Impact Prior to puma version 5.6.2, puma may not always call close on the response body. Rails, prior to version 7.0.2.2, depended on the response body being closed in order for its CurrentAttributes implementation to work correctly.
From Rails:
Under certain circumstances response bodies will not be closed, for example a bug in a webserver[1] or a bug in a Rack middleware. In the event a response is not notified of a close, ActionDispatch::Executor will not know to reset thread local state for the next request. This can lead to data being leaked to subsequent requests, especially when interacting with ActiveSupport::CurrentAttributes.
The combination of these two behaviors (Puma not closing the body + Rails' Executor implementation) causes information leakage.
Patches This problem is fixed in Puma versions 5.6.2 and 4.3.11.
This problem is fixed in Rails versions 7.02.2, 6.1.4.6, 6.0.4.6, and 5.2.6.2.
See: https://github.com/advisories/GHSA-wh98-p28r-vrc9 for details about the rails vulnerability
Upgrading to a patched Rails or Puma version fixes the vulnerability.
Workarounds
Upgrade to Rails versions 7.02.2, 6.1.4.6, 6.0.4.6, and 5.2.6.2.
The Rails CVE includes a middleware that can be used instead.
References
Rails CVE: CVE-2022-23633
For more information If you have any questions or comments about this advisory: Open an issue in puma See our security policy
Puma is a concurrent HTTP 1.1 server for Ruby/Rack applications. The fix for CVE-2019-16770 was incomplete. The original fix only protected existing connections that had already been accepted from having their requests starved by greedy persistent-connections saturating all threads in the same process. However, new connections may still be starved by greedy persistent-connections saturating all threads in all processes in the cluster. A puma server which received more concurrent keep-alive connections than the server had threads in its threadpool would service only a subset of connections, denying service to the unserved connections. This problem has been fixed in puma 4.3.8 and 5.3.1. Setting queuerequests false also fixes the issue. This is not advised when using puma without a reverse proxy, such as nginx or apache, because you will open yourself to slow client attacks (e.g. slowloris). The fix is very small and a git patch is available for those using unsupported versions of Puma.
Impact
Prior to puma version 5.5.0, using puma with a proxy which forwards LF characters as line endings could allow HTTP request smuggling. A client could smuggle a request through a proxy, causing the proxy to send a response back to another unknown client.
This behavior (forwarding LF characters as line endings) is very uncommon amongst proxy servers, so we have graded the impact here as "low". Puma is only aware of a single proxy server which has this behavior.
If the proxy uses persistent connections and the client adds another request in via HTTP pipelining, the proxy may mistake it as the first request's body. Puma, however, would see it as two requests, and when processing the second request, send back a response that the proxy does not expect. If the proxy has reused the persistent connection to Puma to send another request for a different client, the second response from the first client will be sent to the second client.
Patches
This vulnerability was patched in Puma 5.5.1 and 4.3.9.
Workarounds
This vulnerability only affects Puma installations without any proxy in front.
Use a proxy which does not forward LF characters as line endings.
Proxies which do not forward LF characters as line endings:
Nginx Apache (>2.4.25) Haproxy Caddy Traefik
Possible Breakage
If you are dealing with legacy clients that want to send LF as a line ending in an HTTP header, this will cause those clients to receive a 400 error.
References
HTTP Request Smuggling
For more information
If you have any questions or comments about this advisory:
Open an issue in Puma See our security policy