Hackers Scan for Vulnerable Citrix ADC Systems
Weeks after Citrix revealed a critical vulnerability impacting its Application Delivery Controller (ADC) and Gateway products, hackers have started to scan the Internet for vulnerable systems, security researchers report.
Tracked as CVE-2019-19781 and featuring a CVSS score of 9.8, the vulnerability has existed since 2014. Exploitation could result in attackers gaining unauthorized access to internal network resources and executing arbitrary code.
“All supported versions of the product, and all supported platforms” are impacted by the security flaw, Positive Technologies, the company that discovered the bug, says. The affected products were previously known as NetScaler ADC and NetScaler Gateway.
On December 17, Citrix published its own security bulletin on this vulnerability, but hasn’t released a patch as of now. However, the company did publish a support article detailing configuration changes that organizations should make to mitigate the impact of this vulnerability.
Now, only three weeks later, security researchers are already observing online scans for vulnerable systems, as well as exploitation attempts targeting CVE-2019-19781.
“In my Citrix ADC honeypot, CVE-2019-19781 is being probed with attackers reading sensitive credential config files remotely using ../ directory traversal (a variant of this issue). So this is in the wild, active exploitation starting up,” security researcher Kevin Beaumont said on Twitter.
Johannes B. Ullrich, dean of research at the SANS Technology Institute, who also noticed that people are scanning for vulnerable systems, says the scans do not appear sophisticated. The source IPs scanning his honeypot are located on the networks of China Mobile, BACloud (Europe/US colocation servers), CHINA UNICOM Shanghai city network, and OpenIP (France DSL).
Based on the mitigation steps advised by Citrix, exploit attempts are expected to involve a path traversal likely related to the /vpns/ path on Citrix systems, Tripwire security researcher Craig Young notes. In an Apache configuration on an ADC appliance, this path is home to some Perl content, including a directory without an authorization requirement, he discovered.
During his investigation, the researcher learned that it was possible to load a crafted XML file through the Perl Template Toolkit, which could allow an attacker to enumerate usernames and crawl directories to reveal active administrative session IDs.
Young also discovered that the Template Toolkit had an ‘undocumented feature’ that allowed arbitrary command execution when processing a crafted directive, and says that the complete exploit chain only requires two HTTPS requests.
The researcher attempted to determine the number of affected organizations, by looking into how many NetScaler VPN portals are exposed and how many of them have yet to enable the suggested mitigation.
Initially, 80,000 organizations were said to be vulnerable, but Young discovered only 58,620 IP addresses of “likely” NetScaler or ADC VPN portals. 21 days after Citrix published its advisory, less than a third of the exposed appliances had the mitigation enabled, with 39,378 of them remaining vulnerable.
“To put this in perspective, I correlated the IP addresses with their certificate data and found more than 26,000 unique subject common name values. The list contains countless high value targets across a swath of verticals including finance, government, and healthcare,” Young notes.
A total of 141 distinct domain names ending in .gov, along with 351 other distinct names containing .gov. in the domain were discovered, from 24 different ccTLD, but predominantly from *.gov.uk and *.gov.au.
Organizations are advised to apply the suggested mitigation as soon as possible to avoid further exposure.