I've got one stationary computer and two laptops. All of them running slackware 13.37.

When I lock my fluxbox with xlock on my stationary computer, all is well. However, if I do the same on any of my laptops, I get failed login. It's not a clean installation, I've used some slackbuilds and I've got a custom kernel. I fail to see why that should cause this effect. And it's mostly the same slackbuilds/.config for all computers.

Could someone test/"verify" this behavior?

(Protip: If you fail to authenticate, you can kill xlock in a console instead of "zap" the X-session)

My first thought is some PAM configuration issue. I'm not a slacker, but you should have pam configs in /etc/pam.d you can look through. On my system, there is one for "xscreensaver" which uses the "system-auth" config. If you posted both of these perhaps we'd see something.

I had a problem like this, it turned out Xlock was using a different keyboard map, and the charactors I thought I was typing in, were actually different to what the computer was recieving, and so it would fail to unlock. But I cant remember exactly how I solved the problem I usually write everything down, but I cant find the solution among my stuff.

I currently use xlock myself, and it works fine when I type in the password. Try changing the keymap in Xorg?

Ok, I found the difference. On my laptops I've got a ~/.xlock file that contains a hashed version of the password. This file doesn't exist on my workstation. If the file doesn't exist on the laptops I am prompted for a password which is then stored in this file.

I could mark this a solved, but it turned in to a curiosity now. It's the same slackware version on all of them. According to the documentation you'll have to change a file before compilation to choose shadow authentication. So I can't seem to understand the difference in behavior.