Oh no!

[OainjaQakanj]

That's it. I got this in loggin. Too bad, since i never reach Networkmanager to have connection and there is no command line commands. This start after dowgrad gnome schemas. Is it SeLinux what kick me out. And it seem give linux status for droid phone. Hmmm

[DBelton]

?? oooookkkkaaayyyyy.. sooo what's the problem? :D I believe you forgot to post something up there :D

[kurtdriver]

You can also login (maybe), by pressing Alt+Ctrl+F2 at the same time. Connecting to the internet can be done from there. This (http://forums.fedoraforum.org/showthread.php?t=260520) is a good discussion of that subject.

Booting from a liveCD might be your best option, You can access the internet through it and then chroot to the hard drive.

[motnahp00]

?? oooookkkkaaayyyyy.. sooo what's the problem? :D I believe you forgot to post something up there :D

I didn't understand his post either.

[OainjaQakanj]

Oh no! Is ma problem. Log in and this message is given. This hapend after upgrade of schemas. How toget mobile connection on in commandwindow or disable error and kickout? I did read discussion but it is only for ethernet, command line tools are not installed in default for NM. P huu. I like to do and lear but w/o net you are in trouble, what is also downturn of F.

[kurtdriver]

Oh no! Is ma problem. Log in and this message is given.
What message? Please post it here.

Nmcli works from the command line, you can read the manual by typing, man nmcli.

[OainjaQakanj]

Oh no! Something has gone wrong. A problem has ocured and the system can't recover. Please log ott and try again.

[leigh123linux]

Oh no! Something has gone wrong. A problem has ocured and the system can't recover. Please log ott and try again.
Stop wasting our time with crap!
Post some relevant info or I will close this post and review your forum membership :(.

[OainjaQakanj]

Sorry, but that is the message! And in Europe we have freedom to speak. Was it /etc/log where messages goes?

[leigh123linux]

Sorry, but that is the message! And in Europe we have freedom to speak. Was it /etc/log where messages goes?


Post


su
cat /var/log/messages



and

dmesg


Also post


uname -r

as you haven't even posted which release :rolleyes:

[OainjaQakanj]

I nano in messages and 'gnome-shell.desktop' killed and respawning too quicly in Lovelock.

[leigh123linux]

I nano in messages and 'gnome-shell.desktop' killed and respawning too quicly in Lovelock.

Sorry I don't use Gnome3 so I'm unable to help.

[OainjaQakanj]

Is there a way to disble one? Network is the problem since can not control w/o gui. There is no in default command control like it should.

[leigh123linux]

Is there a way to disble one? Network is the problem since can not control w/o gui. There is no in default command control like it should.


The problem isn't network related.

gnome-shell is crashing.

[OainjaQakanj]

In my opinnin, F do not follow time. Half here use mobile broadband only. BFO and not text control in NM do not allow not much mowement. Well I bougt Linux magazine with Knoppix to resque.

[DBelton]

what do you get if you hit <ctrl>-<alt>-<F2>? That should take you to a command line console where you can get more information.

try that and if you can get anything, look in these files for information with your problem:

/var/log/messages

/var/log/Xorg.0.log

If that doesn't work, then when you boot, when the grub prompt comes up, hit enter and edit the kernel line.

add this to the kernel line and you should be able to boot to a command line without starting gnome


systemd.unit=multi-user.target

[kurtdriver]

I believe the error message you're getting is called a fail-whale. It would seem to be this bug. (https://bugzilla.gnome.org/show_bug.cgi?id=642497) I'm not sure that Knoppix can mount LVMs, If you have LVMs and Knoppix is unable to mount them, this won't work, but if Knoppix can, first get the internet going and then try to chroot into Fedora's mount point.
Find out where Knoppix mounts fedora's root filesystem, (if it does)
mount
Assuming that you just have one hard drive plugged in, it'll say something like /dev/sda2 on /media/B963-F283 type ext4 (rw)

Do the chroot, you'll need to be root for this, so type
su -
Then
chroot /media/mountpoint
This should put you inside of Fedora, in root directory on Fedora. You might try to update gsm
yum update gsm-* gnome-session-manager
as that seems to be where the bug is, if I understand this correctly.

Edit: Since you downgraded the schemas, you could update them while chrooted.

[OainjaQakanj]

It was in the end installation failure. I can now tell the story since typping via android phone is not my taste of flavor. I set up Debian sytem to see my other hd and even that I have make installation trough BoF this time something went wrong. If I look the structure of file system there is only wmliux and GRUB but not usual field structure what should appear since it is not crypted. I was planin to install Cnetworkmanager to HD and operate then mobile broadband from ctraltf2 command window. Anyhow even that file system was not correct the original eror was in Gnome3 shell crash. And lesson was that Cnetworkmanager need to have also in default, since there are increasing group of people what are using mobile broadband as an only connection they have.And you can not operate ctrlaltf2 basis or bootting in diferent runlevel to fix problems if there is no net connectio. And kurtdriver was right it was bug what was described in text to kick out and related Gnome3, I will comment if I have fix original nad learn aout it. BTW Deb Squeeze look like F10.

