Bug 176992

Created attachment 122805 [details]
Best. Dialog. Ever!

How are the versions of pam and usermode on the box you see this? The Best
Dialog Ever ;) is displayed when pam returns an unknown exit code usermode is
unable to handle. The versions of both will help me figure out what's going on.

latest rawhide..

pam-0.99.2.1-3
usermode-1.85-1

this has actually been around for a while, but I've been lazy at reporting this.

Could you add here /etc/pam.d/reboot and /etc/pam.d/system-auth to let me figure
out which pam module is the culprit?

The remote boxes I'd ssh'd into are running current rawhide, with no
modifications to these files..


(16:16:02:davej@nemesis:~)$ cat /etc/pam.d/reboot
#%PAM-1.0
auth       sufficient   pam_rootok.so
auth       required     pam_console.so
#auth       include     system-auth
account    required     pam_permit.so
(16:16:09:davej@nemesis:~)$ cat /etc/pam.d/system-auth
#%PAM-1.0
# This file is auto-generated.
# User changes will be destroyed the next time authconfig is run.
auth        required      /lib/security/$ISA/pam_env.so
auth        sufficient    /lib/security/$ISA/pam_unix.so likeauth nullok
auth        required      /lib/security/$ISA/pam_deny.so

account     required      /lib/security/$ISA/pam_unix.so
account     sufficient    /lib/security/$ISA/pam_succeed_if.so uid < 100 quiet
account     required      /lib/security/$ISA/pam_permit.so

password    requisite     /lib/security/$ISA/pam_cracklib.so retry=3
password    sufficient    /lib/security/$ISA/pam_unix.so nullok use_authtok md5
shadow
password    required      /lib/security/$ISA/pam_deny.so

session     required      /lib/security/$ISA/pam_limits.so
session     required      /lib/security/$ISA/pam_unix.so
(16:16:13:davej@nemesis:~)$

Weird, works for me with the same configs. When ssh -X to a FC4/rawhide box and
type reboot I get:

[jnovy@obelix ~]$ reboot
reboot: must be superuser.

There's usermode-1.85-1 and pam-0.99.2.1-3 installed on the box.

Dou you have SELinux enabled btw.?

yes, selinux is enabled (though it makes no difference if its disabled)

I neglected to mention in comment #1, that I was actually running 'sudo reboot',
not just 'reboot'

Created attachment 123861 [details]
New usermode package to test.

Could you please compile and install this package and send output from the
`sudo reboot` here? It'll dump some debuginfo. I can read from it what happens
there. I bet it's unknown pam module or something.

hrmph. there are a bunch of missing build depends there. (libattr-devel,
libICE-devel, libSM-devel)  The build also fails in the end with unpackaged files..

RPM build errors:
    user jnovy does not exist - using root
    group jnovy does not exist - using root
    user jnovy does not exist - using root
    group jnovy does not exist - using root
    Installed (but unpackaged) file(s) found:
   /usr/etc/security/console.apps/halt
   /usr/etc/security/console.apps/poweroff
   /usr/etc/security/console.apps/reboot

If you want to just point me at an x86-64 binary rpm, it'll probably be less
painful.

Created attachment 123909 [details]
usermode testing binary #1

Created attachment 123911 [details]
usermode testing binary rpm #2

Ok, attaching the binaries so that you can test it without messing with the
SRPMS compilation.

Running child pid=2823.
Exec arg 0 = "/usr/sbin/userhelper".
Exec arg 1 = "-w".
Exec arg 2 = "reboot".
userhelper: ruid user = 'root'
userhelper: current user is root
userhelper: ruid user = 'root'
userhelper: user for auth = 'root'
userhelper: authenticating "root"
userhelper: PAM retval = 0 (Success)
userhelper: about to exec "/sbin/reboot"
EOF from child.
Child 2823 exited (looking for 2823).
Child 2823 exited abnormally.
Child 2823 died on signal 11.
Child returned exit status 255.
Status is "Unknown error.".
Quitting main loop 1.
Child exited, continuing.


It dumped a core, which gdb tells me..

Core was generated by `/usr/sbin/userhelper -w reboot'.
Program terminated with signal 11, Segmentation fault.
....

Loaded symbols for /lib64/libnss_files.so.2
#0  0x0000000000403aeb in converse_pipe ()
(gdb) bt
#0  0x0000000000403aeb in converse_pipe ()
#1  0x00000000004040e2 in pipe_conv_exec_start ()
#2  0x000000000040524e in wrap ()
#3  0x0000000000405ce1 in main ()

*** Bug 179374 has been marked as a duplicate of this bug. ***

Fixed in rawhide usermode-1.90-1.  Thanks for your report.