Unable to log into Ubuntu - no GUI!

Thanks for your question.

The swap partition does not hold any data you could loose since it is just an extension to the volatile RAM memory. It will be overwritten every time you boot.

If you believe there is really some problem with you swap, you can boot from your LiveCD and open a command line terminal and execute the command

sudo mkswap -c /dev/sda6

This will definitely re-initialise your swap and also check for bad block in the area.

However, I doubt the swap it really your problem. You might want to look at the file

/var/log/dmesg

or just execute the command "dmesg" and see if there are any error messages that might explain the behaviour.

I hope this helps.

Thanks Ralph for your answer. Here is what I tried following your suggestion:

First,
ubuntu@ubuntu:~$ sudo /var/log/dmesg
sudo: /var/log/dmesg: command not found

So I tried the next
Second,
dmesg
The output is huge and its not fair to post all of it here. But the truncated parts that show up with errors are as follows:
[ 62.631427] ohci1394: fw-host0: SelfID is inconsistent [0x807f8956/0x7f807629]
[ 62.720591] hdd: command error: status=0x51 { DriveReady SeekComplete Error }
[ 62.720596] hdd: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
[ 62.720600] ide: failed opcode was: unknown
[ 62.720603] end_request: I/O error, dev hdd, sector 1414448
[ 62.720606] Buffer I/O error on device hdd, logical block 353612
[ 62.722485] hdd: command error: status=0x51 { DriveReady SeekComplete Error }
[ 62.722490] hdd: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
[ 62.722492] ide: failed opcode was: unknown
[ 62.722495] end_request: I/O error, dev hdd, sector 1414448
[ 62.722498] Buffer I/O error on device hdd, logical block 353612
[ 62.730419] hdd: command error: status=0x51 { DriveReady SeekComplete Error }
[ 62.730424] hdd: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
[ 62.730427] ide: failed opcode was: unknown
[ 62.730430] end_request: I/O error, dev hdd, sector 1414448
[ 62.730433] Buffer I/O error on device hdd, logical block 353612
[ 62.732393] hdd: command error: status=0x51 { DriveReady SeekComplete Error }
[ 62.732397] hdd: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
[ 62.732400] ide: failed opcode was: unknown
[ 62.732403] end_request: I/O error, dev hdd, sector 1414448
[ 62.732406] Buffer I/O error on device hdd, logical block 353612
[ 62.734455] hdd: command error: status=0x51 { DriveReady SeekComplete Error }
[ 62.734459] hdd: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
[ 62.734462] ide: failed opcode was: unknown
[ 62.734464] end_request: I/O error, dev hdd, sector 1414448
[ 62.734467] Buffer I/O error on device hdd, logical block 353612
[ 62.736368] hdd: command error: status=0x51 { DriveReady SeekComplete Error }
[ 62.736372] hdd: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
[ 62.736375] ide: failed opcode was: unknown
[ 62.736377] end_request: I/O error, dev hdd, sector 1414448
[ 62.736380] Buffer I/O error on device hdd, logical block 353612
[ 62.905015] ieee1394: impossible ack_complete from node 65535 (tcode 4)
[ 62.905019] ieee1394: Current remote IRM is not 1394a-2000 compliant, resetting...
[ 62.907056] ohci1394: fw-host0: SelfID is inconsistent [0x807f8956/0x7f807629]
[ 62.921970] hdd: command error: status=0x51 { DriveReady SeekComplete Error }
[ 62.921974] hdd: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
[ 62.921977] ide: failed opcode was: unknown
[ 62.921979] end_request: I/O error, dev hdd, sector 1414440
[ 62.921982] Buffer I/O error on device hdd, logical block 353610
[ 62.924272] hdd: command error: status=0x51 { DriveReady SeekComplete Error }
[ 62.924276] hdd: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
[ 62.924280] ide: failed opcode was: unknown
[ 62.924282] end_request: I/O error, dev hdd, sector 1414444
[ 62.924286] Buffer I/O error on device hdd, logical block 353611
[ 62.974565] hdd: command error: status=0x51 { DriveReady SeekComplete Error }
[ 62.974569] hdd: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
[ 62.974572] ide: failed opcode was: unknown
[ 62.974574] end_request: I/O error, dev hdd, sector 1414440
[ 62.974577] Buffer I/O error on device hdd, logical block 353610
[ 62.976643] hdd: command error: status=0x51 { DriveReady SeekComplete Error }
[ 62.976647] hdd: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
[ 62.976650] ide: failed opcode was: unknown
[ 62.976652] end_request: I/O error, dev hdd, sector 1414444
[ 62.976655] Buffer I/O error on device hdd, logical block 353611
[ 62.985621] hdd: command error: status=0x51 { DriveReady SeekComplete Error }
[ 62.985626] hdd: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
[ 62.985629] ide: failed opcode was: unknown
[ 62.985632] end_request: I/O error, dev hdd, sector 1414448
[ 62.987767] hdd: command error: status=0x51 { DriveReady SeekComplete Error }
[ 62.987771] hdd: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
[ 62.987775] ide: failed opcode was: unknown
[ 62.987777] end_request: I/O error, dev hdd, sector 1414448

