An extremely critical vulnerability affecting most Linux distributions gives attackers the ability to execute malicious code on servers used to deliver e-mail, host webpages, and carry out other vital functions.
The vulnerability in the GNU C Library (glibc) represents a major Internet threat, in some ways comparable to the Heartbleed and Shellshock bugs that came to light last year. The bug, which is being dubbed “Ghost” by some researchers, has the common vulnerability and exposures designation of CVE-2015-0235. While a patch was issued two years ago, most Linux versions used in production systems remain unprotected at the moment. What’s more, patching systems requires core functions or the entire affected server to be rebooted, a requirement that may cause some systems to remain vulnerable for some time to come.
The buffer overflow flaw resides in __nss_hostname_digits_dots(), a glibc function that’s invoked by the gethostbyname() and gethostbyname2() function calls. A remote attacker able to call either of these functions could exploit the flaw to execute arbitrary code with the permissions of the user running the application. In a blog post published Tuesday, researchers from security firm Qualys said they were able to write proof-of-concept exploit code that carried out a full-fledged remote code execution attack against the Exim mail server. The exploit bypassed all existing exploit protections available on both 32-bit and 64-bit systems, including address space layout randomization, position independent executions, and no execute protections. Qualys has not yet published the exploit code but eventually plans to make it available as a Metasploit module.
Read 6 remaining paragraphs | Comments

Leave a Reply