[luv-main] ubuntu hang after upgrade natty to oniric
Hello, When I reboot my computer it hangs with a flashing cursor on a black screen. For ever. However the system hasn't crashed, it just appears to be hanging, e.g. numlock light will toggle. Ctrl+Alt+Delete also works fine. So I reboot it into safe mode, and selected mount filesystems read/write. It does this and then hangs in exactly the same way. So I reboot and enter a command line. I can remount / rewrite and mount everything else fine. I then run "dhclient eth0" and after about 60 seconds I see these messages: [ 88.014507] r8169 0000:02:00.0: eth0: unable to load firmware patch rtl_nic/rtl8168d-2.fw (-2) [ 88.021533] r8169 0000:02:00.0: eth0: link down [ 88.021586] r8169 0000:02:00.0: eth0: link down [ 88.021894] ADDRCONF(NETDEV_UP): eth0: link is not ready [ 90.310097] r8169 0000:02:00.0: eth0: link up [ 90.310433] ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready The firmware exists in the required location: brian@aquitard:~$ ls -l /lib/firmware/rtl_nic/rtl8168d-2.fw -rw-r--r-- 1 root root 1324 2011-08-23 23:23 /lib/firmware/rtl_nic/rtl8168d-2.fw It is possible when mounting filesystems I missed an important step that is required to load the firmware properly (maybe the step that causes the hangs?). However at this point, with the network up, I can continue booting fine. Any ideas? How can I find out what is causing the hang? Is there anyway of creating another virtual console within recovery mode? Thanks -- Brian May <brian@microcomaustralia.com.au>
Hope these ideas are some use. On 27 October 2011 13:06, Brian May <brian@microcomaustralia.com.au> wrote:
Hello,
When I reboot my computer it hangs with a flashing cursor on a black screen. For ever. However the system hasn't crashed, it just appears to be hanging, e.g. numlock light will toggle. Ctrl+Alt+Delete also works fine.
So I reboot it into safe mode, and selected mount filesystems read/write. It does this and then hangs in exactly the same way.
So I reboot and enter a command line. I can remount / rewrite and mount everything else fine. I then run "dhclient eth0" and after about 60 seconds I see these messages:
[ 88.014507] r8169 0000:02:00.0: eth0: unable to load firmware patch rtl_nic/rtl8168d-2.fw (-2) [ 88.021533] r8169 0000:02:00.0: eth0: link down [ 88.021586] r8169 0000:02:00.0: eth0: link down [ 88.021894] ADDRCONF(NETDEV_UP): eth0: link is not ready [ 90.310097] r8169 0000:02:00.0: eth0: link up [ 90.310433] ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
Sounds like it loaded it eventually? Does ifconfig eth0 show it is up?
The firmware exists in the required location:
brian@aquitard:~$ ls -l /lib/firmware/rtl_nic/rtl8168d-2.fw -rw-r--r-- 1 root root 1324 2011-08-23 23:23 /lib/firmware/rtl_nic/rtl8168d-2.fw
It is possible when mounting filesystems I missed an important step that is required to load the firmware properly (maybe the step that causes the hangs?). However at this point, with the network up, I can continue booting fine.
Obvious question - but perhaps not relevant - do you have any networking file system mounts? If the network taking a while to start or not loading is the issue? cat /etc/fstab
Any ideas? How can I find out what is causing the hang?
You should be able to boot in non-graphically mode and get the console output showing the steps. e.g. Does you boot load have a "quiet" or "rgb" or something telling it to do a graphical boot which hides these steps? Perhaps detailing your booting system grub, grub2, lilo et cetera and configuration file ... can allow suggestions
Is there anyway of creating another virtual console within recovery mode?
Don't know what recovery mode is perhaps it is single user mode? Perhaps you can try to switch to another virtual console after the graphics have started (before it has hung?) You can try switching to multiuser mode via telinit - e.g. telinit 2 or telinit 3 and see if you can see what is failing. Andrew
On 27 October 2011 15:28, Andrew Worsley <amworsley@gmail.com> wrote:
Hope these ideas are some use.
Thanks!
Sounds like it loaded it eventually? Does ifconfig eth0 show it is up?
Yes, networking works perfectly when I bring it up manually. Am using it now to send this email in fact.
Obvious question - but perhaps not relevant - do you have any networking file system mounts?
No network file systems.
If the network taking a while to start or not loading is the issue?