Does that look like hdd erroe, I don't know!!

Another error line I came across
[ 125.472039] powernow-k8: BIOS error - no PSB or ACPI _PSS objects
which I don't think is the problem.

Thirdly I have not tried to sort out the swap partitionas you say it is nto the problem. I believe you.

Ralph I tried the suggestion for re-initialising teh swap partition, ie:

sudo mkswap -c /dev/sda6

and this what came up;

ubuntu@ubuntu:~$ sudo mkswap -c /dev/sda6
Setting up swapspace version 1, size = 2146758 kB
no label, UUID=2a1056a5-9633-4215-8f29-09a72318d9af

After that i rebooted and the problem persisited. This time I noticed that teh NVIDIA splash screen came up, disappeared and then there was a peek of the desktop screen for split of a second and then back to the Nvidia splash screen and finally the black screen.

I am beginning to wonder if it is the nvidia drivers again. For so many of them have had problems after the recent kernel upgrade!!

The dmesg output looks like your hdd drive is dying or dead. Is it possible to boot up without this drive being connected?

What is the output of

sudo lshw -C disk

(the -C is a capital C !)

Thanks

Oh God! That looks like something very drastic. Hope its not true. Because I can still boot into windows. Remmber I said its dual boot. I can even access the linux partitions form windows. So the hard disk is fine. But somehow ubuntu thinks its not fine. Or atleast ubuntu cannot access the hard disk and therfore thinks that it is dead or dying!!

Anyway here is the output you asked for:

ubuntu@ubuntu:~$ sudo lshw -C disk
  *-disk
       description: SCSI Disk
       product: Maxtor 6B200M0
       vendor: ATA
       physical id: 0.0.0
       bus info: scsi@0:0.0.0
       logical name: /dev/sda
       version: BANC
       serial: B408AZEH
       size: 189GB
       capabilities: partitioned partitioned:dos
       configuration: ansiversion=5
     *-volume:0
          description: HPFS/NTFS partition
          physical id: 1
          bus info: scsi@0:0.0.0,1
          logical name: /dev/sda1
          capacity: 14GB
          capabilities: primary bootable
     *-volume:1
          description: Extended partition
          physical id: 2
          bus info: scsi@0:0.0.0,2
          logical name: /dev/sda2
          capacity: 175GB
          capabilities: extended partitioned partitioned:extended
  *-cdrom
       description: DVD writer
       product: ATAPI DVD DUAL 8X4X12
       physical id: 1
       bus info: ide@1.1
       logical name: /dev/hdd
       version: B3JC
       capabilities: packet atapi cdrom removable nonmagnetic dma lba iordy audio cd-r cd-rw dvd dvd-r
       configuration: mode=udma2
     *-disc
          physical id: 0
          logical name: /dev/hdd

Ok.. the error come from your DVD writer. You could disconnect it and try what happens without it. However, I probably would rather do the upgrade to the new kernel, since especially the mounting of drives not using UUIDs has been fixed there. It is also a security patch, so should be applied anyway.

