Skip to content
4 Microsoft zero-day vulnerabilities in the Azure cloud platform’s Open Management Infrastructure (OMI) — a software that many don’t know is embedded in a variety of services — show that OMI represents a significant security blind spot, researchers have commented.
Dubbed OMIGOD, a series of vulnerabilities in the Open Management Infrastructure used in Azure on Linux demonstrate hidden security threats, researchers observed.
OMIGOD
Collectively dubbed “OMIGOD” because of the name & the reaction of the researchers who discovered them, the flaws — which were zero-day when found — affect 1,000s of Azure customers & millions of endpoints, according to a blog post published this week by cloud infrastructure security firm Wiz.
Though Microsoft patched them this week in its monthly Patch Tues. raft of updates, their presence in OMI highlights the risk for the supply chain when companies unknowingly run code — particularly open-source code — on their systems that allows for exploitation, researchers observed.
Undetected Flaws
Recent high-profile supply-chain attacks such as SolarWinds & Kaseya demonstrate how much damage can be done when undetected flaws in 3rd-party software that organisations use in larger systems are exploited.
“One of the biggest challenges in preventing them is that our digital supply chain is not transparent,” Senior Security Researcher Nir Ohfeld wrote in the Wiz post. “If you don’t know what’s hidden in the services & products you use every day, how can you manage the risk?
The OMIGOD vulnerabilities discovered by Ohfeld & his colleagues present a security danger to potentially millions of unsuspecting customers of cloud computing services, he stated.
Azure Tenants
“In a small sample of Azure tenants we analysed, over 65% of Azure customers were unknowingly at risk,” Ohfeld wrote.
The vulnerabilities that Wiz researchers discovered include 1 that allows for remote code execution (RCE), CVE-2021-38647. The other 3 are privilege-escalation vulnerabilities (CVE-2021-38648, CVE-2021-38645 & CVE-2021-38649) of lower risk but which are critical for a full attack chain.
“Unless a patch is applied, attackers can easily exploit these 4 vulnerabilities to escalate to root privileges & remotely execute malicious code – for instance, encrypting files for ransom,” Ohfeld cautioned.
Hidden Cloud Security Danger
One reason for the significant alarm over the flaws is that they are found in OMI, an agent automatically deployed when customers set up a Linux virtual machine (VM) in their cloud & enable certain Azure services, researchers explained.
“This happens without customers’ explicit consent or knowledge,” Ohfeld wrote. “Users simply click ‘agree’ to log collection during setup, & they have unknowingly opted in.”
Highest Privileges
OMI is a perilous attack surface because Azure provides “virtually no public documentation” about it, he explained. That means most customers have never heard of it & are unaware that it even exists as an exploitable entity in their deployment.
Moreover, the OMI agent runs as root with the highest privileges, so any user can communicate with it using a UNIX socket or via an HTTP API when configured to allow external access, Ohfeld explained.
“As a result, the vulnerabilities we found would allow external users or low-privileged users to remotely execute code on target machines or escalate privileges,” he wrote.
‘Textbook RCE Vulnerability”
CVE-2021-38647, with a 9.8 severity rating, is the most serious of the flaws, allowing for RCE. However, for it to be exploited, the Azure product using OMI would have to be 1, such as Configuration Management, that exposes an HTTPS port, or port 5986, for interacting with OMI.
“That’s what makes RCE possible,” Ohfeld explained. “Note that most Azure services that use OMI deploy it without exposing the HTTPS port.”
Calling the bug “a textbook RCE vulnerability that you would expect to see in the 90s” not in 2021, the flaw can expose millions of endpoints because “an attacker could use a single packet to become root on a remote machine by simply removing the authentication header,” Ohfeld wrote.
Coding Mistake
“Thanks to the combination of a simple conditional statement coding mistake & an uninitialized auth struct, any request without an Authorization header has its privileges default to uid=0, gid=0, which is root,” he explained.
In situations where the OMI ports are accessible to the internet to allow for remote management, threat players can use the vulnerability co-obtain initial access to a target Azure environment & then move laterally within it, Ohfeld added.
“An exposed HTTPS port is the holy grail for malicious actors,” he observed. “With 1 simple exploit they can get access to new targets, execute commands at the highest privileges & possibly spread to new target machines.”
The other 3 flaws—with severity ratings that range from 7.1 to 7.8—can be used as part of attack chains once attackers gain initial low-privileged access to their targets, Ohfeld added.
Threat Discovery & Mitigations
Wiz researchers reported the 4 vulnerabilities to Microsoft through the responsible disclosure process; the company patched them as of Tues., researchers stated.
Upgrading OMI & thus patch installation happens through the parent Azure service that installed it, they added. “However, we urge customers to verify that their environment is indeed patched & they are running the latest version of OMI (Version 1.6.8.1),” Ohfeld wrote.
Azure Linux Node
Different Azure services have different port numbers, Microsoft noted in its advisory for CVE-2021-38647. However, for customers who want to check that their Azure Linux Node does not have an exposed port, they should look for the command ‘netstat -an | grep <port-number>‘ on most Linux distributions, which will indicate if any processes are listening on an open port, the company concluded.
