Re: [Dots] Target-Attack-type expansion: more discussion

[Nik Teague <[e-mail]>]

-------- Original Message --------
On 9 May 2019, 04:33, MeiLing Chen wrote:
body { line-height: 1.5; }blockquote { margin-top: 0px; margin-bottom: 0px; margin-left: 0.5em; }div.FoxDiv20190509112907302659 { }body { font-size: 10.5pt; font-family: 微软雅黑; color: rgb(0, 0, 0); line-height: 1.5; }
Hi, Nik and Töma,
"procotol layer" is a field designed for attack type definition and classification, but the actual mitigation operation depends on the mitigation method provided by mitigators.

I'm not sure of the value then - from a reporting standpoint the information may be inferred from other characteristics. Taking memcached again, If I'm classifying that I would probably drop it in the reflection bucket not application. That rings true of most if not all reflection attacks - an application weakness may be exploited but the attack in that instance is volumetric - the payload is somewhat secondary to the goal of making lots of big things and throwing them at systems.

-------- Original Message --------
On May 6, 2019, 11:52 AM, Töma Gavrichenkov wrote:

On Mon, May 6, 2019 at 1:10 PM MeiLing Chen wrote:
> Actually, It is more inclined to use TCP/IP four-layer protocol.

Which layer is QUIC then?

The Internet protocol suite is not really layered. OSI model is, but
the IETF as a whole tends to slip away from the layered model. To
quote Christian Huitema:

"There is also beauty in *not* having a layered architecture [..]. It
is great to see transport functions like acknowledgement or flow
control fully contained in the Quic transport. Quic is about transport
innovation, and that pretty much requires direct access to the network
API. In practice, layered implementation hide that API, so the
transport developers have to constantly negotiate with the
intermediate layer developers."

I would strongly oppose a classification based on "exploited protocol
layers". As attractive as it is academically, it makes operational
issues more opaque.

--
Töma

-----
Layers have no real relevance... Using the memcached example - the exploitation may be at the application but the mitigation could easily be handled by a standard ACL.