If you could upgrade that and see if that fixes your problem (I assume here you have feisty).

Otherwise could you please also past the contents of your /etc/fstab file.

Thanks

Just in case, you want to do the upgrade from command line. Just execute the following commands

sudo apt-get update
sudo apt-get dist-upgrade

Just like you told me Ralph!

I ran the commands from the command line. First time around it did some updating. And then when i re booted, the problem persisted.

I tried booting form an earlier kernel image, and this time no updates at all.

One question, though, which is the most recent kernel image. The ones that show up on my grub list are:
2.6.20-16-generic and 2.6.20-15-generic including the recovery modes.

What next now??

2.6.20-16 is the most recent kernel for feisty.

Do you have any errors in the logfile /var/log/Xorg.0.log ?

I did check the Xorg.0.lof file. Its quite a long file I must say. No error in particular at all except one about a wacom device which I don't have connected to the PC and another about not finding come fonts. I don't think they are significant.

So I am updated to a new kernel. Do you think there is a way to regress into an old kernel version and then upgrade? Will that help?

If you update the new kernel, both kernels will be available. You probably can regress to an old kernel, but there is no tested procedure for it as fas as I know.

Ok Ralph, looks like its the end of the tunnel and I see no light from any corner.

Two issues come right away to my mind. I have almost made up my mind to do a reinstall. My personal data which is in the /home folder is in a different partition which I am planning to leave untouched (unformatted). Hopefully a re-install will help me use the home folder contents like before again. Any suggestions in this matter from your side is most welcome.

Secondly do I file this problem as a bug? If so, I need your help in doing so. Since we did not get to the bottom of the problem, I do not know what kind of a bug this will qualify as. Maybe the re-install will make things clearer ...maybe its a hardware problem and not a software one!

Could you please try to boot using the open source (nv) drivers? You can do this by modifying the driver line in your xorg.conf to read nv instead of nvidia.

Also, could you please post the full output of dmesg, preferable with the cd drive disconnected, or atleast the cd/dvd taken out of the drive. I've had these read errors on DVDs quite often, and by leaving the drive empty maybe dmesg would be a little less cluttered.

Also run "cat /var/logs/Xorg.0.log |grep EE" to see if there are any errors you missed earlier.

You should also run "df -h" and make sure you have free space in your root partition as well as your home directory. While the problem caused by the absolute lack of disk space is the inability to logon (gdm keeps coming back each time you try to login), you could just check it for the sake of it.

Keep in mind that if you use the nVidia beta drivers (or if for any reason you manually installed your drivers) you will have to reinstall them each time the kernel is upgraded.

In addition to all of these, editing your boot options to remove the quiet and splash parameters could let you see any errors during bootup. (when in the grub menu, select your ubuntu option and hit 'e', then select the line starting with kernel, hit e again, remove the two words from the end of the line. Don't make any other changes. These changes are only temporary so if it gets screwed up you can always reboot.

Thanks Shashank for your replies.

While reconfiguring the xorg.conf file I have chosen the nv drivers and the vesa drivers and none of them worked.

The full output of the dmesg was very long and hence I decided against posting it here. I am not adept enough to look into my CPU to be able to disconnect the cd drive. However I could indeed take out the CD before obtaining the output ...but unfortunately the situation is a little changed now!