[marko]

what do you get if you hit <ctrl>-<alt>-<F2>? That should take you to a command line console where you can get more information.

try that and if you can get anything, look in these files for information with your problem:

/var/log/messages

/var/log/Xorg.0.log

If that doesn't work, then when you boot, when the grub prompt comes up, hit enter and edit the kernel line.

add this to the kernel line and you should be able to boot to a command line without starting gnome


systemd.unit=multi-user.target


Are you sure about the using systemd.unit? Fedora 14 (Lovelock) was not defaulted to systemd, it uses upstart

[OainjaQakanj]

As I used systemd it works, but why you need that? Since you got ctrlaltf2 anyway to operate?

[DBelton]

Are you sure about the using systemd.unit? Fedora 14 (Lovelock) was not defaulted to systemd, it uses upstart


systemd is the default in Fedora 15. (we are in the F15 Development forum :D )


and OainjaQakanj, if you couldn't get to the console with ctrl-alt-f2, the you could have tried booting in multi-user mode and not tried starting the graphical stuff in case that is where the hang up was.

[bronson]

F15 was working for me last night. I yum updated this morning and now I get the same thing as OainjaQakanj ("Oh no! Something has gone wrong." with a sad compy icon).

/var/log/messages contains: "gnome-session[1820]: WARNING: App 'gnome-shell.desktop' respawning too quickly".

Ten seconds earlier I see:

gnome-session[1744]: WARNING: Invalid method call: Argument 0 is specified to be of type "boolean" but is actually of type "uint32"#012
gnome-session[1744]: WARNING: Client '/org/gnome/SessionManager/Client1' failed to reply before timeout
gnome-session[1744]: WARNING: Invalid method call: Argument 0 is specified to be of type "boolean" but is actually of type "uint32"#012

I also see a bunch of:

NetworkManager[1020]: <error> [1201335270.114420] [nm-manager.c:1368] user_proxy_init(): could not init user settings proxy: (3) Could not get owner of name 'org.freedesktop.NetworkManagerUserSettings': no such name

Tried creating a brand new user and I get the same thing. Don't see anything in Xorg.0.log.

Hope this helps someone figure this out!

---------- Post added at 11:19 AM ---------- Previous post was at 11:08 AM ----------

Also, when I try to yum upgrade in the console, it bombs out with:

Error: Package: 1:control-center-2.91.92-4.fc15.i686 (updates testing) Requires: libnm-glib.so.4
Error: Package: anaconda-15.25-1.fc15.i686 (updates-testing) Requires: libnm-glib.so.4
Error: Package: anaconda-15.25-1.fc15.i686 (updates-testing) Requires: libnm-util.so.2

It may or may not be related.

[Dan]

Hmmmm.

I just got notice of 30 some updates since last night, but with several threads about broken systems cropping up ... I'm thinking I'll wait 24 or so hours before I apply them.

[mariuszs]

I have similar error and cannot login into gnome after latest updates:

NetworkManager[1079]: <error> [1301339634.98618] [nm-manager.c:1368] user_proxy_init(): could not init user settings proxy: (3) Could not get owner of name 'org.freedesktop.NetworkManagerUserSettings': no such name


2.6.38-1.fc15.i686.PAE
FC15

[Dan]

Dang! Ran updates an hour ago, now I can't get it to boot all the way to a useable GUI.

I get wallpaper, and a nifty little mouse pointer ... and that's it.

Ayup! It's a fedora Alpha all right! <..:p..>

[bronson]

Hit control-alt-f2, logged in as root, yum upgrade, and I'm back in business. Ayup, 48 hour Fedora flu.

Really liking the integrated networkmanager!

[tuxor]

Really liking the integrated networkmanager!
Yes, fits in quite nicely. But I can't change VPN settings or add VPN configs (I guess, it's a permissions problem). I'm really glad, I set up VPN before updating. If I hadn't I couldn't set it up now and therefore had no internet connection :-/

[ovadyah]

Dang! Ran updates an hour ago, now I can't get it to boot all the way to a useable GUI.

I get wallpaper, and a nifty little mouse pointer ... and that's it.

Ayup! It's a fedora Alpha all right! <..:p..>

Same for me last night, happened to do updates while listening to a lecture, ooops :doh:. I could not run updates for a conflict,
so went to bed and ran update (in root) this morning (I could get to login from cntl-alt-f3) and yum updated and now back to working very well :) Yea Alpha!

[rlengland]

I'm seeing the fail whale still. I have not been able to get the Gnome GUI to start on this Dell Precision 360


lspci indicates
nVidia Corp. NV11GL (Quadro2 MXR/EX/Go) (rev 02)

/var/log/gdm/:0.log contains, as the last line:

NOUVEAU(0): NVLeaveVT is called.

But no errors are indicated.


Latest directory in /var/spool/abrt/ccpp* file contained:
reason file contains
Process /usr/bin/gnome-shell was killed by signal 11 (SIGSEGV)

Is there any one with a similar machine that can verify that this should work?

---------- Post added at 10:51 AM ---------- Previous post was at 10:09 AM ----------

I was suspicious, given the age of the system, that the graphics card was not going to be able to support Gnome 3 but was assuming that the fail over would simply eliminate that option.

My assumption was wrong, at least in this case. When I added "nomodeset" to my boot line in grub there was a complaint about the graphics card and the option to continue.

Is this the expect behaviour? From what I've had time to read I suspect not.

Opinions?

~~R