On Tue, Feb 7, 2012 at 11:25 AM, Jakub Hrozek <jhrozek(a)redhat.com> wrote:
On Tue, Feb 07, 2012 at 11:04:53AM +0100, Marco Pizzoli wrote:
> Hi Jakub,
>
> On Tue, Feb 7, 2012 at 10:58 AM, Jakub Hrozek <jhrozek(a)redhat.com>
wrote:
>
> On Tue, Feb 07, 2012 at 10:48:59AM +0100, Marco Pizzoli wrote:
> > Hi John,
> >
> > On Tue, Feb 7, 2012 at 10:42 AM, John Hodrien
> <J.H.Hodrien(a)leeds.ac.uk>
> > wrote:
> >
> > On Tue, 7 Feb 2012, Marco Pizzoli wrote:
> >
> > Hi,
> > I'm using sssd-1.7.0-5.fc16.x86_64 and during my sssd
startup I
> see
> > this output:
> >
> > [cut]
> > (Tue Feb 7 10:25:16 2012) [sssd] [mark_service_as_started]
> (0x0100):
> > Now starting services!
> > (Tue Feb 7 10:25:16 2012) [sssd] [start_service] (0x0100):
> Queueing
> > service nss for startup
> > (Tue Feb 7 10:25:16 2012) [sssd] [start_service] (0x0100):
> Queueing
> > service pam for startup
> > (Tue Feb 7 10:25:16 2012) [sssd[be[my_ldap]]]
> [sbus_toggle_watch]
> > (0x4000): 0x7f0043097590/0x7f004309c2a0 (17), -/W
(disabled)
> > (Tue Feb 7 10:25:16 2012) [sssd[be[my_ldap]]]
> [sbus_remove_timeout]
> > (0x2000): 0x7f004309a510
> > (Tue Feb 7 10:25:16 2012) [sssd[be[my_ldap]]]
[sbus_dispatch]
> > (0x4000): dbus conn: 7F0043099BF0
> > (Tue Feb 7 10:25:16 2012) [sssd[be[my_ldap]]]
[sbus_dispatch]
> > (0x4000): Dispatching.
> > (Tue Feb 7 10:25:16 2012) [sssd[be[my_ldap]]]
[id_callback]
> (0x0100):
> > Got id ack and version (1) from Monitor
> > (Tue Feb 7 10:25:16 2012) [sssd] [service_startup_handler]
> (0x0010):
> > Could not exec sssd_nss, reason: No such file or directory
> > (Tue Feb 7 10:25:16 2012) [sssd] [service_startup_handler]
> (0x0010):
> > (Tue Feb 7 10:25:16 2012) [sssd] [mt_svc_exit_handler]
> (0x0040):
> > Child [nss] exited with code [1]
> > (Tue Feb 7 10:25:16 2012) [sssd] [start_service] (0x0100):
> Could not
> > exec sssd_pam, reason: No such file or directory
> > Queueing service nss for startup
> > (Tue Feb 7 10:25:16 2012) [sssd] [service_startup_handler]
> (0x0010):
> > (Tue Feb 7 10:25:16 2012) [sssd] [mt_svc_exit_handler]
> (0x0040):
> > Child [pam] exited with code [1]
> > (Tue Feb 7 10:25:16 2012) [sssd] [start_service] (0x0100):
> Could not
> > exec sssd_nss, reason: No such file or directory
> > Queueing service pam for startup
> > (Tue Feb 7 10:25:16 2012) [sssd] [service_startup_handler]
> (0x0010):
> > (Tue Feb 7 10:25:16 2012) [sssd] [mt_svc_exit_handler]
> (0x0040):
> > Child [nss] exited with code [1]
> > (Tue Feb 7 10:25:16 2012) [sssd] [start_service] (0x0100):
> Queueing
> > service nss for startup
> > Could not exec sssd_pam, reason: No such file or directory
> > (Tue Feb 7 10:25:16 2012) [sssd] [service_startup_handler]
> (0x0010):
> > (Tue Feb 7 10:25:16 2012) [sssd] [mt_svc_exit_handler]
> (0x0040):
> > Could not exec sssd_nss, reason: No such file or directory
> > Child [pam] exited with code [1]
> > (Tue Feb 7 10:25:16 2012) [sssd] [start_service] (0x0100):
> Queueing
> > service pam for startup
> > (Tue Feb 7 10:25:16 2012) [sssd] [mt_svc_exit_handler]
> (0x0040):
> > Child [nss] exited with code [1]
> > (Tue Feb 7 10:25:16 2012) [sssd] [start_service] (0x0100):
