Baby Blue Screen Of Permanent Death

Home » CentOS » Baby Blue Screen Of Permanent Death
CentOS 10 Comments

I clicked on a window to try to raise it and get a good look. Instead, my screen turned the baby blue mentioned in the subject and I could do nothing else. The mouse cursor was not even visible. It’s possible an alternate terminal was available, but I didn’t think of it at the time. Eventually I hit the reset button. My Daktech machine has a reset button separate from the power button. The blue went away, but it hung after the CentOS logo appeared. The watch mouse cursor appeared and stayed. I could move it with the mouse, but that was all I could do on that terminal. The possilbility that it might have some major file system checking to do caused me to wait for a few hours, but the nothing changed.

It’s my expectation that when the logo appears, any file system checking has already been done. Is that correct?

What is the incantation to get a text view of the start up so that I can get a hint of what CentOS is waiting for?

I manged to boot an old Fedora 14 install, but could not login because I had forgotten the passwords. I’m doing this from Knoppix.

On another terminal, I can login to CentOS as root, but do not know what else to do.

root@Knoppix:~# fsck LABEL=sata400-3-slash fsck 1.40-WIP (14-Nov-2006)
e2fsck 1.40-WIP (14-Nov-2006)
sata400-3-slash: clean, 364784/977280 files, 2005224/3907811 blocks root@Knoppix:~# fsck LABEL=sata400-5-var fsck 1.40-WIP (14-Nov-2006)
e2fsck 1.40-WIP (14-Nov-2006)
sata400-5-var: clean, 5650/428240 files, 289518/1710914 blocks root@Knoppix:~# fsck LABEL=sata400-12-homes fsck 1.40-WIP (14-Nov-2006)
e2fsck 1.40-WIP (14-Nov-2006)
sata400-12-homes: clean, 279637/5242880 files, 10138872/20971520 blocks root@Knoppix:~#

They ran rather quickly. Should I trust the result?
Here is my fstab:
#
# /etc/fstab
# Created by anaconda on Sun Jun 24 18:14:46 2012
#
# Accessible filesystems, by reference, are maintained under ‘/dev/disk’
# See man pages fstab(5), findfs(8), mount(8) and/or blkid(8) for more info
#
LABEL=sata400-3-slash / ext3 defaults 1 1
LABEL=sata400-5-var /var ext3 defaults 1 2
LABEL=sata400-12-homes /homes ext3 defaults 1 2
LABEL=ide5-swap swap swap defaults 0 0
LABEL=sata400-6-swap swap swap defaults 0 0
tmpfs /dev/shm tmpfs defaults 0 0
devpts /dev/pts devpts gid=5,modeb0 0 0
sysfs /sys sysfs defaults 0 0
proc /proc proc defaults 0 0

As you can see, I changed it to use labels instead of UUIDs.
/home is a soft link into /homes .

How do I change passwords on an install, e.g. my F14, into which I cannot login?
I can edit the files from either Knoppix or from a CentOS terminal.

10 thoughts on - Baby Blue Screen Of Permanent Death

  • edit /mnt/…/etc/shadow and delete the password hash between the first two :’s, so it looks like username::…..

    then boot that system, log into it without any password, and set a password immediately.

  • To view the startup, when you boot, hit any key, then hit e as in edit (I
    think–otherwise, just use the arrow key to get down to the line beginning with Linux and when you highlight that line hit e to edit.

    At the end of that line you will see rhgb quiet. Remove those two words, hit enter, which should take you back the main menu, and then hit b for boot. (I’m doing this from memory, but I think there are instructions on the screen once you get to the e for edit part.

    That will give you text, at least for awhile. However, if you boot into GUI mode, you will also have to change /etc/inittab, where it says id:5:default—change the 5 to a 3, which will boot into text mode and end at a login prompt. Once you log in, you can type startx to start the GUI.

  • Here is the last bit of the startup output, copied by hand:
    Starting abrt daemon OK
    user had insufficient privilege Starting crond OK
    Starting atd OK
    Starting virt-who OK
    Starting libvird daemon OK
    Bridge firewalling registered tun: Universal TUN/TAP device driver 1.6
    tun: (C) 1994-2004 Max Krasnyansky
    device virb0-nic entered promiscous mode virb0: starting userspace STP failed, starting kernel STP
    Ebtable v20 registered lo: Disabled Provacy Extensions
    _

    At that point the watch mouse cursor appeared. The text stayed. After going back and forth between virtual erminals, the mouse cursor was gone.

