[SOLVED] "Oh no! Something has gone wrong." on Restart

[OtagoHarbour]

I have Debian 7.0 installed on a Dell Insperion. It has been working fine for a while but I have been making some changes suggested by lynis. The changes that may have caused this problem were most likely changes suggested here to deal with unreliable ntpq peers.

But, whatever caused it, when I rebooted the PC I got this message after seeing a watch cursor for a long time.

Code: Select all

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

I cannot even log on since it has not taken me to the log-in screen. Rebooting again just has the same result.

I have read that it is a bug in Gnu but no one seems that clear about how to fix it. I have also read that it is called "fail whale" and you should be able to get past it with alt-F4. When I use alt-F4, all I get is a normal, white cursor on a black background. The only way I have found to get past it is to do ctrl-alt-F4 and then I can log on but I do not get a Gnu desktop, only the command line.

Once I logged on to the command line, I looked at /var/log/messages and it includes the line.

Code: Select all

gnome-settings-[3770]:seg fault at 8 ip 00007f1f221a50f8 sp 00007fff6b5e80 error 4 in libpower.so

I have looked this error message up but did not find anything relating it to the "fail whale" error. Most of the posts I found were about VMs which I am not using.

Any assistance would be greatly appreciated.
OH

[dasein]

https://www.startpage.com/do/search?cmd ... ine0=v1all

(*grumble* Damn GNOME devs... this has got to be the most useless damn error message in the entire history of computing on this planet.)

[OtagoHarbour]

https://www.startpage.com/do/search?cmd ... ine0=v1all

I tried the suggestions I found there.

Code: Select all

sudo apt-get install firmware-linux-nonfree

gave

Code: Select all

Reading package lists... Done
Building Dependency Tree
Reading State information...Done
E: Unable to locate firmware-linux-nonfree

I tried ping and it shows I have Internet connectivity. The solution here was about the first thing I tried. I didn't work. Just gave a black screen with a white cursor.

I was able to start up the desktop with

Code: Select all

sudo startx 2>startx_root.errors

but it logged me in as root which seem pretty insecure. When I tried

Code: Select all

startx 2>startx_me.errors

and it gave me the "Oh no! Something has gone wrong." screen. Doing

Code: Select all

diff startx_root.errors startx_me.errors

just gave the time stamps.

I tried the solution here by doing

Code: Select all

ls -s /run/dbus /var/run/dbus
ls -s /run/ /var/run/
ls -s /run/lock /var/run/lock
sudo apt-get clean
sudo apt-get update
sudo dpkg --configure -a

but it made no difference.

There were no errors messages in /var/log/Xorg.1.log

Thanks,
OH

[mor]

The reason why you can't install "firmware-linux-nonfree" is most likely due to the fact you don't have the non-free component in your sources.

You will need to add non-free to /etc/apt/sources.list as explained here:
https://wiki.debian.org/SourcesList#Component

[OtagoHarbour]

The reason why you can't install "firmware-linux-nonfree" is most likely due to the fact you don't have the non-free component in your sources.

You will need to add non-free to /etc/apt/sources.list as explained here:
https://wiki.debian.org/SourcesList#Component

I edited /etc/apt/sources.list and added "contrib non-free" after every main (end of 8 lines).

Code: Select all

sudo apt-get install firmware-linux-nonfree

still results in

Code: Select all

Unable to locate package firmware-linux-nonfree

Code: Select all

sudo apt-get install firmware-linux-non-free

still results in

Code: Select all

Unable to locate package firmware-linux-non-free

Thanks,
OH

[dilberts_left_nut]

Code: Select all

sudo apt-get install firmware-linux-nonfree

still results in

Code: Select all

Unable to locate package firmware-linux-nonfree

Because it still doesn't appear in the list of packages available to apt-get.
Posting your sources.list might help.

Code: Select all

sudo apt-get install firmware-linux-non-free

still results in

Code: Select all

Unable to locate package firmware-linux-non-free

Because no such package exists (and consequently, it doesn't appear in the list of packages available to apt-get).

[OtagoHarbour]

Thank you for your reply.

Code: Select all

sudo apt-get install firmware-linux-nonfree

still results in

Code: Select all

Unable to locate package firmware-linux-nonfree

Because it still doesn't appear in the list of packages available to apt-get.
Posting your sources.list might help.

