Red Hat Bugzilla – Bug 437499
silently rejects certain encryption passphrases
Last modified: 2014-03-16 23:12:58 EDT
Description of problem:
Hit the spacebar eight times for your encryption passphrase. anaconda doesn't
throw an error, but appears to have rejected it.
Version-Release number of selected component (if applicable):
It appears to be treating it as blank - any leading or trailing spaces on
passphrases are removed as well.
I'm so used to stripping leading/trailing whitespace from input that I did it
for the passphrases as well. You think we should preserve it, I take it?
Not sure. Do we do it for passwords?
If nothing else, you may want to move the character count until after the strip.
To my surprise we do not do any stripping of whitespace from the root password.
The more I think about it, the more I think we should not do any unrequested
modification of the passphrase. It's exactly the sort of thing that pisses me
off as a user.
While I'm at it, any thoughts on enforcing a limited character set like we do
for the password? (string.digits + string.ascii_letters + string.punctuation + "
") It doesn't seem necessary to me, but I know nothing of such things.
Well, it would obviate some of the keymap issues if we did. Don't know it's 100%
Should be fixed in anaconda-126.96.36.199-1.