Apache Hive Metastore (HMS) uses SerializationUtilities#deserializeObjectWithTypeInformation method when filtering and fetching partitions that is unsafe and can lead to Remote Code Execution (RCE) since it allows the deserialization of arbitrary data.
In real deployments, the vulnerability can be exploited only by authenticated users/clients that were able to successfully establish a connection to the Metastore. From an API perspective any code that calls the unsafe method may be vulnerable unless it performs additional prerechecks on the input arguments.
The product deserializes untrusted data without sufficiently verifying that the resulting data will be valid.
It is often convenient to serialize objects for communication or to save them for later use. However, deserialized data or code can often be modified without using the provided accessor functions if it does not use cryptography to protect itself. Furthermore, any cryptography would still be client-side security – which is a dangerous security assumption. Data that is untrusted can not be trusted to be well-formed. When developers place no restrictions on “gadget chains,” or series of instances and method invocations that can self-execute during the deserialization process (i.e., before the object is returned to the caller), it is sometimes possible for attackers to leverage them to perform unauthorized actions, like generating a shell.