Apache : Security Vulnerabilities, CVEs, Published In 2017

CouchDB administrative users can configure the database server via HTTP(S). Some of the configuration options include paths for operating system-level binaries that are subsequently launched by CouchDB. This allows an admin user in Apache CouchDB before 1.7.0 and 2.x before 2.1.1 to execute arbitrary shell commands as the CouchDB user, including downloading and executing scripts from the public internet.

Due to differences in the Erlang-based JSON parser and JavaScript-based JSON parser, it is possible in Apache CouchDB before 1.7.0 and 2.x before 2.1.1 to submit _users documents with duplicate keys for 'roles' used for access control within the database, including the special case '_admin' role, that denotes administrative users. In combination with CVE-2017-12636 (Remote Code Execution), this can be used to give non-admin users access to arbitrary shell commands on the server as the database system user. The JSON parser differences result in behaviour that if two 'roles' keys are available in the JSON, the second one will be used for authorising the document write, but the first 'roles' key is used for subsequent authorization for the newly created user. By design, users can not assign themselves roles. The vulnerability allows non-admin users to give themselves admin privileges.

When running Apache Tomcat versions 9.0.0.M1 to 9.0.0, 8.5.0 to 8.5.22, 8.0.0.RC1 to 8.0.46 and 7.0.0 to 7.0.81 with HTTP PUTs enabled (e.g. via setting the readonly initialisation parameter of the Default servlet to false) it was possible to upload a JSP file to the server via a specially crafted request. This JSP could then be requested and any code it contained would be executed by the server.

When running Apache Tomcat 7.0.0 to 7.0.79 on Windows with HTTP PUTs enabled (e.g. via setting the readonly initialisation parameter of the Default to false) it was possible to upload a JSP file to the server via a specially crafted request. This JSP could then be requested and any code it contained would be executed by the server.

The REST Plugin in Apache Struts 2.1.1 through 2.3.x before 2.3.34 and 2.5.x before 2.5.13 uses an XStreamHandler with an instance of XStream for deserialization without any type filtering, which can lead to Remote Code Execution when deserializing XML payloads.

Apache httpd allows remote attackers to read secret data from process memory if the Limit directive can be set in a user's .htaccess file, or if httpd.conf has certain misconfigurations, aka Optionsbleed. This affects the Apache HTTP Server through 2.2.34 and 2.4.x through 2.4.27. The attacker sends an unauthenticated OPTIONS HTTP request when attempting to read secret data. This is a use-after-free issue and thus secret data is not always sent, and the specific data depends on many factors including configuration. Exploitation with .htaccess can be blocked with a patch to the ap_limit_section function in server/core.c.

The Struts 1 plugin in Apache Struts 2.1.x and 2.3.x might allow remote code execution via a malicious field value passed in a raw message to the ActionMessage.

The Jakarta Multipart parser in Apache Struts 2 2.3.x before 2.3.32 and 2.5.x before 2.5.10.1 has incorrect exception handling and error-message generation during file-upload attempts, which allows remote attackers to execute arbitrary commands via a crafted Content-Type, Content-Disposition, or Content-Length HTTP header, as exploited in the wild in March 2017 with a Content-Type header containing a #cmd= string.

The code in Apache Tomcat 9.0.0.M1 to 9.0.0.M11, 8.5.0 to 8.5.6, 8.0.0.RC1 to 8.0.38, 7.0.0 to 7.0.72, and 6.0.0 to 6.0.47 that parsed the HTTP request line permitted invalid characters. This could be exploited, in conjunction with a proxy that also permitted the invalid characters but with a different interpretation, to inject data into the HTTP response. By manipulating the HTTP response the attacker could poison a web-cache, perform an XSS attack and/or obtain sensitive information from requests other then their own.

In Apache Synapse, by default no authentication is required for Java Remote Method Invocation (RMI). So Apache Synapse 3.0.1 or all previous releases (3.0.0, 2.1.0, 2.0.0, 1.2, 1.1.2, 1.1.1) allows remote code execution attacks that can be performed by injecting specially crafted serialized objects. And the presence of Apache Commons Collections 3.2.1 (commons-collections-3.2.1.jar) or previous versions in Synapse distribution makes this exploitable. To mitigate the issue, we need to limit RMI access to trusted users only. Further upgrading to 3.0.1 version will eliminate the risk of having said Commons Collection version. In Synapse 3.0.1, Commons Collection has been updated to 3.2.2 version.

In Apache Struts 2.5 to 2.5.14, the REST Plugin is using an outdated JSON-lib library which is vulnerable and allow perform a DoS attack using malicious request with specially crafted JSON payload.