    Does this help with diagnosis?

  • That worked after I also deleted the x from passwd. I still do not know how to fix anything. I forgot to edit the kernal line the last time I boot CentOS, but if the output oof dmesg is an indication, it got a bit further this time:

    ip6_tables: (C) 2000-2006 Netfilter Core Team nf_conntrack version 0.5.0 (16384 buckets, 65536 max)
    ip_tables: (C) 2000-2006 Netfilter Core Team e100 0000:02:08.0: eth0: NIC Link is Up 100 Mbps Full Duplex ADDRCONF(NETDEV_UP): eth0: link is not ready ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready p4-clockmod: P4/Xeon(TM) CPU On-Demand Clock Modulation available RPC: Registered named UNIX socket transport module. RPC: Registered udp transport module. RPC: Registered tcp transport module. RPC: Registered tcp NFSv4.1 backchannel transport module. SELinux: initialized (dev rpc_pipefs, type rpc_pipefs), uses genfs_contexts eth0: no IPv6 routers present
    802.1Q VLAN Support v1.8 Ben Greear
    All bugs added by David S. Miller
    SELinux: initialized (dev autofs, type autofs), uses genfs_contexts SELinux: initialized (dev autofs, type autofs), uses genfs_contexts SELinux: initialized (dev autofs, type autofs), uses genfs_contexts eth0: no IPv6 routers present Bridge firewalling registered tun: Universal TUN/TAP device driver, 1.6
    tun: (C) 1999-2004 Max Krasnyansky
    device virbr0-nic entered promiscuous mode virbr0: starting userspace STP failed, starting kernel STP
    Ebtables v2.0 registered SELinux: initialized (dev mqueue, type mqueue), uses transition SIDs SELinux: initialized (dev proc, type proc), uses genfs_contexts SELinux: initialized (dev mqueue, type mqueue), uses transition SIDs lo: Disabled Privacy Extensions SELinux: initialized (dev proc, type proc), uses genfs_contexts

    The SELinux line is new.

    Is there somewhere I can put print statements to try to figure out what is going on?

  • Have you looked at /var/log/Xorg.0.log file – it sounds like there is a problem with X.

    Also someone mentioned editing /etc/inittab and setting the run level to 3. id:5:initdefault: – change the 5 to a 3.
    If it boots to a non-gui login prompt then the problem is definitely X related.

    HTH, Steve

  • n Thu, 24 Oct 2013, Steve Clark wrote:

    Not yet, but I will.

    I get a non-gui login prompt on other virtual terminals, so that would seem to be the case.

  • I have now, but I do not know what to do with the information. I understand line 15 and 111. Any ideas?

    [root@localhost log]# grep -n EE Xorg.0.log
    15: (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
    111:[ 57.736] Initializing built-in extension MIT-SCREEN-SAVER
    692:[ 59.651] (EE) PreInit returned 8 for “HDA ATI HDMI HDMI/DP,pcm=3”
    694:[ 59.651] (EE) config/hal: NewInputDeviceRequest failed (8)
    [root@localhost log]#

