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-2026-110723 Overview
CVE-2026-110723 is a rate-limiting bypass vulnerability in the Eclipse ECSP api-gateway component through version 1.7.3. The default key resolver, ClientIpKeyResolver, derives per-client rate-limit bucket keys from the leftmost value of the X-Forwarded-For HTTP header. Because the gateway does not validate this header against a trusted-proxy list, a single attacker can forge arbitrary values and generate an unlimited number of independent rate-limit buckets. This defeats the request-rate protections on every route that uses the shipped default configuration, enabling resource exhaustion and brute-force attacks against downstream services. The weakness is tracked as CWE-290: Authentication Bypass by Spoofing.
Critical Impact
Attackers can bypass gateway rate limiting entirely by rotating the X-Forwarded-For header, exposing backend services to abuse and denial-of-service conditions.
Affected Products
- Eclipse ECSP api-gateway versions up to and including 1.7.3
- Deployments using the default ClientIpKeyResolver rate-limit key resolver
- Any downstream route protected only by the gateway's default request-rate limiter
Discovery Timeline
- 2026-10-15 - CVE-2026-110723 published to the National Vulnerability Database
- 2026-10-15 - Last updated in NVD database
Technical Details for CVE-2026-110723
Vulnerability Analysis
The Eclipse ECSP api-gateway ships with a default rate-limiting mechanism that identifies clients by IP address. The ClientIpKeyResolver component builds this identifier by reading the leftmost entry of the X-Forwarded-For request header. HTTP clients fully control that header value, and the gateway performs no verification that the request arrived through a known reverse proxy. Attackers can therefore submit each request with a different fabricated address and receive a fresh rate-limit bucket for every value.
The net result is complete evasion of the shipped request-rate controls. Routes that rely on default configuration will accept unbounded request volumes from a single source, enabling credential brute force, scraping, and resource exhaustion against protected upstream services. The impact scope is limited to availability of downstream systems, since the flaw does not expose confidentiality or integrity of gateway data.
Root Cause
The root cause is trust in a client-controlled HTTP header without proxy allow-listing. X-Forwarded-For is intended for use between the client and a trusted edge proxy that appends the real source address. When the gateway consumes the leftmost element directly, it treats attacker-supplied data as an authoritative identity. This matches the CWE-290 pattern of authentication bypass by spoofing an identifier that should have been validated.
Attack Vector
Exploitation requires only network access to the gateway and the ability to set arbitrary HTTP headers. An attacker iterates through random or sequential X-Forwarded-For values and issues requests at any rate. Each unique header value maps to a new bucket in the token-bucket store, so no bucket reaches its threshold. No authentication, user interaction, or special privileges are required.
Technical specifics are documented in the GitHub Security Advisory GHSA-m468-w7pr-f39w and the upstream fix pull request.
Detect Identity Attacks the Moment Authentication Fails
Singularity™ Identity surfaces credential misuse, lateral movement, and privilege escalation in real time, so you can contain the breach before it spreads.
Stop Identity AttacksDetection Methods for CVE-2026-110723
Indicators of Compromise
- High volumes of requests to the same route where each request carries a unique or sequential X-Forwarded-For value.
- Rate-limit metrics showing bucket counts far exceeding the number of plausible unique clients.
- Access logs where the leftmost X-Forwarded-For entry is not consistent with the TCP peer address observed by the edge proxy.
Detection Strategies
- Compare the leftmost X-Forwarded-For value against the real source IP of each connection and alert on mismatches when the connection did not traverse a known proxy.
- Baseline the cardinality of client IPs per route per minute, then alert on sudden spikes that indicate header rotation.
- Enable gateway debug or audit logging for the ClientIpKeyResolver to record resolved bucket keys for forensic review.
Monitoring Recommendations
- Forward gateway access logs and rate-limit telemetry to a centralized SIEM for correlation with upstream service error rates.
- Monitor downstream services for authentication-failure bursts, which may indicate credential-stuffing attempts proxied through the bypass.
- Track backend latency and connection-pool saturation metrics, as these are early signs of successful rate-limit evasion.
How to Mitigate CVE-2026-110723
Immediate Actions Required
- Upgrade Eclipse ECSP api-gateway to a release above 1.7.3 that includes the fix from pull request #330.
- Replace the default ClientIpKeyResolver with a resolver that uses the TCP remote address or a validated header from a trusted proxy.
- Configure an explicit trusted-proxy allow-list so X-Forwarded-For is only honored when the request originates from a known edge device.
Patch Information
The maintainers addressed the issue upstream. Review the GitHub Security Advisory GHSA-m468-w7pr-f39w and the GitLab CVE assignment work item for the authoritative fix version and patch notes. Apply the patched release across all gateway instances and redeploy configurations that rely on default rate-limit behavior.
Workarounds
- Implement a custom KeyResolver that derives the bucket key from the authenticated principal or the direct TCP peer address instead of a client-controlled header.
- Terminate TLS and enforce request-rate limits at an upstream reverse proxy or web application firewall that strips untrusted X-Forwarded-For values before forwarding.
- Apply per-route request-rate limits based on authentication tokens or API keys so that unauthenticated traffic cannot bypass controls through header manipulation.
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

