I am guessing it is well known that the gdm process runs but does not get the login screen started in the current Rawhide?
I am guessing it is well known that the gdm process runs but does not get the login screen started in the current Rawhide?
It could be this: https://bugzilla.redhat.com/show_bug.cgi?id=1331668
On Fri, 2016-04-29 at 08:20 -0400, Kamil Paral wrote:
I am guessing it is well known that the gdm process runs but does not get the login screen started in the current Rawhide?
It could be this: https://bugzilla.redhat.com/show_bug.cgi?id=1331668 --
I suspect you are right, I rebooted with the selinux=0 boot option and GDM does what it should. Using enforcing=0 does not allow DGM to do its job unlike what the bug report post claims, so maybe there are differences. We shall see…
On 04/29/2016 06:44 AM, Russel Winder wrote:
On Fri, 2016-04-29 at 08:20 -0400, Kamil Paral wrote:
I am guessing it is well known that the gdm process runs but does not get the login screen started in the current Rawhide?
It could be this: https://bugzilla.redhat.com/show_bug.cgi?id=1331668 --
I suspect you are right, I rebooted with the selinux=0 boot option and GDM does what it should. Using enforcing=0 does not allow DGM to do its job unlike what the bug report post claims, so maybe there are differences. We shall see…
It's interesting you mention that. I've seen selinux behave differently between disabled state and permissive for a long, LONG time.
As I understand it, permissive should allow all operations but give warnings while disabled means, well, disabled. However, I've seen permissive mode _block_ some operations and not issue any warnings about those blocked operations.
I can't give details about what has been blocked in the past. I've lost track of the records. I do recall a lot of it had to do with networking operations. This issue seems somehow related. ---------------------------------------------------------------------- - Rick Stevens, Systems Engineer, AllDigital ricks@alldigital.com - - AIM/Skype: therps2 ICQ: 226437340 Yahoo: origrps2 - - - - A squeegee, by any other name, wouldn't sound as funny. - ----------------------------------------------------------------------
On Fri, 2016-04-29 at 09:49 -0700, Rick Stevens wrote:
On 04/29/2016 06:44 AM, Russel Winder wrote:
On Fri, 2016-04-29 at 08:20 -0400, Kamil Paral wrote:
I am guessing it is well known that the gdm process runs but does not get the login screen started in the current Rawhide?
It could be this: https://bugzilla.redhat.com/show_bug.cgi?id=1331668 --
I suspect you are right, I rebooted with the selinux=0 boot option and GDM does what it should. Using enforcing=0 does not allow DGM to do its job unlike what the bug report post claims, so maybe there are differences. We shall see…
It's interesting you mention that. I've seen selinux behave differently between disabled state and permissive for a long, LONG time.
As I understand it, permissive should allow all operations but give warnings while disabled means, well, disabled. However, I've seen permissive mode _block_ some operations and not issue any warnings about those blocked operations.
I can't give details about what has been blocked in the past. I've lost track of the records. I do recall a lot of it had to do with networking operations. This issue seems somehow related.
This is known, there are *some* special forms of SELinux filtering that can't be made 'permissive'. It works for most stuff, though. I think Dan has a blog post on it somewhere.
On Fri Apr 29 2016 11:05:47 GMT-0600 (MDT) Adam Williamson adamwill@fedoraproject.org wrote:
On Fri, 2016-04-29 at 09:49 -0700, Rick Stevens wrote:
As I understand it, permissive should allow all operations but give warnings while disabled means, well, disabled. However, I've seen permissive mode _block_ some operations and not issue any warnings about those blocked operations.
Does anything get logged when 'dontaudit' is disabled?
This is known, there are *some* special forms of SELinux filtering that can't be made 'permissive'. It works for most stuff, though. I think Dan has a blog post on it somewhere.
Improving/refreshing SELinux knowledge is never a bad thing ;) so I did some reading and have come across:
http://danwalsh.livejournal.com/67855.html
Is that it?
-
Adam Williamson -
Kamil Paral -
Rick Stevens -
Russel Winder -
Viorel Tabara