles

GlassFish Administration Console Authenticated RCE (CVE-2026-2586)

An authenticated Remote Code Execution vulnerability in GlassFish's Administration Console (CVSS 9.1) allows users with panel access to execute arbitrary...

5 min read

CVE-2026-44050 — Netatalk CNID Daemon Heap Buffer Overflow RCE (CVSS 9.9)

A heap-based buffer overflow in the Netatalk CNID daemon comm_rcv() function allows a remote authenticated attacker to execute arbitrary code with...

6 min read

CVE-2025-15379: MLflow Command Injection in Model Serving (CVSS 10.0)

A maximum-severity command injection vulnerability in MLflow's model serving container initialization allows attackers to execute arbitrary OS commands...

6 min read
Back to all Security Alerts