Have wondered if it is a network issue, not entire convinced either way yet. It would seem strange that mounting the filesystems by the recovery menu should also try to bring up the network, yet this process hangs unless I do it manually at the command line.
You should be able to boot in non-graphically mode and get the console output showing the steps. e.g. Does you boot load have a "quiet" or "rgb" or something telling it to do a graphical boot which hides these steps? Perhaps detailing your booting system grub, grub2, lilo et cetera and configuration file ... can allow suggestions
I boot with the boot option in Grub that has the recover. This turns on all the kernel debugging stuff. However the problem happens after user space starts.
Don't know what recovery mode is perhaps it is single user mode? Perhaps you can try to switch to another virtual console after the graphics have started (before it has hung?)
No virtual consoles configured in recovery mode.
You can try switching to multiuser mode via telinit - e.g.
telinit 2 or telinit 3
and see if you can see what is failing.
Hmmm. Could try that I guess. However Ubuntu start up scripts seem to be very quiet, and I think that is the problem with trying to work out what step it is failing on. -- Brian May <brian@microcomaustralia.com.au>
Brian May <brian@microcomaustralia.com.au> wrote:
I boot with the boot option in Grub that has the recover. This turns on all the kernel debugging stuff. However the problem happens after user space starts.
You should be able to get boot messages on tty1 after user space starts. Removing any quiet option etc., from the parameters that the boot loader passes to the kernel is necessary, however.
Brian May <brian@microcomaustralia.com.au> wrote:
It is possible when mounting filesystems I missed an important step that is required to load the firmware properly (maybe the step that causes the hangs?). However at this point, with the network up, I can continue booting fine.
Any ideas? How can I find out what is causing the hang?
Watch the boot messages, modifying your boot loader's configuration to ensure that they are displayed if this is not the case by default.
Is there anyway of creating another virtual console within recovery mode?
If you mean single user mode, then yes, you can run, e.g., /sbin/getty 38400 tty2 (note that /sbin would normally be in your $PATH thus strictly needn't be specified as above). Then you should be able to type alt-f2 to switch to that console and get a login prompt. If you have the openvt command then openvt -c 2 bash should work too, and obviously you can substitue any virtual console number for 2 in the above example. After that, leave single user mode and you should be able to switch at will to other consoles to run ps or whatever you have in mind.
On Thu, Oct 27, 2011 at 01:06:59PM +1100, Brian May wrote:
Any ideas?
more info needed: text mode booting. boot logging. verbose boot. IMO, most of the settings below should be default, not disabled or hidden or undocumented. 1. edit /etc/default/grub and get rid of the "quiet". i routinely change the top of my /etc/default grub to: GRUB_DEFAULT=saved GRUB_SAVEDEFAULT=true GRUB_TIMEOUT=5 GRUB_DISTRIBUTOR=`lsb_release -i -s 2> /dev/null || echo Debian` #GRUB_CMDLINE_LINUX_DEFAULT="quiet" GRUB_CMDLINE_LINUX_DEFAULT="" GRUB_CMDLINE_LINUX="" # Uncomment to disable graphical terminal (grub-pc only) GRUB_TERMINAL=console 2. edit /etc/default/bootlogd to enable boot logging # Run bootlogd at startup ? BOOTLOGD_ENABLE=Yes 3. edit /etc/default/rcS and enable verbose booting. here's what mine looks like: # Time files in /tmp are kept in days. TMPTIME=0 # Set to yes if you want sulogin to be spawned on bootup SULOGIN=no # Set to no if you want to be able to login over telnet/rlogin # before system startup is complete (as soon as inetd is started) DELAYLOGIN=yes # Set UTC=yes if your system clock is set to UTC (GMT), and UTC=no if not. UTC=yes # Set VERBOSE to "no" if you would like a more quiet bootup. VERBOSE=yes # Set EDITMOTD to "no" if you don't want /etc/motd to be editted automatically EDITMOTD=yes # Set FSCKFIX to "yes" if you want to add "-y" to the fsck at startup. FSCKFIX=yes
Is there anyway of creating another virtual console within recovery mode?
open or openvt, depending on which version you have installed. probably openvt. craig -- craig sanders <cas@taz.net.au> BOFH excuse #42: spaghetti cable cause packet failure
participants (4)
-
Andrew Worsley -
Brian May -
Craig Sanders -
Jason White