Re: CVE request: uglify-js node.js module <2.4.24 incorrectly handles non-boolean comparisons during minification

[Reed Loden <reed () reedloden com>]

On Mon, Aug 24, 2015 at 12:50 PM, Florian Weimer <fweimer () redhat com> wrote:

> On 08/24/2015 08:26 PM, Reed Loden wrote:
> > As seen on Hacker News --
> https://zyan.scripts.mit.edu/blog/backdooring-js/
> > Blog post has all the details, but basically the UglifyJS node module
> has a
> > problem where the combination of De Morgan’s Law and non-boolean values
> can
> > lead to a case where code is incorrectly minified, which can lead to
> > possibly malicious minified JS code.
> >
> > UglifyJS is a "JavaScript parser / mangler / compressor / beautifier
> > toolkit" for Node.js.
>
> How is this different from a any other compiler bug?  They can be abused
> in similar ways by crafted code, and we don't treat them as
> vulnerabilities (unless there is actual application impact beyond
> synthetic test cases).
>
> Note that I'm not saying this isn't a nice find, I'm just not sure if it
> should be considered as a security by itself.

I fully agree not every compiler bug is a security issue, but when one is
used for (theoretically) malicious purposes (including a full write-up and
example PoC), seems like there should be some middle ground here. If we're
always being reactive and waiting for an issue to actually be exploited,
we'll never win. Sadly, sometimes it takes an issue being classified as
'security' for folks (including distros) to update their
applications/software to pull in updated dependencies.

~reed