PYSEC-2026-3594

Open WebUI: Account enumeration via observable login timing discrepancy

### Summary The `/api/v1/auths/signin` endpoint leaked whether an email address belonged to a registered account through a response-time side channel. Password verification ran bcrypt only when the email was found in the database; for a non-existent email the request returned early without hashing. The expensive bcrypt comparison therefore made valid-account attempts respond significantly slower (~180 ms) than non-existent ones (~5 ms), so an unauthenticated attacker could enumerate valid accounts by measuring response time. ### Details On signin the backend looked the user up by email and only performed the bcrypt password comparison if a record existed. A missing email short-circuited before any hashing, producing the timing gap. The built-in brute-force throttling did not prevent it: sending one request at a time with a small delay between requests stays under the rate limit while still exposing the difference. Observed in the reporter's run (HTTP 400 for every attempt, the response time is the signal): ``` Email Status Response time joe@example.com 400 186 ms <- valid account larry@example.com 400 9 ms jose@example.com 400 6 ms james@example.com 400 5 ms ``` ### Impact An unauthenticated attacker can enumerate which email addresses are registered accounts, which enables targeted password-spraying against confirmed accounts. The impact is amplified by MFA not being enabled by default. No data is read or modified; the disclosure is limited to account existence. ### Patched The authentication path now runs a bcrypt verification against a constant placeholder hash whenever the email does not resolve to an active credential, so a real hash comparison executes on every attempt and the response time is the same whether or not the account exists. Fixed in 0.10.0. ### Credits @dievus