Package Instance

Circumvention of file size limits in ActiveStorage
There is a vulnerability in ActiveStorage's S3 adapter that allows the Content-Length of a direct file upload to be modified by an end user.

Versions Affected:  rails < 5.2.4.2, rails < 6.0.3.1
Not affected:       Applications that do not use the direct upload functionality of the ActiveStorage S3 adapter.
Fixed Versions:     rails >= 5.2.4.3, rails >= 6.0.3.1

Impact
------

Utilizing this vulnerability, an attacker can control the Content-Length of an S3 direct upload URL without receiving a new signature from the server. This could be used to bypass controls in place on the server to limit upload size.

Workarounds
-----------

This is a low-severity security issue. As such, no workaround is necessarily until such time as the application can be upgraded.

Untrusted users can run pending migrations in production in Rails
There is a vulnerability in versions of Rails prior to 6.0.3.2 that allowed an untrusted user to run any pending migrations on a Rails app running in production.

This vulnerability has been assigned the CVE identifier CVE-2020-8185.

Versions Affected:  6.0.0 < rails < 6.0.3.2
Not affected:       Applications with `config.action_dispatch.show_exceptions = false` (this is not a default setting in production)
Fixed Versions:     rails >= 6.0.3.2

Impact
------

Using this issue, an attacker would be able to execute any migrations that are pending for a Rails app running in production mode. It is important to note that an attacker is limited to running migrations the application developer has already defined in their application and ones that have not already run.

Workarounds
-----------

Until such time as the patch can be applied, application developers should disable the ActionDispatch middleware in their production environment via a line such as this one in their config/environment/production.rb:

`config.middleware.delete ActionDispatch::ActionableExceptions`

Ability to forge per-form CSRF tokens in Rails
It is possible to, given a global CSRF token such as the one present in the authenticity_token meta tag, forge a per-form CSRF token for any action for that session.

Impact
------

Given the ability to extract the global CSRF token, an attacker would be able to construct a per-form CSRF token for that session.

Workarounds
-----------

This is a low-severity security issue. As such, no workaround is necessarily until such time as the application can be upgraded.

Possible Strong Parameters Bypass in ActionPack
There is a strong parameters bypass vector in ActionPack.

Versions Affected:  rails <= 6.0.3
Not affected:       rails < 5.0.0
Fixed Versions:     rails >= 5.2.4.3, rails >= 6.0.3.1

Impact
------
In some cases user supplied information can be inadvertently leaked from
Strong Parameters.  Specifically the return value of `each`, or `each_value`,
or `each_pair` will return the underlying "untrusted" hash of data that was
read from the parameters.  Applications that use this return value may be
inadvertently use untrusted user input.

Impacted code will look something like this:

```
def update
  # Attacker has included the parameter: `{ is_admin: true }`
  User.update(clean_up_params)
end

def clean_up_params
   params.each { |k, v|  SomeModel.check(v) if k == :name }
end
```

Note the mistaken use of `each` in the `clean_up_params` method in the above
example.

Workarounds
-----------
Do not use the return values of `each`, `each_value`, or `each_pair` in your
application.

rack-protection gem timing attack vulnerability when validating CSRF token
Sinatra rack-protection versions 1.5.4 and 2.0.0.rc3 and earlier contains a timing attack vulnerability in the CSRF token checking that can result in signatures can be exposed. This attack appear to be exploitable via network connectivity to the ruby application. This vulnerability appears to have been fixed in 1.5.5 and 2.0.0.

Possible Information Leak / Session Hijack Vulnerability in Rack
There's a possible information leak / session hijack vulnerability in Rack. Attackers may be able to find and hijack sessions by using timing attacks targeting the session id. Session ids are usually stored and indexed in a database that uses some kind of scheme for speeding up lookups of that session id. By carefully measuring the amount of time it takes to look up a session, an attacker may be able to find a valid session id and hijack the session.

The session id itself may be generated randomly, but the way the session is indexed by the backing store does not use a secure comparison.

### Impact

The session id stored in a cookie is the same id that is used when querying the backing session storage engine.  Most storage mechanisms (for example a database) use some sort of indexing in order to speed up the lookup of that id.  By carefully timing requests and session lookup failures, an attacker may be able to perform a timing attack to determine an existing session id and hijack that session.

## Releases

The 1.6.12 and 2.0.8 releases are available at the normal locations.

### Workarounds

There are no known workarounds.

### Patches

To aid users who aren't able to upgrade immediately we have provided patches for
the two supported release series. They are in git-am format and consist of a
single changeset.

* 1-6-session-timing-attack.patch - Patch for 1.6 series
* 2-0-session-timing-attack.patch - Patch for 2.6 series

### Credits

Thanks Will Leinweber for reporting this!

