This article is a part of our Vulnerability Database (back to index)
Path Traversal occurrences in Http4s
Http4s is a Scala interface for HTTP services. `StaticFile.fromUrl` can leak the presence of a directory on a server when the `URL` scheme is not `file://`, and the URL points to a fetchable resource under its scheme and authority. The function returns `F[None]`, indicating no resource, if `url.getFile` is a directory, without first checking the scheme or authority of the URL. If a URL connection to the scheme and URL would return a stream, and the path in the URL exists as a directory on the server, the presence of the directory on the server could be inferred from the 404 response. The contents and other metadata about the directory are not exposed. This affects http4s versions: 0.21.7 through 0.21.23, 0.22.0-M1 through 0.22.0-M8, 0.23.0-M1, and 1.0.0-M1 through 1.0.0-M22. The [patch](https://github.com/http4s/http4s/commit/52e1890665410b4385e37b96bc49c5e3c708e4e9) is available in the following versions: v0.21.24, v0.22.0-M9, v0.23.0-M2, v1.0.0-M23. As a workaround users can avoid calling `StaticFile.fromUrl` with non-file URLs. (2021-05-27, CVE-2021-32643)
http4s before versions 0.18.26, 0.20.20, and 0.21.2 has a local file inclusion vulnerability. This vulnerability applies to all users of org.http4s.server.staticcontent.FileService, org.http4s.server.staticcontent.ResourceService and org.http4s.server.staticcontent.WebjarService. URI normalization is applied incorrectly. Requests whose path info contain ../ or // can expose resources outside of the configured location. This issue is patched in versions 0.18.26, 0.20.20, and 0.21.2. Note that 0.19.0 is a deprecated release and has never been supported. (2020-03-25, CVE-2020-5280)
Why Path Traversal can be dangerous
Relative Path Confusion means that your web server is configured to serve responses to ambiguous URLs. This configuration can possibly cause confusion about the correct relative path for the URL. It is also an issue of resources, such as images, styles etc., which are specified in the response using relative path, not the absolute URL.
If the web browser permits to parse "cross-content" response, the attacker may be able to fool the web browser into interpreting HTML into other content types, which can then lead to a cross site scripting attack (link do XSS).