Apache Karaf allows monitoring of applications and the Java runtime by using the Java Management Extensions (JMX). JMX is a Java RMI based technology that relies on Java serialized objects for client server communication. Whereas the default JMX implementation is hardened against unauthenticated deserialization attacks, the implementation used by Apache Karaf is not protected against this kind of attack. The impact of Java deserialization vulnerabilities strongly depends on the classes that are available within the targets class path. Generally speaking, deserialization of untrusted data does always represent a high security risk and should be prevented. The risk is low as, by default, Karaf uses a limited set of classes in the JMX server class path. It depends of system scoped classes (e.g. jar in the lib folder).

Affected: Maven: org.apache.karaf.management:org.apache.karaf.management.server

Apache Karaf allows monitoring of applications and the Java runtime by using the Java Management Extensions (JMX). JMX is a Java RMI based technology that relies on Java serialized objects for client server communication. Whereas the default JMX implementation is hardened against unauthenticated deserialization attacks, the implementation used by Apache Karaf is not protected against this kind of attack. The impact of Java deserialization vulnerabilities strongly depends on the classes that are available within the targets class path. Generally speaking, deserialization of untrusted data does always represent a high security risk and should be prevented. The risk is low as, by default, Karaf uses a limited set of classes in the JMX server class path. It depends of system scoped classes (e.g. jar in the lib folder).

elasticsearch: not properly preserving security permissions when executing complex queries may lead to information disclosure tomcat: deserialization flaw in session persistence storage leading to RCE junit4: TemporaryFolder is shared between all users across system which could result in information disclosure wildfly-core: memory leak in WildFly host-controller in domain mode while not able to reconnect to domain-controller kotlin: vulnerable Java API was used for temporary file and folder creation which could result in information disclosure jackson-databind: denial of service via a large depth of nested objects mysql-connector-java: unauthorized access to critical undertow: potential security issue in flow control over HTTP/2 may lead to DOS wildfly-elytron: possible timing attack in ScramServer wildfly-core: Invalid Sensitivity Classification of Vault Expression nodejs-ansi-regex: Regular expression denial of service (ReDoS) matching ANSI escape codes undertow: client side invocation timeout raised when calling over HTTP2 kubernetes-client: Insecure deserialization in unmarshalYaml method springframework: Additional Log Injection in Spring Framework (follow-up to CVE-2021-22096) springframework: malicious input leads to insertion of additional log entries spring-security: Denial-of-Service (DoS) attack via initiation of Authorization Request protobuf-java: potential DoS in the parsing procedure for binary data google-oauth-client: Token signature not verified tomcat: Inf…

Ein entfernter, anonymer, authentisierter oder lokaler Angreifer kann mehrere Schwachstellen in Red Hat FUSE ausnutzen, um vertrauliche Informationen offenzulegen, beliebigen Code auszuführen, einen Denial of Service Zustand herbeizuführen, Sicherheitsmaßnahmen zu umgehen, Daten und Informationen zu manipulieren und seine Privilegien zu erweitern.