XSS in Action View
There is a potential Cross-Site Scripting (XSS) vulnerability in Action View's translation helpers. Views that allow the user to control the default (not found) value of the `t` and `translate` helpers could be susceptible to XSS attacks.

### Impact

When an HTML-unsafe string is passed as the default for a missing translation key [named `html` or ending in `_html`](https://guides.rubyonrails.org/i18n.html#using-safe-html-translations), the default string is incorrectly marked as HTML-safe and not escaped. Vulnerable code may look like the following examples:

```erb
<%# The welcome_html translation is not defined for the current locale: %>
<%= t("welcome_html", default: untrusted_user_controlled_string) %>

<%# Neither the title.html translation nor the missing.html translation is defined for the current locale: %>
<%= t("title.html", default: [:"missing.html", untrusted_user_controlled_string]) %>
```

### Patches

Patched Rails versions, 6.0.3.3 and 5.2.4.4, are available from the normal locations.

The patches have also been applied to the `master`, `6-0-stable`, and `5-2-stable` branches on GitHub. If you track any of these branches, you should update to the latest.

To aid users who aren’t able to upgrade immediately, we’ve provided patches for the two supported release series. They are in git-am format and consist of a single changeset.

* [5-2-translate-helper-xss.patch](https://gist.github.com/georgeclaghorn/a466e103922ee81f24c32c9034089442#file-5-2-translate-helper-xss-patch) — patch for the 5.2 release series
* [6-0-translate-helper-xss.patch](https://gist.github.com/georgeclaghorn/a466e103922ee81f24c32c9034089442#file-6-0-translate-helper-xss-patch) — patch for the 6.0 release series

Please note that only the 5.2 and 6.0 release series are currently supported. Users of earlier, unsupported releases are advised to update as soon as possible, as we cannot provide security fixes for unsupported releases.

### Workarounds

Impacted users who can’t upgrade to a patched Rails version can avoid this issue by manually escaping default translations with the `html_escape` helper (aliased as `h`):

```erb
<%= t("welcome_html", default: h(untrusted_user_controlled_string)) %>
```

ActiveSupport potentially unintended unmarshalling of user-provided objects in MemCacheStore and RedisCacheStore
In ActiveSupport, there is potentially unexpected behaviour in the MemCacheStore and RedisCacheStore where, when
untrusted user input is written to the cache store using the `raw: true` parameter, re-reading the result
from the cache can evaluate the user input as a Marshalled object instead of plain text. Vulnerable code looks like:

```
data = cache.fetch("demo", raw: true) { untrusted_string }
```
Versions Affected:  rails < 5.2.5, rails < 6.0.4
Not affected:       Applications not using MemCacheStore or RedisCacheStore. Applications that do not use the `raw` option when storing untrusted user input.
Fixed Versions:     rails >= 5.2.4.3, rails >= 6.0.3.1
  
Impact
------
Unmarshalling of untrusted user input can have impact up to and including RCE. At a minimum,
this vulnerability allows an attacker to inject untrusted Ruby objects into a web application.
In addition to upgrading to the latest versions of Rails, developers should ensure that whenever
they are calling `Rails.cache.fetch` they are using consistent values of the `raw` parameter for both
reading and writing, especially in the case of the RedisCacheStore which does not, prior to these changes,
detect if data was serialized using the raw option upon deserialization.

Workarounds
-----------
It is recommended that application developers apply the suggested patch or upgrade to the latest release as
soon as possible. If this is not possible, we recommend ensuring that all user-provided strings cached using
the `raw` argument should be double-checked to ensure that they conform to the expected format.

rest-client Gem Vulnerable to Session Fixation
REST client for Ruby (aka rest-client) versions 1.6.1.a until 1.8.0 allow remote attackers to conduct session fixation attacks or obtain sensitive cookie information by leveraging passage of cookies set in a response to a redirect.

CSRF Vulnerability in rails-ujs
There is a vulnerability in rails-ujs that allows attackers to send CSRF tokens to wrong domains.

Versions Affected:  rails <= 6.0.3
Not affected:       Applications which don't use rails-ujs.
Fixed Versions:     rails >= 5.2.4.3, rails >= 6.0.3.1

Impact
------

This is a regression of CVE-2015-1840.

In the scenario where an attacker might be able to control the href attribute of an anchor tag or the action attribute of a form tag that will trigger a POST action, the attacker can set the href or action to a cross-origin URL, and the CSRF token will be sent.

Workarounds
-----------

To work around this problem, change code that allows users to control the href attribute of an anchor tag or the action attribute of a form tag to filter the user parameters.

For example, code like this:

    link_to params

to code like this:

    link_to filtered_params

    def filtered_params
      # Filter just the parameters that you trust
    end