>Date: Tue, 3 Feb 2009 20:11:23 +0100
>From: Jan Pazdziora <[email protected]>
<snip>
>Therefore I do not think that putting SELinux into permissive mode
>cause any change in behaviour. So it must have been the missing
>Oracle setup steps that caused the problem.
I think, rather, making /etc/rhn readable by apache did it. BUT, as I keep
saying, I *still* get the 500 error when I try to manage software channels.
100% of the time. And after debugging via the logs, and reading some code, it
always comes back to
[Tue Feb 03 11:59:46 2009] [error] Execution of
/var/www/html/network/software/channels/manage/index.pxt failed at Tue Feb 3
11:59:46 2009: user_role_acl_test called with no $pxt->user authenticated at
/usr/lib/perl5/site_perl/5.8.8/RHN/Access.pm line 60.
And in the code of that perl module is
sub user_role_acl_test {
my $pxt = shift;
my $role = shift;
die "user_role_acl_test called with no \$pxt->user authenticated" unless
$pxt->user;
return $pxt->user->is($role) ? 1 : 0;
}
so what's happening is that, for some reason $pxt->user has not been
created/initialized, so the attempt to go there fails and dies.
Now, unless someone has any idea soon (I *really* need to get this damn thing
working, and soon, or my manager's not going to be happy), I may just mod the
function to just return true.
Yes, I'm *sure* that's not a Good Idea, since it would let anyone, I presume,
perform this, but for now, until I get this resolved, I can depend on the fact
that the machine is pretty secure, on an internal network, and only I, the
other admin, and our manager have the password to log onto the box, and that
only we know the password I've put in for the organization manager role.
Comments? Solutions?
mark
_______________________________________________
Spacewalk-list mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/spacewalk-list
