On 2025-03-03 23:26:17 +0100, Vincent Lefevre wrote:
> BTW, the adduser code looks suspicious:
>
> my $ret = existing_user_status($new_name, $new_uid);
> if ($ret == (EXISTING_FOUND|EXISTING_SYSTEM)) {
> # a user with this name already exists; it's a problem when it's not
> a system user
> log_fatal( mtx("The user `%s' already exists, but is not a system
> user. Exiting."), $new_name );
> exit( RET_WRONG_OBJECT_PROPERTIES );
> }
>
> Concerning the test $ret == (EXISTING_FOUND|EXISTING_SYSTEM),
> if I understand correctly, if EXISTING_SYSTEM is set, this means
> that this is a system user: in sub existing_user_status:
>
> $ret |= EXISTING_SYSTEM if \
> ($uid >= $config{"first_system_uid"} && $uid <=
> $config{"last_system_uid"});
>
> However, I still don't understand the difference between the
> two machines.
Thanks the Michael Musenbrock, I now understand: on the other
machine, the account is locked!
Machine where I got the failure:
# grep colord /etc/shadow
colord:*:16714:0:99999:7:::
(with /bin/false for the shell in /etc/passwd).
Machine where the upgrade worked:
# grep colord /etc/shadow
colord:!:19637::::::
(with /usr/sbin/nologin for the shell in /etc/passwd).
So, in short, the test has (at least) 2 bugs:
1. It reverses the meaning of EXISTING_SYSTEM.
2. It depends on the lock status of the account.
Now, I don't understand why the account is not locked on the old
machine and the account is locked on the more recent machine.
A past bug of adduser?
--
Vincent Lefèvre <[email protected]> - Web: <https://www.vinc17.net/>
100% accessible validated (X)HTML - Blog: <https://www.vinc17.net/blog/>
Work: CR INRIA - computer arithmetic / Pascaline project (LIP, ENS-Lyon)
