Hi,
after a boot into Gnome and opening a root console, switching to runlevel 3 (or whatever) fails. "init 3" or "telinit 3" does simply nothing, and the box begins to lock up. There's no problem to boot into runlevel 3 or whatever.
Does anybody here encounter the same, and know a solution?
On 15.11.2010, Heinz Diehl wrote:
after a boot into Gnome and opening a root console, switching to runlevel 3 (or whatever) fails. "init 3" or "telinit 3" does simply nothing, and the box begins to lock up. There's no problem to boot into runlevel 3 or whatever.
Ok, I'm back on F13. Couldn't figure out what's the cause of this nasty behaviour, will try F14 in some months again.
Heinz Diehl wrote:
Hi,
after a boot into Gnome and opening a root console, switching to runlevel 3 (or whatever) fails. "init 3" or "telinit 3" does simply nothing, and the box begins to lock up. There's no problem to boot into runlevel 3 or whatever.
Does anybody here encounter the same, and know a solution?
The next time I'm booted on 14 I'll try that. I did try it in a VM and it worked back and forth between 5 and 3 multiple times.
On Wed, 17 Nov 2010 14:08:43 -0500, Bill wrote:
Heinz Diehl wrote:
Hi,
after a boot into Gnome and opening a root console, switching to runlevel 3 (or whatever) fails. "init 3" or "telinit 3" does simply nothing, and the box begins to lock up. There's no problem to boot into runlevel 3 or whatever.
Does anybody here encounter the same, and know a solution?
The next time I'm booted on 14 I'll try that. I did try it in a VM and it worked back and forth between 5 and 3 multiple times.
"init 1" in GNOME doesn't work https://bugzilla.redhat.com/640925
Could be the wrong component (after having been reassigned), but the current owner has yet to respond. :/
When I asked about it on test list on Oct 1st, nobody added any insight: http://lists.fedoraproject.org/pipermail/test/2010-October/094304.html
On 17.11.2010, Michael Schwendt wrote:
"init 1" in GNOME doesn't work https://bugzilla.redhat.com/640925
Could be the wrong component (after having been reassigned), but the current owner has yet to respond. :/
When I asked about it on test list on Oct 1st, nobody added any insight: http://lists.fedoraproject.org/pipermail/test/2010-October/094304.html
Is there some new information on this one?
I did a "yum update" today on my F14 machine, the bug is still present and unresolved. This is deeply disgusting, I have to switch the runlevels quite often and this bug renders F14 completely unusuable for me.
On Mon, 29 Nov 2010 21:16:42 +0100, Heinz wrote:
On 17.11.2010, Michael Schwendt wrote:
"init 1" in GNOME doesn't work https://bugzilla.redhat.com/640925
Could be the wrong component (after having been reassigned), but the current owner has yet to respond. :/
When I asked about it on test list on Oct 1st, nobody added any insight: http://lists.fedoraproject.org/pipermail/test/2010-October/094304.html
Is there some new information on this one?
I did a "yum update" today on my F14 machine, the bug is still present and unresolved. This is deeply disgusting, I have to switch the runlevels quite often and this bug renders F14 completely unusuable for me.
New information? Not that I know of. Except, there's a second ticket about it, filed a month later for "gdm": http://bugzilla.redhat.com/649940
On 01.12.2010, Michael Schwendt wrote:
New information? Not that I know of. Except, there's a second ticket about it, filed a month later for "gdm": http://bugzilla.redhat.com/649940
Yes, I know.
This bug is ugly as hell, and in absence of a working fix or workaround, I'll be forced to use another distro on my machines...
On Wed, 1 Dec 2010 15:50:42 +0100 Heinz Diehl htd@fritha.org wrote:
On 01.12.2010, Michael Schwendt wrote:
New information? Not that I know of. Except, there's a second ticket about it, filed a month later for "gdm": http://bugzilla.redhat.com/649940
Yes, I know.
This bug is ugly as hell, and in absence of a working fix or workaround, I'll be forced to use another distro on my machines...
I noticed this today, and thought I would do a trial to test it, see if I too have the error. I'm running F14 x86_64, fully updated to updates-testing as of today.
I boot into runlevel 5, login to Gnome as a user. Switch to a console (4, if that matters) (*not* an xterm), login as root. Run telinit 1 and the system switches to runlevel 1 Run telinit 5 and the system switches to runlevel 5 Run telinit 3 and the system switches to runlevel 3 Run telinit 5 and the system switches to runlevel 5
The error must have some dependency on the hardware you are running, or I have a software combination that negates it.
On 01.12.2010, stan wrote:
The error must have some dependency on the hardware you are running, or I have a software combination that negates it.
I had F14 on four quite different machines:
- an older Celeron based Laptop - Asus UJ45C (core i5) - AMD Quadcore - an old P4
The same behaviour was present on all 4 machines: booting into Gnome, opening a console, "su root", init 3 -> nothing happens, the logout and shutdown buttons disappear and the machine becomes unstable. This was not present in F13, which were on those 4 machines about 1/2 year ago.
Three of the machines are (aehh, were; they contain opensuse 11.3 and Gentoo now) freshly installed/updated ones, the Celeron laptop is preupgraded. The bug remains the same, though.
On 3 December 2010 11:36, Heinz Diehl htd@fritha.org wrote:
The same behaviour was present on all 4 machines: booting into Gnome, opening a console, "su root", init 3 -> nothing happens, the logout and shutdown buttons disappear and the machine becomes unstable. This was not present in F13, which were on those 4 machines about 1/2 year ago.
I get slightly different behaviour with a KDE desktop (started from gdm though). In my case I tried "init 1" and most processes, including the X server, were eventually killed. However, I had waaay too many processes left by the time I got to the single user prompt. As I recall, there were about half a dozen of which the only one I can remember is the openoffice.org quick start thing. Obviousy the only things that should have been there are the single user shell and the kernel-context processes.
This looks like a problem with the upstart-init not actually killing off everything it should be killing off.
jch
On Wed, 1 Dec 2010 11:05:44 -0700, stan wrote:
On Wed, 1 Dec 2010 15:50:42 +0100 Heinz Diehl wrote:
On 01.12.2010, Michael Schwendt wrote:
New information? Not that I know of. Except, there's a second ticket about it, filed a month later for "gdm": http://bugzilla.redhat.com/649940
Yes, I know.
This bug is ugly as hell, and in absence of a working fix or workaround, I'll be forced to use another distro on my machines...
I noticed this today, and thought I would do a trial to test it, see if I too have the error. I'm running F14 x86_64, fully updated to updates-testing as of today.
I boot into runlevel 5, login to Gnome as a user. Switch to a console (4, if that matters) (*not* an xterm), login as root.
Well, the original test-case is to use a graphical terminal, not a virtual console.
On Sat, 4 Dec 2010 22:37:33 +0100 Michael Schwendt mschwendt@gmail.com wrote:
On Wed, 1 Dec 2010 11:05:44 -0700, stan wrote:
I boot into runlevel 5, login to Gnome as a user. Switch to a console (4, if that matters) (*not* an xterm), login as root.
Well, the original test-case is to use a graphical terminal, not a virtual console.
That explains it. I (almost) always do root operations in a virtual console if they are command line. If I have a time when I don't mind rebooting, I'll try this the original way. But it probably won't work, given your experience.
-
Bill Davidsen -
Heinz Diehl -
John Haxby -
Michael Schwendt -
stan