The following is the content of my sources.list after I edited the file

Code: Select all

# 

# deb cdrom:[Debian GNU/Linux 7.0.0 _Wheezy_ - Official amd64 NETINST Binary-1 20130504-14:43]/ wheezy main contrib non-free

#deb cdrom:[Debian GNU/Linux 7.0.0 _Wheezy_ - Official amd64 NETINST Binary-1 20130504-14:43]/ wheezy main contrib non-free

deb http://ftp.us.debian.org/debian/ wheezy main contrib non-free
deb-src http://ftp.us.debian.org/debian/ wheezy main contrib non-free

deb http://security.debian.org/ wheezy/updates main contrib non-free
deb-src http://security.debian.org/ wheezy/updates main contrib non-free

# wheezy-updates, previously known as 'volatile'
deb http://ftp.us.debian.org/debian/ wheezy-updates main contrib non-free
deb-src http://ftp.us.debian.org/debian/ wheezy-updates main contrib non-free

Thanks,
OH

[dilberts_left_nut]

That looks OK.
You probably didn't do 'apt-get update' to sync your package lists before trying to install?

[OtagoHarbour]

That looks OK.
You probably didn't do 'apt-get update' to sync your package lists before trying to install?

Thank you for your reply. Yes 'apt-get update' resolved the problem with installing firmware-linux-nonfree. Unfortunately, installing firmware-linux-nonfree did not fix the problem with my running "startx" as a non-root user. I still get the "Oh no! Something has gone wrong" message.

Since I do not have the problem when I use "sudo startx", I created a new user account and tried "startx" from that account. I still got the "Oh no! Something has gone wrong" message.

Thanks,
OH

[vbrummond]

Switch to KDE, it actually works. And hey presto! a nice fluid 60fps over the solid 5-10fps of Gnome. Gnome is a disease.

[OtagoHarbour]

Switch to KDE, it actually works. And hey presto! a nice fluid 60fps over the solid 5-10fps of Gnome. Gnome is a disease.

Thanks very much! That seems to have fixed the problem. (For now at least. I'm new to KDE.) I did

Code: Select all

sudo apt-get install kubuntu-full

I then edited /etc/X11/xinit/xinitrc and made

Code: Select all

exec startkde

The first uncommeted line

I then called

Code: Select all

startx

and my desktop started up. A few of questions I still have.

1/ When it started up, I got the message

Code: Select all

We are sorry.  Plasma Desktop Shell closed unexpectedly.  You can help us improve KDE software by reporting this error.

I reported the problem and it eventually went away. However, the first time, all I got was a cursor on a black background and this didn't change when I left it for at least half an hour. I restarted the PC and the KDE desktop asked me to log into the terminal mode. I entered

Code: Select all

startx

and the desktop started up. I got the "We are sorry" message again but this time the desktop stayed up. However, I did get the message "KDE power management system could not be initialized". I then restarted the PC and it went to the command line. I logged in and entered

Code: Select all

startx

and the desktop started up.

Question 2. Should it always boot to the command line so that I need to enter

Code: Select all

startx

to get the desktop. Gnome used to go straight to the desktop (when it worked).

Question 3. I have noticed that a third alternative is Xfce. I have never used it and have just started with KDE. Do you know which is more stable: KDE or Xfce.

Thanks again,
OH

[vbrummond]

Ubuntu is not Debian you need to go here for help: http://ubuntuforums.org/

[dasein]

I did

Code: Select all

sudo apt-get install kubuntu-full

Trying to mix Debian and Ubuntu won't work. Pick either one.

P.S. Xfce is a lightweight DE. If you're curious about it, try it. No one but you can decide whether it's what you prefer.

[OtagoHarbour]

I did

Code: Select all

sudo apt-get install kubuntu-full

Trying to mix Debian and Ubuntu won't work. Pick either one.

P.S. Xfce is a lightweight DE. If you're curious about it, try it. No one but you can decide whether it's what you prefer.

Installing kubuntu was a faux pas on my part. I had not heard of kubuntu before and thought it was a variety of KDE. I had been searching the web for how to install KDE.

I want to use Debian, not Ubuntu. Should I uninstall kubuntu before installing KDE or is there a danger that that may remove files used by Debian?

