A remote code injection vulnerability exists in the Ambari Metrics and
AMS Alerts feature, allowing authenticated users to inject and execute
arbitrary code. The vulnerability occurs when processing alert
definitions, where malicious input can be injected into the alert script
execution path. An attacker with authenticated access can exploit this
vulnerability to execute arbitrary commands on the server. The issue has
been fixed in the latest versions of Ambari.
An XML External Entity (XXE) vulnerability exists in the Ambari/Oozie
project, allowing an attacker to inject malicious XML entities. This
vulnerability occurs due to insecure parsing of XML input using the
`DocumentBuilderFactory` class without disabling external entity
resolution. An attacker can exploit this vulnerability to read arbitrary
files on the server or perform server-side request forgery (SSRF)
attacks. The issue has been fixed in both Ambari 2.7.9 and the trunk
branch.
A code injection vulnerability exists in the Ambari Alert Definition
feature, allowing authenticated users to inject and execute arbitrary
shell commands. The vulnerability arises when defining alert scripts,
where the script filename field is executed using `sh -c`. An attacker
with authenticated access can exploit this vulnerability to inject
malicious commands, leading to remote code execution on the server. The
issue has been fixed in the latest versions of Ambari.
Malicious code injection in Apache Ambari in prior to 2.7.8. Users are recommended to upgrade to version 2.7.8, which fixes this issue.
Impact:
A Cluster Operator can manipulate the request by adding a malicious code injection and gain a root over the cluster main host.
In Apache Ambari versions 2.6.2.2 and earlier, malicious users can construct file names for directory traversal and traverse to other directories to download files.
Apache Ambari, versions 1.4.0 to 2.6.1, is susceptible to a directory traversal attack allowing an unauthenticated user to craft an HTTP request which provides read-only access to any file on the filesystem of the host the Ambari Server runs on that is accessible by the user the Ambari Server is running as. Direct network access to the Ambari Server is required to issue this request, and those Ambari Servers that are protected behind a firewall, or in a restricted network zone are at less risk of being affected by this issue.
In Ambari 1.2.0 through 2.2.2, it may be possible to execute arbitrary system commands on the Ambari Server host while generating SSL certificates for hosts in an Ambari cluster.
The File Browser View in Apache Ambari before 2.2.1 allows remote authenticated administrators to read arbitrary files via a file: URL in the WebHDFS URL configuration.
The agent in Apache Ambari before 2.1.2 uses weak permissions for the (1) /var/lib/ambari-agent/data and (2) /var/lib/ambari-agent/keys directories, which allows local users to obtain sensitive information by reading files in the directories.