Memory acquisition is difficult and often unsuccessful, I have done it many times for incident response on behalf of federal LEO. Any password stash will autolock after a period of inactivity, making memory acquisition fruitless. I also can't imagine a password stash that does not use a salt.
The human element of writing down and typing by hand hundreds of unique password is the weakness, not password stashes.
As for rainbow tables, they only work if there is no salt, and once again I know of no password stash that does not use a salt.
But for arguments sake let's say that a site or password stash uses a simple MD5 (one way hash) to store passwords and the database gets dumped. As humans you'll choose a weak password like "broncos!".
So if the site just stores MD5s lets see what happens.
$ echo -n broncos! | md5
c49728934b41c8f8ce932480d01a0ff0
We google that, it's found.
LMGTFY
So if we use a password manager with a random password and it get's dumped, do we find it?
$ echo -n 'W*z(v7apowim=;v^Bh)3P=.&JQ[:Uf"' | md5
c20d5344fa641523b1bd4b1515b76c6f
No it is not found in rainbow tables.
LMGTFY
So then let's play with using a password manager and a weak password. The password management will use a salt, so if the DB gets dumped can you find it?
$ export SALT="velocipede salon"
$ echo -n 'broncos!'$SALT | md5
21e852b94de31c78b172b0361ed44acd
No it is not found.
LMGTFY
I argue the human element of creating, storing and entering unique password per site far outweighs the likelihood your PW stash will be compromised. I have been doing DFIR for 15 years now and never once have I been called in to help out when a password stash was compromised. I have been asked to help because of XSS, SQLi and phishing more than anything else. Unique passwords per site helps more in all the cases I have seen.
-Joe
Bookmarks