Its yet another new spin on a pervasive attack -- this time using the old standby Simple Network Management Protocol (SNMP) to stage cross-site scripting (XSS) attacks.
XSS, which basically forces a Website to echo malicious code that then gets loaded into a user's browser, is one of the most common vulnerabilities in Web applications. Researchers with ProCheckUp Ltd. recently discovered what they think may be a new type of attack vector, using SNMP to create a persistent XSS attack. Persistent XSS is a more powerful XSS attack where malicious code is stored on a Website for a period of time, and all the user has to do is view the page to get infected.
With SNMP, the attacker changes parameters in the device to then launch a persistent XSS attack. ProCheckUp found the SNMP-XSS vulnerability, as well as several others, while researching ZyXELs Prestige router products, which are commonly used in home, SOHO, and ISP networks.
"I believe that this is a totally brand new attack which I suspect affects many other appliances from other vendors," says Adrian Pastor, a security consultant with ProCheckUp.
According to Pastor's report, a persistent XSS attack is launched "when the parameters containing the payload are printed on the browser via the web interface of the device. The problem lies in part with the fact that ZyXELs Prestige products run with SNMP, HTTP, and telnet in default mode on its WAN interface, Pastor says. This is at least true among the ISPs used by some of our customers who we offer penetration testing services for, he writes.
XSS expert RSnake -- aka Robert Hansen, CEO of SecTheory -- says using SNMP to launch an XSS attack is definitely an interesting approach. It's hard to say that it's new since lots of exploits use logging variables to instantiate the XSS attack, but that's definitely the first time I've heard SNMP [being] used."
RSnake says SNMP and Web application hackers dont typically intersect, so the attack method is an interesting mix: "Lots of hackers use SNMPwalk, but not many Web app hackers."
Such an attack would most likely be a targeted one rather than a large-scale one, he says. "It would be almost impossible to automate an attack like that because you'd need to set up a Website to pull the credentials or log the output, etc. I think it's less accessible than most forms of XSS, so it's less likely to see wide scale abuse. But it is a very interesting vector for the same old attack weve known and seen affect so many things, so many times.
Meanwhile, ProCheckUp also discovered several other vulnerabilities in the ZyXEL products, including remote war-driving via the Internet, privilege escalation, SNMP read and write flaws, credential disclosure weaknesses, and authentication vulnerabilities.
Have a comment on this story? Please click "Discuss" below. If you'd like to contact Dark Reading's editors directly, send us a message.