[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: pwdb breakage

On Fri, 30 May 1997,  "Michael K. Johnson" <johnsonm@redhat.com> wrote:

> Well, pwdb/pam_pwdb appear to be very broken.  I'm dealing with
> pwdb 0.54 preC and pam 0.57 as we shipped with Red Hat Linux 4.2 --
> if any of these bugs have been fixed since, please kindly tell
> me...  :-)  Thanks!

Why was 0.57 shipped with RedHat v4.2?  I made the idiotic mistake of
upgrading from the version shipped with RedHat v4.0 (0.50?) to the 0.57
and I got completely locked out of my machine (thank god for ssh).  I
imagine RedHat support will be swamped by anyone who does the same.

> In general, it appears that if /etc/shadow contains a * password,
> the password is modified in /etc/passwd instead.

It should be noted that if there is something other than a "*" in
/etc/shadow the passwd command works fine.  However, anyone adding a user
with /usr/sbin/adduser, which creates an entry in /etc/shadow using
password "*", will not be able to change the newly created user's
password.  Again, how did this ship without being checked first?

Just a suggestion: perhaps PAM developers could adopt the stable vs. 
unstable numberring system?  Something along the lines of the way the
kernel is developed?  PAM a VERY useful package but unsuspecting (and
arguably stupid) users like myself will really stumble with it -
especially with bugs like this. 


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index] []