> Queueing
> > service nss for startup
> > (Tue Feb 7 10:25:16 2012) [sssd] [service_startup_handler]
> (0x0010):
> > Could not exec sssd_pam, reason: No such file or directory
> > (Tue Feb 7 10:25:16 2012) [sssd] [mt_svc_exit_handler]
> (0x0040):
> > Child [pam] exited with code [1]
> > (Tue Feb 7 10:25:16 2012) [sssd] [start_service] (0x0100):
> Queueing
> > service pam for startup
> > (Tue Feb 7 10:25:16 2012) [sssd] [service_startup_handler]
> (0x0010):
> > Could not exec sssd_nss, reason: No such file or directory
> > (Tue Feb 7 10:25:16 2012) [sssd] [mt_svc_exit_handler]
> (0x0040):
> > Child [nss] exited with code [1]
> > (Tue Feb 7 10:25:16 2012) [sssd] [mt_svc_exit_handler]
> (0x0010):
> > Process [nss], definitely stopped!
> > (Tue Feb 7 10:25:16 2012) [sssd] [service_startup_handler]
> (0x0010):
> > Could not exec sssd_pam, reason: No such file or directory
> > (Tue Feb 7 10:25:16 2012) [sssd] [mt_svc_exit_handler]
> (0x0040):
> > Child [pam] exited with code [1]
> > (Tue Feb 7 10:25:16 2012) [sssd] [mt_svc_exit_handler]
> (0x0010):
> > Process [pam], definitely stopped!
> > [cut]
> >
> > [root@fedora16 sssd]# locate sssd_pam
> > /usr/lib/debug/usr/libexec/sssd/sssd_pam.debug
> > /usr/libexec/sssd/sssd_pam
> >
> > [root@fedora16 sssd]# locate sssd_nss
> > /usr/lib/debug/usr/libexec/sssd/sssd_nss.debug
> > /usr/libexec/sssd/sssd_nss
> >
> > Could you help me in finding the problem?
> >
> > AVC denials listed in /var/log/audit/audit.log? SELinux was
my
> first
> > thought
> > looking at this.
> >
> > If you do:
> >
> > setenforce 0
> > service sssd restart
> >
> > Does it work then? If not, it's something else ;)
> >
> > [root@fedora16 sssd]# getenforce
> > Disabled
> >
> >
> >
> > stracing the sssd process could give you a pointer as to
exactly
> what
> > it's
> > trying to do that's failing, in case it's looking in the
wrong
> place or
> > something.
> >
> > Thanks to pointing me to this. Found the problem.
> > [pid 1853] execve("/usr/local/bin/sssd_pam",
["sssd_pam"],
[/* 26
> vars
> > */]) = -1 ENOENT (No such file or directory)
>
> This ^^ is wrong. Did you build the package yourself or is this an
> offictial one? Does your sssd.conf use the "command" parameter?
>
> This is the rpm Stephen asked me to try last week.
>
> Yes, I used the command parameter on both cases. It has been a fault?
I
> was only trying expliciting default parameters.
This is a manpage bug.
The manpage says the default is sssd_${service}, but it's in fact
${libexec}/sssd_${service} which expands to absolute path pointing to
/usr/libexec/sssd/sssd_${service} by default.
When you specified a relative command name, execvp() probably tried to
execute the command somewhere in $PATH and failed.
I understand.
We should either remove the command parameter altogether or at least
fix
the default.
Thanks, this is tracked in
https://fedorahosted.org/sssd/ticket/1174
now.
Now I'm feeling less embarrassment in mailing you my help requests... ;-)
Removing the command parameter will make the sssd work.
Done and effectively now it's working.
Thanks
Marco