MessagePack allows untrusted data to lead to DoS attack due to hash collisions and stack overflow

Impact

When this library is used to deserialize messagepack data from an untrusted source, there is a risk of a denial of service attack by an attacker that sends data contrived to produce hash collisions, leading to large CPU consumption disproportionate to the size of the data being deserialized.

This is similar to a prior advisory, which provided an inadequate fix for the hash collision part of the vulnerability.

Patches

The following steps are required to mitigate this risk.

1. Upgrade to a version of the library where a fix is available. If upgrading from v1, check out our migration guide.
2. Review the steps in this previous advisory to ensure you have your application configured for untrusted data.

Workarounds

If upgrading MessagePack to a patched version is not an option for you, you may apply a manual workaround as follows:

1. Declare a class that derives from MessagePackSecurity.
2. Override the GetHashCollisionResistantEqualityComparer<T> method to provide a collision-resistant hash function of your own and avoid calling base.GetHashCollisionResistantEqualityComparer<T>().
3. Configure a MessagePackSerializerOptions with an instance of your derived type by calling WithSecurity on an existing options object.
4. Use your custom options object for all deserialization operations. This may be by setting the MessagePackSerializer.DefaultOptions static property, if you call methods that rely on this default property, and/or by passing in the options object explicitly to any Deserialize method.

References

• Learn more about best security practices when reading untrusted data with MessagePack 1.x or MessagePack 2.x.
• The .NET team's discussion on hash collision vulnerabilities of their HashCode struct (or in the pull request that merges this into the dotnet org).

For more information

If you have any questions or comments about this advisory:

• Start a public discussion
• Email us privately