Re: memcached UDP amplification attacks

[Patrick Forsberg <fors () cert sunet se>]

On 2018-03-07 22:57, Seaman, Chad wrote:

> Tomas,
>
> You’re not wrong, that was a typo in the blog.
>
> Regards,
> Chad
On the other hand I think the blog is wrong. (Just reading the protocol
specs)

The default maximum size of an item is 1M, but you can easily request
more than one item using the gets command.
   gets <key>*\r\n
where <key>* is one or several keys separated by whitespace.
A key is a text string that must not include control characters or
whitespace. The maximum length of a key is 250 characters but I don't
think there's a minimum size so one byte keys should be fine.

/Patrick
> On 3/7/18, 5:10 AM, "Tomas Hoger" <thoger () redhat com> wrote:
>
>     On Fri, 2 Mar 2018 21:42:30 -0700 Kurt Seifried wrote:
>     
>     > I have assigned CVE-2018-1000115 to this issue:
>     > 
>     > Memcached version 1.5.5 contains an Insufficient Control of Network
>     > Message Volume (Network Amplification, CWE-406) vulnerability in the
>     > UDP support of the memcached server that can result in denial of
>     > service via network flood (traffic amplification of 1:50,000 has been
>     > reported by reliable sources). This attack appear to be exploitable
>     > via network connectivity to port 11211 UDP. This vulnerability
>     > appears to have been fixed in 1.5.6 due to the disabling of the UDP
>     > protocol by default.
>     
>     Minor nitpick, the description mentions 1:50,000 ratio, apparently
>     based on the information in the following reference:
>     
>     > https://blogs.akamai.com/2018/03/memcached-fueled-13-tbps-attacks.html
>     
>     where it's mentioned as:
>     
>     """
>     Worse, memcached can have an amplification factor of over 50,000,
>     meaning a 203 byte request results in a 100 megabyte response.
>     """
>     
>     However, 200 * 50k = 10m, not 100m.  Wonder if I'm doing my math wrong.
>     
>     -- 
>     Tomas Hoger / Red Hat Product Security
>