    [ 59.628] (**) HDA ATI HDMI HDMI/DP,pcm=3: always reports core events
    [ 59.628] (**) evdev: HDA ATI HDMI HDMI/DP,pcm=3: Device: “/dev/input/event6”
    [ 59.635] (–) evdev: HDA ATI HDMI HDMI/DP,pcm=3: Vendor 0 Product 0
    [ 59.635] (WW) evdev: HDA ATI HDMI HDMI/DP,pcm=3: Don’t know how to use device
    [ 59.651] (EE) PreInit returned 8 for “HDA ATI HDMI HDMI/DP,pcm=3”
    [ 59.651] (II) UnloadModule: “evdev”
    [ 59.651] (EE) config/hal: NewInputDeviceRequest failed (8)
    [ 59.661] AUDIT: Wed Oct 23 15:22:43 2013: 2401: client 1 connected from local host ( uid=0 gid=0 pid#99 )
    Auth name: MIT-MAGIC-COOKIE-1 ID: 364
    [ 59.722] (II) RADEON(0): EDID vendor “AOC”, prod id 42881
    [ 59.722] (II) RADEON(0): Using EDID range info for horizontal sync
    [ 59.722] (II) RADEON(0): Using EDID range info for vertical refresh
    [ 59.722] (II) RADEON(0): Printing DDC gathered Modelines:
    [ 59.722] (II) RADEON(0): Modeline “1280×1024″x0.0 135.00 1280 1296 1440 1688 1024 1025 1028 1066 -hsync -vsync (80.0 kHz eP)
    [ 59.722] (II) RADEON(0): Modeline “800×600″x0.0 40.00 800 840 968 1056 600 601 605 628 +hsync +vsync (37.9 kHz e)

    The last three lines are:
    [ 1594.436] (II) evdev: ImPS/2 Generic Wheel Mouse: Close
    [ 1594.436] (II) UnloadModule: “evdev”
    [ 1594.487] Server terminated successfully (0). Closing log file.

    The string ucc does not occur elsewhere in the file.

    In case it helps:
    [root@localhost log]# grep -n hal Xorg.0.log
    566:[ 59.462] (II) config/hal: Adding input device ImPS/2 Generic Wheel Mouse
    585:[ 59.509] (**) Option “config_info” “hal:/org/freedesktop/Hal/devices/platform_i8042_i8042_AUX_port_logicaldev_input”
    592:[ 59.520] (II) config/hal: Adding input device Logitech Logitech Illuminated Keyboard
    609:[ 59.525] (**) Option “config_info” “hal:/org/freedesktop/Hal/devices/usb_device_46d_c318_noserial_if1_logicaldev_input”
    623:[ 59.569] (II) config/hal: Adding input device Logitech Logitech Illuminated Keyboard
    630:[ 59.573] (**) Option “config_info” “hal:/org/freedesktop/Hal/devices/usb_device_46d_c318_noserial_if0_logicaldev_input”
    638:[ 59.584] (II) config/hal: Adding input device Sleep Button
    645:[ 59.589] (**) Option “config_info” “hal:/org/freedesktop/Hal/devices/computer_logicaldev_input_0”
    653:[ 59.600] (II) config/hal: Adding input device Power Button
    660:[ 59.606] (**) Option “config_info” “hal:/org/freedesktop/Hal/devices/computer_logicaldev_input”
    668:[ 59.615] (II) config/hal: Adding input device Macintosh mouse button emulation
    679:[ 59.619] (**) Option “config_info” “hal:/org/freedesktop/Hal/devices/computer_logicaldev_input_1”
    686:[ 59.628] (II) config/hal: Adding input device HDA ATI HDMI HDMI/DP,pcm=3
    694:[ 59.651] (EE) config/hal: NewInputDeviceRequest failed (8)
    [root@localhost log]#

    Any ideas?

  • Absent other ideas, I might try re-installing CentOS or re-installing X. I have a pretty good idea how to do the former, but the latter might be harder despite, in principle, being less intrusive.

    My understand is that unistalling X will normally take all its dependents will it. That means That I will have to re-install said dependents, possibly listing each one separately. My thought is to /usr/bin/script yum’s output into a file. I’d use the file to produce another yum commmand to reinstall X’s dependents. Does that seem like a good plan?

    To reinstall CentOS, I would back up things that needed backing up. I would use yum to list all installed packages. I would use my grub menu to select the same stanza that I used to do a net-install of CentOS. It might be nice to use a kickstart file, but I do not know how.

    After the install, I would restore the directory that listed all my repositories. This is a step I am not sure about. I have a vague recollection that that is not sufficient. What else would I need to do?
    I would use the yum listing to install everything I have now. Does this seem like a good plan?

  • Would yum reinstall X
    reinstall X’s dependents?
    If not, I still need the list of package names that depend on X. Telling yum, at least tentatively, to remove X would seem to be the simplest way.

    I really do not want to do any more experiments than I have to.

LEAVE A COMMENT