Il giorno gio 26 mar 2020 alle ore 17:37 Peter Schober <
peter.schober at univie.ac.at> ha scritto:
* Giuseppe De Marco <giuseppe.demarco at
unical.it> [2020-03-26 17:00]:
I never thought that a signature could be too
much computational
intensive but, it could be so.
Not a single one, but thousands of "cheap" GET requests to a
protected resource, as easily generated by a single unauthenicated
client on the network, resulting in "expensive" signing operations.
It's this asymmetry in the attacker's favor you'd want to avoid.
that's interesting Peter,
Probably that kind of abuse should be handled in a SIEM or a next-gen
firewall, or locally on the same server with those ip2ban or better a httpd
firewall, or an anti brute force applicative prevention (many on django, my
favourite framework).
I think that this aspect would be better handled in the domain of the
global security treatment of an IT infrastructure.
My point of view, simply, is the following:
the signature prevents that a fake entity could access to the credential
prompt, this latter is the way to do bruteforce attacks.
Probably a sane LDAP Enterprise deployment have a password policy with a
temporary lockout or whatever, but bruteforce happened. The MFA help in
that, but is a second step once the credentials have been stolen.
I always seen the signature as a good tool for preventing bruteforce
attacks, if some performance goes down the IdP should just scaled up soon
as possible and a good SIEM/httpd_firewall will do the rest helping us
seeing the bads and the uglies.
personal point of view, still in search of the truth ...
--
____________________
Dott. Giuseppe De Marco
CENTRO ICT DI ATENEO
University of Calabria
87036 Rende (CS) - Italy
Phone: +39 0984 496961
e-mail: giuseppe.demarco at unical.it