News for hackers. Stuff that matters.
Re:CVE-2015-0235 – GHOST: glibc gethostbyname buffer overflow by windsurfer on Tuesday January 27, 2015 @03:18PM (#8955511)
I always laugh when people give you a URL to C code to test for remote code execution...
Re:CVE-2015-0235 – GHOST: glibc gethostbyname buffer overflow by fragmede on Tuesday January 27, 2015 @04:07PM (#8955896)
Why?

Reproducers are standard fare and it's not like the code in this case is obfuscated. Are magic code goblins going to come and invoke Ken Thompson's untrustable computing and make your computer install windows and join a botnet or something?

Re:CVE-2015-0235 – GHOST: glibc gethostbyname buffer overflow by graemian on Wednesday January 28, 2015 @01:36PM (#8961617)
Because if you give someone a URL to C code and they run it, you have effective gotten them to remotely execute your code ;-)
Re:CVE-2015-0235 – GHOST: glibc gethostbyname buffer overflow by Natsu on Tuesday January 27, 2015 @04:31PM (#8956097)
You're piping random executables from the internet without even looking at them to see what they do if you run that command.
Re:CVE-2015-0235 – GHOST: glibc gethostbyname buffer overflow by Xylakant on Tuesday January 27, 2015 @04:38PM (#8956148)
There's no pipe involved in the GPs posting. It downloads a source file and the post even points out to check the file before compiling. I actually copied and pasted the test code from the original advisory. Should I have typed it in with my own bare fingers to be more secure? I agree with your sentiment in general, you just picked the wrong example to bash here.
Re:CVE-2015-0235 – GHOST: glibc gethostbyname buffer overflow by Natsu on Tuesday January 27, 2015 @08:15PM (#8957425)
I was responding to the later posts who were confused as to why someone would laugh at seeing that. You're right that you said to check and I wasn't trying to bash that post at all.

That aside, I actually did type it into my disposable VM. The theory being that if there was something subtle, it would force me either to type it wrong and not be exploited due to cognitive blindness or I'd catch the problem and avoid it.

I've read too many IOCC entries and I probably am a bit paranoid.

Re:CVE-2015-0235 – GHOST: glibc gethostbyname buffer overflow by acveilleux on Tuesday January 27, 2015 @05:04PM (#8956332)
The average Linux IT guy will not read the C code. Many of them wouldn't be able to really understand what it does either. And this is best case code is simple and easy to grok.
Re:CVE-2015-0235 – GHOST: glibc gethostbyname buffer overflow by Xylakant on Tuesday January 27, 2015 @05:34PM (#8956579)
If you're not reading the code you can just as well curl-and-pipe it. However, we run so much code on our computers that is trusted-by-association (oh, that's from apache.org, that's probably safe!) that it probably does not matter anyways.
Re:CVE-2015-0235 – GHOST: glibc gethostbyname buffer overflow by peteretep on Wednesday January 28, 2015 @01:06AM (#8958387)

    > You're piping random executables from the internet
    > without even looking at them to see what they do if you
    > run that command
... and then in your profile:

    > Just another Perl hacker.
You audit every CPAN module you install, line by line, right?
Re:CVE-2015-0235 – GHOST: glibc gethostbyname buffer overflow by toddsiegel on Tuesday January 27, 2015 @03:41PM (#8955681)
I agree, it is funny, but we run code we can't even see all day long. This is only 38 LOC. Anyone who is going to run it, however, should make sure they understand it.
Re:CVE-2015-0235 – GHOST: glibc gethostbyname buffer overflow by cellover on Wednesday January 28, 2015 @08:24AM (#8959622)
You know you can also read the source code right? Even if you are not proficient with c you can understand what it's doing.
Re:CVE-2015-0235 – GHOST: glibc gethostbyname buffer overflow by tedunangst on Tuesday January 27, 2015 @04:05PM (#8955878)
As opposed to code in some other language?
Re:CVE-2015-0235 – GHOST: glibc gethostbyname buffer overflow by fletchowns on Tuesday January 27, 2015 @03:40PM (#8955669)
Might as well throw a --no-check-certificate in there