CVE-2026-32595: 
NixOS vulnerability analysis and mitigation

CVE-2026-32595 is a timing attack vulnerability in Traefik's BasicAuth middleware that enables unauthenticated username enumeration. The flaw affects Traefik versions 2.11.40 and below (including v1.x), 3.0.0-beta1 through 3.6.10, and 3.7.0-ea.1. It was disclosed on March 20, 2026, by researcher f1veT via GitHub's security advisory process. The vulnerability carries a CVSS v3.1 score of 3.7 (Low) and a CVSS v4.0 score of 6.3 (Medium) (Github Advisory, Traefik Advisory).

The root cause is classified as CWE-208 (Observable Timing Discrepancy). When a submitted username exists in the BasicAuth configuration, the middleware performs a bcrypt password comparison that takes approximately 166ms due to bcrypt's intentional computational cost. When the username does not exist, the middleware returns an authentication failure immediately in approximately 0.6ms — a ~298x timing difference — because the code performs an early return without executing a dummy bcrypt comparison to normalize response time. This discrepancy is reliably observable over a network connection, allowing an attacker to distinguish valid from invalid usernames by measuring HTTP response latency. A public proof-of-concept Python script was included in the advisory, demonstrating the attack using the requests library and statistics.median() over 20 samples per username (Traefik Advisory).

Successful exploitation allows an unauthenticated attacker to enumerate valid usernames protected by Traefik's BasicAuth middleware, exposing information about the system's user structure. This enumerated data can be leveraged for targeted credential brute-force attacks against confirmed accounts or for social engineering campaigns. There is no direct integrity or availability impact; the confidentiality impact is limited to username disclosure on downstream/subsequent systems (Github Advisory, Traefik Advisory).

A public proof-of-concept script was included in the original security advisory, demonstrating reliable username enumeration using standard HTTP requests and response time measurement. There is no evidence of in-the-wild exploitation or threat actor attribution at this time. The EPSS score is approximately 0.015% (3rd percentile), indicating a low near-term exploitation probability. The vulnerability is not listed in the CISA Known Exploited Vulnerabilities (KEV) catalog (Github Advisory).

1. Reconnaissance: Identify internet-facing services protected by Traefik's BasicAuth middleware. This can be done via Shodan, Censys, or by observing WWW-Authenticate: Basic headers in HTTP 401 responses.
2. Prepare username wordlist: Compile a list of candidate usernames to test (e.g., admin, root, user, operator, application-specific names).
3. Send timed authentication requests: For each candidate username, send multiple HTTP GET requests with a wrong password (e.g., wrongpassword) to the protected endpoint, measuring response time using high-resolution timers.
4. Analyze timing results: Calculate the median response time over ~20 samples per username. Responses taking ~150–200ms indicate the username exists (bcrypt comparison performed); responses returning in ~0.6ms indicate the username does not exist.
5. Classify valid usernames: Apply a threshold (e.g., >50ms median = username exists) to build a confirmed list of valid usernames.
6. Follow-on attack: Use the enumerated valid usernames as targets for password brute-force, credential stuffing, or social engineering attacks (Traefik Advisory).

• Network: High volume of HTTP requests to BasicAuth-protected endpoints with varying usernames but a consistent wrong password; requests originating from a single IP or small IP range in rapid succession.
• Logs: Traefik access logs showing repeated HTTP 401 responses to the same protected path from the same source IP, with response times alternating between ~0.6ms and ~166ms, suggesting automated timing measurement.
• Behavioral: Unusual patterns of authentication failures where the same password is reused across many different usernames — characteristic of username enumeration rather than password guessing (Traefik Advisory).

Traefik has released patched versions that make BasicAuth check timing constant by performing a dummy bcrypt comparison even when the username is not found (commit #12803 by @rtribotte). Organizations should upgrade to v2.11.41 (for the 2.x branch), v3.6.11 (for the 3.x branch), or v3.7.0-ea.2 (for the 3.7 early access branch). As a temporary workaround while patching, implement network-level rate limiting and IP-based throttling on authentication endpoints to reduce the feasibility of timing measurements (Traefik v2.11.41 Release, Traefik v3.6.11 Release, Traefik v3.7.0-ea.2 Release).

Traefik published an official security advisory and community forum post announcing the fix across all affected branches simultaneously (Traefik Community). Red Hat tracked the vulnerability via Bugzilla and published a CVE page, indicating relevance to downstream distributions (Red Hat CVE). OpenSUSE issued security announcements for updated Traefik packages, and the vulnerability was picked up by Linux security news aggregators (openSUSE Announce).