Comment 2 for bug 305882

Dustin Kirkland  (kirkland) wrote :

Subscribing Steve Langasek...


This is yet another case of badness regarding PAM's architecture. I can confirm this problem, and it's a nasty one.

The ecryptfs PAM module thinks that the password change succeeded, and so we go ahead and rewrap the passphrase, but this isn't so.

We need to get this fixed...