[VoiceOps] 1. Large VoIP Attacks?

J. Oquendo sil at infiltrated.net
Tue Nov 26 14:12:20 EST 2013


On Tue, 26 Nov 2013, Pawlowski, Adam wrote:

> I don't normally post to this list, as it is more for the SP side of things.
> A few weeks back we had a publically exposed LifeSize video conference unit
> compromised. It was not installed particularly well, given the limitations to
>  the administrative PIN code, but it has shell access enabled, and a flash portal
> which purportedly has rights execution issues. The vendor has a firewall
> device that would place these behind it on private space - was that in play
> where your units were attacked? We've since removed these until we can
> provide a full video call control infrastructure, to keep from exposing these
> appliances, but I figured it was the poor PIN code that was hit, not the device
> itself.
> 
> Adam Pawlowski
> SUNYAB

This LifeSize was "theoretically" behind a firewall. I will
explain 'that' one. We manage(d) the FWs (SSGs btw) there
however... Client 'demanded' they have access to be able to
change things as they saw fit. So whatever rules are, or
were in place, is who knows. That became the Pottery Barn
rule for us (you touched it you bought it).

Now... Depending on which version of LifeSize you have,
you're either running Apache, or now Lighttpd, both of
which are "DoS'able." So that's fine and dandy, where DoS
comes into play however, the user "Test()" is an anomaly
because this user was ALSO coming into the PBXNSIP systems'
CDR logs.

We try whenever not to ever make devices public, but when
clients want to be able to have their PBXs have their calls
forwarded from their PBX, to their cells, refrigerators, and
toilets, we have to let them know: "hey this is what can
(and usually does) happen." They're all chipper with it
until they get a bill for $50,000.00 because someone went
ahead and used a strong password of 12349876 even though we
have shown them: "look, these attackers, can test hundreds
of thousands of password in a minute..."

LifeSize: I want to say a lot, but it would be stepping on
toes. My suggestion, if you need to put it online, do so
for the duration of use, then pull the plug. Same goes for
the Polycoms. One of these days, I will stop being lazy and
write up some advisories I have been holding onto for some
time...

Maybe we should all get together and keep a running WIP
(work in progress) of a "Best Practice" which would be a no
bs, vendor neutral: "DO THIS NOW" whitepaper of sorts. I'd
be willing to spew theories/thoughts/practices. (By the way
chucking VoIPSA back on this thread, I know many criss cross
that list, but some on that list, don't traverse this one).

-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+
J. Oquendo
SGFA, SGFE, C|EH, CNDA, CHFI, OSCP, CPT, RWSP, GREM

"Where ignorance is our master, there is no possibility of
real peace" - Dalai Lama

42B0 5A53 6505 6638 44BB  3943 2BF7 D83F 210A 95AF
http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x2BF7D83F210A95AF


More information about the VoiceOps mailing list