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-2026-111386 Overview
CVE-2026-111386 is an information disclosure vulnerability in Nezha, an open-source server monitoring and management dashboard. The flaw affects versions from 1.7.0 before 2.3.15 and allows authenticated tenants to enumerate server IDs belonging to other tenants. The weakness is tracked as CWE-204: Observable Response Discrepancy. An authenticated member can iterate sequential server IDs against the getServerConfig and setServerConfig endpoints and distinguish permission denied responses from successful payload responses. This differential behavior exposes the existence and identifiers of resources owned by other tenants in multi-tenant deployments.
Critical Impact
Authenticated attackers can map other tenants' server inventories by sweeping sequential IDs and observing response differences, undermining multi-tenant isolation.
Affected Products
- Nezha dashboard versions 1.7.0 through 2.3.14
- Multi-tenant Nezha deployments exposing getServerConfig and setServerConfig
- Self-hosted Nezha instances prior to patch commit 36b742a
Discovery Timeline
- 2026-10-16 - CVE-2026-111386 published to the National Vulnerability Database (NVD)
- 2026-10-16 - Last updated in NVD database
Technical Details for CVE-2026-111386
Vulnerability Analysis
The Nezha dashboard exposes server configuration endpoints that enforce ownership checks after resolving the target server by ID. When a caller requests a server ID they do not own, the backend returns a distinct permission denied response. When the ID does not exist, the backend returns a different error. This difference in observable responses lets any authenticated member distinguish valid tenant-owned IDs from unused IDs.
An attacker can script a numeric sweep across the getServerConfig and setServerConfig endpoints and classify each response. The result is a reliable enumeration of other tenants' server IDs, which can seed follow-on reconnaissance against the exposed assets. The flaw does not grant direct read access to the configuration payloads of other tenants, but it breaks the confidentiality boundary around resource identifiers.
Root Cause
The root cause is an observable response discrepancy [CWE-204]. The handlers return different error signals for not found versus permission denied, instead of returning a uniform response for any ID the caller is not entitled to see. The patch in commit 36b742a7e9a81ed782363dd19efd77b7030fbd0f hardens dashboard security boundaries, including scrubbing sensitive fields such as AccessSecret, WebhookURL, WebhookRequestBody, and WebhookHeaders from shared singleton data.
Attack Vector
Exploitation requires network access to the dashboard and valid low-privilege tenant credentials. No user interaction is needed. The attacker iterates server IDs and parses response bodies or status differences to build a map of cross-tenant resources.
// Patch excerpt - cmd/dashboard/controller/ddns.go
// Scrubs sensitive fields from shared singleton data to prevent cross-tenant leakage.
// 不影响 singleton 内原始数据。
for _, p := range ddnsProfiles {
p.AccessSecret = ""
+ p.WebhookURL = ""
+ p.WebhookRequestBody = ""
p.WebhookHeaders = ""
}
Source: GitHub Commit 36b742a
// Patch excerpt - cmd/dashboard/controller/fm.go
// Caps file-manager WebSocket input to the official 1 MiB protocol boundary.
+const fileManagerWebSocketInputLimit int64 = 1024 * 1024
+
+type websocketReadLimiter interface {
+ SetReadLimit(limit int64)
+}
+
+func limitFileManagerWebSocketInput(conn websocketReadLimiter) {
+ conn.SetReadLimit(fileManagerWebSocketInputLimit)
+}
Source: GitHub Commit 36b742a
Detection Methods for CVE-2026-111386
Indicators of Compromise
- High-volume sequential requests to /api/v1/server/config or equivalent getServerConfig and setServerConfig routes from a single authenticated session.
- Repeated permission denied responses interleaved with occasional successful responses tied to the same client.
- Authenticated tenant accounts issuing requests for server IDs that fall outside their known assigned range.
Detection Strategies
- Log every call to getServerConfig and setServerConfig with the requesting tenant, target server ID, and response status, then alert on sweep patterns.
- Baseline normal server-ID access per tenant and flag sessions that query IDs across a wide numeric range in a short window.
- Correlate authentication events with API request volume to surface low-privilege accounts behaving like scanners.
Monitoring Recommendations
- Forward Nezha dashboard access logs to a central SIEM or data lake and retain enough history to spot slow enumeration.
- Add rate limits and anomaly rules on authenticated API endpoints that take resource IDs as parameters.
- Review tenant membership audit logs for recently added members immediately followed by bursts of configuration lookups.
How to Mitigate CVE-2026-111386
Immediate Actions Required
- Upgrade Nezha to version 2.3.15 or later, which contains the fix from commit 36b742a.
- Rotate any AccessSecret, webhook URLs, webhook headers, and webhook request bodies that may have been exposed through shared singleton data.
- Audit tenant memberships and remove accounts that are no longer required to reduce the authenticated attack surface.
Patch Information
The fix is delivered in Nezha 2.3.15 through commit 36b742a7e9a81ed782363dd19efd77b7030fbd0f, which hardens dashboard security boundaries. Refer to the GitHub Security Advisory GHSA-228w-p4h4-6rvw and the VulnCheck Advisory for Nezha for complete remediation details.
Workarounds
- Restrict dashboard network exposure to trusted operators until the upgrade can be applied.
- Place the dashboard behind a reverse proxy that enforces per-user rate limiting on configuration endpoints.
- Temporarily reduce multi-tenant usage by consolidating active server IDs into non-sequential ranges to make enumeration less productive.
# Upgrade Nezha to the patched release
git fetch --tags
git checkout v2.3.15
# Rebuild and restart the dashboard container
docker compose pull && docker compose up -d
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