In Apache Qpid Broker-J 0.18 through 0.32, if the broker is configured with different authentication providers on different ports one of which is an HTTP port, then the broker can be tricked by a remote unauthenticated attacker connecting to the HTTP port into using an authentication provider that was configured on a different port. The attacker still needs valid credentials with the authentication provider on the spoofed port. This becomes an issue when the spoofed port has weaker authentication protection (e.g., anonymous access, default accounts) and is normally protected by firewall rules or similar which can be circumvented by this vulnerability. AMQP ports are not affected. Versions 6.0.0 and newer are not affected.

In Apache Qpid Broker-J versions 6.1.0 through 6.1.4 (inclusive) the broker does not properly enforce a maximum frame size in AMQP 1.0 frames. A remote unauthenticated attacker could exploit this to cause the broker to exhaust all available memory and eventually terminate. Older AMQP protocols are not affected.

A flaw in the org.apache.sling.auth.core.AuthUtil#isRedirectValid method in Apache Sling Authentication Service 1.4.0 allows an attacker, through the Sling login form, to trick a victim to send over their credentials.

The camel-castor component in Apache Camel 2.x before 2.19.4 and 2.20.x before 2.20.1 is vulnerable to Java object de-serialisation vulnerability. De-serializing untrusted data can lead to security flaws.

The camel-hessian component in Apache Camel 2.x before 2.19.4 and 2.20.x before 2.20.1 is vulnerable to Java object de-serialisation vulnerability. De-serializing untrusted data can lead to security flaws.

Apache CXF Fediz ships with a number of container-specific plugins to enable WS-Federation for applications. A CSRF (Cross Style Request Forgery) style vulnerability has been found in the Spring 2, Spring 3 and Spring 4 plugins in versions before 1.4.3 and 1.3.3. The vulnerability can result in a security context that is set up using a malicious client's roles for the given enduser.

In Apache Drill 1.11.0 and earlier when submitting form from Query page users are able to pass arbitrary script or HTML which will take effect on Profile page afterwards. Example: after submitting special script that returns cookie information from Query page, malicious user may obtain this information from Profile page afterwards.

Remote code execution occurs in Apache Solr before 7.1 with Apache Lucene before 7.1 by exploiting XXE in conjunction with use of a Config API add-listener command to reach the RunExecutableListener class. Elasticsearch, although it uses Lucene, is NOT vulnerable to this. Note that the XML external entity expansion vulnerability occurs in the XML Query Parser which is available, by default, for any query request with parameters deftype=xmlparser and can be exploited to upload malicious data to the /upload request handler or as Blind XXE using ftp wrapper in order to read arbitrary local files from the Solr server. Note also that the second vulnerability relates to remote code execution using the RunExecutableListener available on all affected versions of Solr.

The JMX server embedded in Apache James, also used by the command line client is exposed to a java de-serialization issue, and thus can be used to execute arbitrary commands. As James exposes JMX socket by default only on local-host, this vulnerability can only be used for privilege escalation. Release 3.0.1 upgrades the incriminated library.

Apache Hive 2.1.x before 2.1.2, 2.2.x before 2.2.1, and 2.3.x before 2.3.1 expose an interface through which masking policies can be defined on tables or views, e.g., using Apache Ranger. When a view is created over a given table, the policy enforcement does not happen correctly on the table for masked columns.

Apache CXF supports sending and receiving attachments via either the JAX-WS or JAX-RS specifications. It is possible to craft a message attachment header that could lead to a Denial of Service (DoS) attack on a CXF web service provider. Both JAX-WS and JAX-RS services are vulnerable to this attack. From Apache CXF 3.2.1 and 3.1.14, message attachment headers that are greater than 300 characters will be rejected by default. This value is configurable via the property "attachment-max-header-size".

An authorized user could upload a template which contained malicious code and accessed sensitive files via an XML External Entity (XXE) attack. The fix to properly handle XML External Entities was applied on the Apache NiFi 1.4.0 release. Users running a prior 1.x release should upgrade to the appropriate release.

During Jelly (xml) file parsing with Apache Xerces, if a custom doctype entity is declared with a "SYSTEM" entity with a URL and that entity is used in the body of the Jelly file, during parser instantiation the parser will attempt to connect to said URL. This could lead to XML External Entity (XXE) attacks in Apache Commons Jelly before 1.0.1.

When loading models or dictionaries that contain XML it is possible to perform an XXE attack, since Apache OpenNLP is a library, this only affects applications that load models or dictionaries from untrusted sources. The versions 1.5.0 to 1.5.3, 1.6.0, 1.7.0 to 1.7.2, 1.8.0 to 1.8.1 of Apache OpenNLP are affected.