On Thu, 2012-08-30 at 17:46 +0000, Tyler Hicks wrote:
> > > * The handling of session_pid doesn't look right to me. Do we really
> > > want to
> > > blindly kill a PID that we stored in a global variable at some point
> > > in the
> > > past? I think there are probably PID wrap around issues here.
> >
> > I see where there could be issues there, but I'm unsure of how else to
> > handle the case of something like this:
> >
> > open_session();
> > do_something(&error);
> > if (error) {
> > close_session();
> > }
> >
> > Where if we don't go into the session, and we don't read the socket, the
> > fork'ed process would stay around forever.
>
> I'll have to give this some more thought. Can we at least issue the kill
> after dropping privileges? Doing it when privileged makes it scary.
Done.
The only one I've had issue with is it seems I don't have permissions to
do a setgroups(). Either my user or the guest user. So I've removed
that from the merge proposal. Not sure what I could be doing wrong
there other than it's just not allowed by default on Ubuntu.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1039634
Title:
[MIR] libpam-freerdp
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/libpam-freerdp/+bug/1039634/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
