Skip to main content
CVE Vulnerability Database
Vulnerability Database/CVE-2025-13927

CVE-2025-13927: GitLab Denial of Service Vulnerability

CVE-2025-13927 is a denial of service vulnerability in GitLab CE/EE that allows unauthenticated attackers to disrupt services via malformed authentication requests. This article covers technical details, affected versions, and mitigation.

Updated:

CVE-2025-13927 Overview

GitLab has patched a denial of service vulnerability affecting GitLab Community Edition (CE) and Enterprise Edition (EE). The flaw exists in all versions from 11.9 before 18.6.4, 18.7 before 18.7.2, and 18.8 before 18.8.2. An unauthenticated remote attacker can trigger a denial of service condition by sending crafted requests containing malformed authentication data. The vulnerability is categorized under CWE-770: Allocation of Resources Without Limits or Throttling.

Critical Impact

Unauthenticated attackers can disrupt GitLab availability across all supported deployments by submitting malformed authentication payloads over the network.

Affected Products

  • GitLab Community Edition (CE) — all versions from 11.9 before 18.6.4
  • GitLab Enterprise Edition (EE) — 18.7 before 18.7.2
  • GitLab CE/EE — 18.8 before 18.8.2

Discovery Timeline

  • 2026-01-21 - GitLab releases patch versions 18.8.2, 18.7.2, and 18.6.4
  • 2026-01-22 - CVE-2025-13927 published to NVD
  • 2026-01-26 - Last updated in NVD database

Technical Details for CVE-2025-13927

Vulnerability Analysis

The vulnerability is a denial of service flaw in GitLab CE/EE authentication processing. An unauthenticated attacker can send crafted HTTP requests with malformed authentication data to exhaust server resources. The condition is reachable over the network without user interaction, which expands the exposed attack surface to any internet-facing GitLab instance.

The issue is classified as CWE-770: Allocation of Resources Without Limits or Throttling. GitLab processes authentication payloads without sufficiently constraining resource consumption when input is malformed. This allows attackers to degrade availability for legitimate users, including developers, CI/CD pipelines, and integrated services.

Root Cause

The root cause is improper handling of malformed authentication data during request processing. GitLab's authentication pipeline does not adequately validate or throttle resource allocation when parsing invalid inputs. Repeated or large malformed requests trigger resource exhaustion conditions on the GitLab application server.

Attack Vector

The attack vector is network-based and requires no authentication or user interaction. An attacker sends a stream of HTTP requests containing crafted authentication headers or payloads to a vulnerable GitLab endpoint. The server consumes CPU, memory, or worker threads while processing these malformed inputs, eventually rendering the instance unresponsive. Refer to the HackerOne Security Report #3439683 and the GitLab Issue Discussion for additional technical context.

No public proof-of-concept exploit is available at the time of publication.

Detection Methods for CVE-2025-13927

Indicators of Compromise

  • Sudden spikes in failed or malformed authentication requests against GitLab endpoints such as /users/sign_in, /oauth/token, or API authentication paths.
  • Elevated CPU, memory, or Puma/Unicorn worker saturation on GitLab application nodes without a corresponding increase in legitimate traffic.
  • HTTP 5xx error rates rising on authentication endpoints, often accompanied by request timeouts.

Detection Strategies

  • Inspect GitLab production.log and auth.log for high volumes of authentication failures originating from a small set of source IPs or user agents.
  • Correlate web server access logs with application performance metrics to identify request patterns consistent with resource exhaustion.
  • Deploy WAF or reverse proxy rules to flag malformed Authorization headers and abnormally large authentication payloads.

Monitoring Recommendations

  • Alert on sustained increases in request latency on /api/v4/ and session endpoints relative to baseline.
  • Monitor Puma worker queue depth, GitLab Workhorse error counts, and Redis/Sidekiq backlog growth.
  • Track outbound 429 and 5xx response rates to detect attacker probing or rate limiting bypass attempts.

How to Mitigate CVE-2025-13927

Immediate Actions Required

  • Upgrade self-managed GitLab instances to 18.8.2, 18.7.2, or 18.6.4 as appropriate for your deployment branch.
  • Restrict network exposure of GitLab authentication endpoints to trusted networks or VPNs where feasible.
  • Enable rate limiting on authentication endpoints under Admin Area > Settings > Network to reduce abuse potential.

Patch Information

GitLab released fixed versions on January 21, 2026. Administrators should install GitLab CE/EE 18.8.2, 18.7.2, or 18.6.4 to remediate the vulnerability. Full details are available in the GitLab Release Patch Announcement. GitLab.com SaaS customers are running the patched version and require no action.

Workarounds

  • No vendor-supplied workaround exists; upgrading is the supported remediation path.
  • As a compensating control, place GitLab behind a WAF or reverse proxy configured to drop requests with malformed Authorization headers.
  • Tighten per-IP rate limits on sign-in and OAuth endpoints to slow exhaustion attempts until patching is complete.
bash
# Configuration example: enforce rate limits in /etc/gitlab/gitlab.rb
gitlab_rails['rate_limit_requests_per_period'] = 10
gitlab_rails['rate_limit_period'] = 60
gitlab_rails['throttle_authenticated_api_enabled'] = true
gitlab_rails['throttle_unauthenticated_enabled'] = true
gitlab_rails['throttle_unauthenticated_requests_per_period'] = 100
gitlab_rails['throttle_unauthenticated_period_in_seconds'] = 60

# Apply changes
sudo gitlab-ctl reconfigure

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

Default Legacy - Prefooter | Experience the World’s Most Advanced Cybersecurity Platform

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.