When I reboot the computer, I still only have the options of Debian and Debian-Restore. Also, when I go to the KDE desktop and enter

Code: Select all

uname -a

in a terminal, it says that I am using Debian.

In terms of Xfce versus KDE, I am mainly interested in what is most stable although I like the appearance of KDE. I would go with KDE if it is as stable as Xfce.

Thanks,
OH

[dasein]

I want to use Debian, not Ubuntu. Should I uninstall kubuntu before installing KDE or is there a danger that that may remove files used by Debian?

The problem is that your system is now in some weird, unknown (and ultimately unknowable) state. It may be fine, or it may be a ticking time bomb. No one can know for sure.

If you actually care about having a Stable system, the smart thing to do would be to reinstall, and to resolve never again to blindly type in stuff you read on Some Random Blog. (Or, at the very least, make a backup before you start mucking around.)

In terms of Xfce versus KDE, I am mainly interested in what is most stable although I like the appearance of KDE. I would go with KDE if it is as stable as Xfce.

Both DEs are mature and reasonably robust, but "stable" is a squishy concept. At the risk of seeming to repeat myself, you are the only person on the entire planet who can decide if any given DE is suitable for your particular uses, or even to your liking. (Some folks adore GNOME Shell, for instance, whereas I think it's the Devil's own spawn.) For any DE you're curious about, give it a try (either installed directly or in a VM) and decide for yourself.

[OtagoHarbour]

I want to use Debian, not Ubuntu. Should I uninstall kubuntu before installing KDE or is there a danger that that may remove files used by Debian?

The problem is that your system is now in some weird, unknown (and ultimately unknowable) state. It may be fine, or it may be a ticking time bomb. No one can know for sure.

The time bomb apparently had a short fuse since the system would not see an external drive that was in ntfs-g3 format. With CenOS I needed to enter

Code: Select all

sudo rpm -Uvh epel-release-6-8.noarch.rpm
sudo yum install ntfs-3g

to make it work but with Debian, it worked off the bat.

If you actually care about having a Stable system, the smart thing to do would be to reinstall, and to resolve never again to blindly type in stuff you read on Some Random Blog. (Or, at the very least, make a backup before you start mucking around.)

Actually I had saved an image of the disk, using Clonezilla, about a month ago (when I had pure Debian with the Gnome desktop) so I just restored that image to the disk. Gnome is working properly now but may not be if I do everything that Lynis suggests. That seemed to be the problem last time.

Thanks,
OH

[OtagoHarbour]

I have Debian 7.0 installed on a Dell Insperion. It has been working fine for a while but I have been making some changes suggested by lynis. The changes that may have caused this problem were most likely changes suggested here to deal with unreliable ntpq peers.

But, whatever caused it, when I rebooted the PC I got this message after seeing a watch cursor for a long time.

Code: Select all

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

I cannot even log on since it has not taken me to the log-in screen. Rebooting again just has the same result.

I have read that it is a bug in Gnu but no one seems that clear about how to fix it. I have also read that it is called "fail whale" and you should be able to get past it with alt-F4. When I use alt-F4, all I get is a normal, white cursor on a black background. The only way I have found to get past it is to do ctrl-alt-F4 and then I can log on but I do not get a Gnu desktop, only the command line.

Once I logged on to the command line, I looked at /var/log/messages and it includes the line.

Code: Select all

gnome-settings-[3770]:seg fault at 8 ip 00007f1f221a50f8 sp 00007fff6b5e80 error 4 in libpower.so

I have looked this error message up but did not find anything relating it to the "fail whale" error. Most of the posts I found were about VMs which I am not using.

Any assistance would be greatly appreciated.
OH

My initial way of resolving this was to cut the knot rather than untying it. That is, I just restored the clone of the disk from before I started messing around with lynis. This time, I think I might have resolved what caused the original problem.

I ran lynis and it said to change umask, in /etc/init.d/rc, to 027. Prior to doing that, Gnome was starting up fine. However, after I made that recommended change, restart produced the "Oh no!" screen. I used ctrl-alt-F4 to get the command line, edited /etc/init.d/rc and changed its umask value back to 022. When I restarted, the Gnome desktop started up just fine. So, setting umask, in /etc/init.d/rc, to 027 was what caused my problem