The output of the cat call is as follows:
 cat /var/log/Xorg.0.log |grep EE
        (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
(II) Loading extension MIT-SCREEN-SAVER
(EE) AIGLX: Screen 0 is not DRI capable
(EE) xf86OpenSerial: Cannot open device /dev/input/wacom
(EE) xf86OpenSerial: Cannot open device /dev/input/wacom
(EE) xf86OpenSerial: Cannot open device /dev/input/wacom

My root partition is 15 gb and the entire installation took up 3+gb and the additional upgrades amde it go up by another 1+gb. My home partition is 50gb and about 35+ gb was available. So disk space unavailability certainly was not the problem.

I did see the nosplash option being discussed in the forums and gave it a try as well. But unfortunately it did not work out. In fact i used nano to edit the grub menu, but no luck!

After waiting for so long, I couldn't take it any more so I decide to go in for a reinstall. As I mentioned earlier, /home is located in a different partition. So I decide to leave it unformatted. So I went ahead and designated original partition where ubuntu was installed as the root partition and had it formatted as well. Call it serendipity or whatever ...I completely forgot to designate the partition where the previous /home was located as /home. So it was mounted simply as another partition and not as /home. The installation was smooth and everything else went on as I planned. I rebooted and everything was intact. The rig was working fine and suddenly it dawned on me that I had forgotten to place the /home folder on a different partition. I cursed myself and went onto to do a second re-install. This time around, I marked the partition I wanted as the /home partition. The installation was smooth. I re-booted. The splash screen showed up. The login screen showed up. I typed in my user id and password and hit enter. I was in, the screen blanked and then a completely white screen showed up with only the arrow cursor. Horrors of horrors from a completely black screen I had gone full circle onto a white screen!!

One peculiar issue that I noticed was when I brought the arrow cursor to one edge of the screen it would suddenly end up on the opposite end of the screen. This is exactly how it was on the first installation with the desktop effects enabled, read compiz., the so-called cube effect. Now it seemed as though the earlier settings that were stored in the /home folder and have been left untouched (unformatted) were playing havoc with the fresh installation. This is because the second installation when i had forgotten about the /home folder, everything worked perfect. There were no desktop effects as they do not come on by default!!

So what I would like to do now is to disable desktop effects through the command line and see if this helps.

Ralph and Shashank, sorry for being so verbose, but I would really appreciate your comments and suggestions in thsi matter. Nothing would interest me more than getting to the root of this bug!

At workplace I have access to a pc and attempted ubuntu install to dual boot with win xp. The installation process went smoothly. And I had feisty running soon after. The first thing I did was enable desktop effects to see if anything happened. All the wobbly windows and the cube were working. Wonderful, I thought to myself. Since its release there have been numerous updates and this is what showed up. I set up the update process and went off. When i returned after a while the pc was in hibernation. So i powered it on and the screen came to life. All the updates had been installed and a restart was all that was required to bring them into effect. Ok! I rebooted the machine ...surprise!surprise! The white screen of death!?

So I popped my live cd into the machine and rebooted. I searched a while in the Ubuntu forums and I came up with three options for the white screen error:
Option 1 - sudo apt-get install nvidia-glx
Option 2 - sudo dpkg-reconfigure xserver-xorg
and uncheck glx
Option 3 - killall compiz.real
and Ctrl+Alt+F7 to log into Gnome again.

This PC at workplace does not have an Nvidia card so the Options 2 and 3 is what I should be trying out. It is also a i386 based.

I reconfigured the xorg.conf file and unchecked glx. No luck. Then I unchecked 24-bit and chose 16-bit display instead. ( I read that in the posts somewhere along). This time i rebooted and the familiar brown was back again. I thought I had it solved. I tried Desktop Effects but it would not be enabled this time. I even went back to reconfigure the Xorg.conf file and had glx enabled and the 24-bit display chosen. Nothing went wrong but the desktop effects would not be enabled.

Looks to me like there is some conflict between the desktop effects / compiz with the most recent kernel update. Now I can't wait to get back home and see if I can sort out the white screen of death in a similar way.

This question was expired because it remained in the 'Open' state without activity for the last 15 days.