We discovered a new Mirai variant (detected as IoT.Linux.MIRAI.VWISI) that exploits nine vulnerabilities, most notable of which is CVE-2020-10173 in Comtrend VR-3033 routers which we have not observed exploited by past Mirai variants.
This discovery is a new addition to the Mirai variants that appeared in the past few months, that include SORA, UNSTABLE, and Mukashi. The case, however, showcases the ever-expanding arsenal of vulnerabilities new Mirai variants are equipped with by their developers.
The vulnerabilities used by this Mirai variant consist of a combination of old and new that help cast a wide net encompassing different types of connected devices. The nine vulnerabilities used in this campaign affect specific versions of IP cameras, smart TVs, and routers, among others.
As mentioned earlier, the most notable of these vulnerabilities is CVE-2020-10173, a Multiple Authenticated Command injection vulnerability found in Comtrend VR-3033 routers. Remote malicious attackers can use this vulnerability to compromise the network managed by the router.
Only a proof of concept (POC) has been released for this vulnerability, with no reported exploit at large before this Mirai variant. Figure 1 serves as evidence of how this vulnerability is used by the sample.
Figure 1. Code snippet that shows the use of CVE-2020-10173
Another relatively recent vulnerability also used in this campaign is Netlink GPON Router 1.0.11 RCE. Discovered this year, it was reportedly exploited by the Bashlite/Gafgyt variant Hoaxcalls.
Figure 2. Code snippet that shows the use of Netlink GPON Router 1.0.11 RCE
Aside from the two, the variant makes use of mostly old vulnerabilities which have been used in past campaigns. The two code snippets shown in figures 3 and 4 serve as examples of old vulnerabilities written in the variant’s code.
Figure 3. Code snippet that shows the use of LG SuperSign EZ CMS 2.5 - Remote Code Execution
Figure 4. Code snippet that shows the use of Linksys E-series - Remote Code Execution
In addition to these examples, the remaining five old vulnerabilities that were exploited by the variant are the following:
- AVTECH IP Camera / NVR / DVR Devices - Multiple Vulnerabilities
- D-Link Devices - UPnP SOAP Command Execution
- MVPower DVR TV-7104HE 1.8.4 115215B9 - Shell Command Execution
- Symantec Web Gateway 22.214.171.124 Remote Code Execution
- ThinkPHP 5.0.23/5.1.31 - Remote Code Execution
A hallmark of Mirai variants is the use of Telnet and Secure Shell (SSH) brute-forcing as evidenced by our sample. This variant also used the typical XOR encryption (with the XOR key: 0x04) to hide the credentials that it uses to attack vulnerable devices. The credentials we extracted are listed here.
Conclusion and security recommendations
The use of CVE-2020-10173 in this variant’s code shows how botnet developers continue to expand their arsenal to infect as many targets as possible and take advantage of the opening afforded by unpatched devices. Newly discovered vulnerabilities, in particular, offer better chances for cybercriminals. Users, not knowing that a vulnerability even exists, might be unable to patch the device before it is too late.
In the future, it would be wise to expect this vulnerability might be used in new DDoS botnets like Mirai. As monitoring of such botnets show, handlers tend to copy each other’s techniques, including lists of vulnerabilities and credentials that increase their chance of success.
For devices to remain safe from the usual tactics of botnet malware, users need to follow best practices in securing their connected devices. These include the following:
- Patch vulnerabilities and apply updates as soon as they become available.
- Use network segmentation to limit the spread of potential infections.
- Use strong passwords and quickly change default ones.
- Apply secure configurations for devices to limit unforeseen openings for infection.
Connected devices can also be protected by security software such as the Trend Micro™ Home Network Security and Trend Micro™ Home Network Security SDK solutions, which can check internet traffic between the router and all connected devices as well as help users asses for vulnerabilities.
Trend Micro™ Deep Discovery™ Inspector also protects customers from this attack via these DDI rules:
- 2452 - Wget Commandline Injection
- 2544 - JAWS Remote Code Execution Exploit - HTTP (Request)
- 2575 - Command Injection via UPnP SOAP Interface - HTTP (Request)
- 2692 - LINKSYS Unauthenticated Remote Code Execution Exploit - HTTP (Request)
- 2713 - AVTECH Command Injection Exploit - HTTP (Request)
- 2786 - ThinkPHP 5x Remote Code Execution - HTTP (Request)
- 2865 - CVE-2018-17173 LG Supersign Remote Code Execution - HTTP (Request)
- 4689 - Comtrend - Remote Command Execution Exploit - HTTP (REQUEST)
Indicators of compromise (IoCs)
|SHA256||Old Detection||New Detection|
Command and control (C&C) servers