|Summary:||XFS Causes Kernel Panic on Reboot|
|Product:||[Fedora] Fedora||Reporter:||Gary A. McGee <clscience>|
|Component:||kernel||Assignee:||Dave Jones <davej>|
|Status:||CLOSED DUPLICATE||QA Contact:||Brian Brock <bbrock>|
|Fixed In Version:||Doc Type:||Bug Fix|
|Doc Text:||Story Points:||---|
|Last Closed:||2005-05-25 00:30:59 UTC||Type:||---|
|oVirt Team:||---||RHEL 7.3 requirements from Atomic Host:|
Description Gary A. McGee 2005-04-26 12:53:29 UTC
Description of problem: During reboot, the system will hang while shutting down XFS. When the system hangs, I observe the following message: Shutting down xfs: Kernel panic - not syncing: drivers/block/cfq-iosched.c:1065: spin_is_locked on uninitialized spinlock ed9d4ck. (Tainted: P ) Version-Release number of selected component (if applicable): kernel-2.6.11-1.14_FC3 kernel-utils-2.4-13.1.49_FC3 xorg-x11-xfs-6.8.2-1.FC3.13 How reproducible: Sometimes Steps to Reproduce: 1. Reboot/Shutdown 2. 3. Actual results: System Hangs on reboot/shutdown. Expected results: System should shutdown cleanly. Additional info: The log files did not seem to reveal anything, except that shutdown got as far as XFS. I am still trying to determine whether this problem is associated with the use of a usbdisk. Seems like when I have used my usbdisk (1 GB memory stick) this occurs, but I must confirm it.
Comment 1 Gary A. McGee 2005-04-26 14:13:28 UTC
I can reproduce the problem reliably now, but I am not sure what this has to do with shutting down XFS. The problem is associated with the use of a usbdisk removable memory. Here is what happens. When I insert the device into the usb port, it is mounted as one would expect. I then right click on the desktop icon to unmount the device. The icon goes away, and "df -h" shows that the device has been unmounted. After the device is umounted, I notice that the device light is still on, but not flashing. Assuming the device is safe to remove, I take it out. I log out and then initiate a reboot. When the shutdown process gets to XFS, it hangs as described in the initial report. I have observed that there is a work-around for this. If I follow the same procedure, as in the preceding paragraph, but do not take the device out of the usb port during the reboot, it does not hang. It seems to me that when the device is unmounted that it should be safe to remove without impacting the system.
Comment 2 Gary A. McGee 2005-04-26 14:20:28 UTC
Created attachment 113668 [details] /var/log/Xorg.0.log Here is the Xorg.log.
Comment 3 Rik van Riel 2005-04-26 16:21:45 UTC
Gary, which 3rd party kernel module are you using ?
Comment 4 Mike A. Harris 2005-04-26 16:44:14 UTC
A kernel panic is not an X font server bug, nor an Xorg bug at all. A kernel panic is due to a bug in the kernel or a hardware fault. Your kernel is tainted by 3rd party kernel modules, so it is not a Red Hat kernel anymore. If you can reproduce the problem with a stock kernel with no 3rd party modules loaded whatsoever, reopen the bug.
Comment 5 Gary A. McGee 2005-04-26 17:17:18 UTC
I am using the Linuxant DriverLoader for my internal Broadcom Wireless card.
Comment 6 Gary A. McGee 2005-04-26 18:03:36 UTC
I removed the Linuxant driverloader-2.27_k2.6.11_1.14_FC3-1fdr.i686.rpm with the following command: $ rpm -e driverloader Removed driverloader from /lib/modules/
Comment 7 Gary A. McGee 2005-04-26 18:08:24 UTC
After reboot "lsmod" shows the following: $ lsmod Module Size Used by radeon 74177 1 parport_pc 28421 0 lp 12489 0 parport 40201 2 parport_pc,lp autofs4 26181 0 i2c_dev 11201 0 i2c_core 21953 1 i2c_dev sunrpc 164485 1 pcmcia 26465 2 ipt_REJECT 7105 1 ipt_state 1857 1 ip_conntrack 40601 1 ipt_state iptable_filter 2881 1 ip_tables 19777 3 ipt_REJECT,ipt_state,iptable_filter md5 4161 1 ipv6 259201 12 dm_mod 59221 0 video 15813 0 button 6609 0 battery 9285 0 ac 4805 0 ohci1394 39129 0 ieee1394 309145 1 ohci1394 yenta_socket 21065 1 rsrc_nonstatic 10433 1 yenta_socket pcmcia_core 47993 3 pcmcia,yenta_socket,rsrc_nonstatic uhci_hcd 33497 0 ehci_hcd 39501 0 snd_ali5451 28421 2 snd_ac97_codec 71097 1 snd_ali5451 snd_pcm_oss 51953 0 snd_mixer_oss 18241 2 snd_pcm_oss snd_pcm 99657 3 snd_ali5451,snd_ac97_codec,snd_pcm_oss snd_timer 33093 1 snd_pcm snd 56741 8 snd_ali5451,snd_ac97_codec,snd_pcm_oss,snd_mixer_oss,snd_pcm,snd_timer soundcore 10785 2 snd snd_page_alloc 9669 1 snd_pcm natsemi 33697 0 floppy 64753 0 ext3 131145 9 jbd 82777 1 ext3
Comment 8 Rik van Riel 2005-04-26 18:11:44 UTC
Can the bug be reproduced without the Linuxant drivers present ? If yes, please reopen this bug and reassign it to the "kernel" component. If the bug is dependant on the Linuxant driver, please open a bug report with the Linuxant developers. They will be the only ones who can help you at that point - after all they have our source, we don't have theirs.
Comment 9 Gary A. McGee 2005-04-26 18:17:27 UTC
I followed the same procedure as outlined in comment #1, paragraph #2 and received the following results right after XFS was successfully shutdown. Shutting down console mouse services: Kernel panic - not syncing: drivers/block/cfg-iosched.c:1065: spin_is_locked on uninitialized spinlock ee86341c. (Not tainted) It is obvious to me now that this is not an XFS issue, and that the issue exists, for some reason, even though I am using a stock kernel. Other than DriverLoader, which I uninstalled, this is a stock kernel. I did the FC3 installation yesterday, followed by "yum update".
Comment 10 Gary A. McGee 2005-04-26 18:18:20 UTC
Please advise what I should do next.
Comment 11 Rik van Riel 2005-04-26 18:22:43 UTC
Did the kernel print out anything else than that panic ? Any backtraces, etc... ?
Comment 12 Gary A. McGee 2005-04-26 18:46:06 UTC
No, that was the last message on the console when it froze up. Is there another place I should look in order to glean information? The logs did not seem to reveal anything.
Comment 13 Dave Jones 2005-04-26 22:08:33 UTC
http://people.redhat.com/davej/kernels/Fedora/FC3 has a work-in-progress kernel that should fix this.
Comment 14 Gary A. McGee 2005-05-23 23:22:13 UTC
I upgraded to kernel-2.6.11-1.27_FC3 today, and was not able to reproduce the problem. Thank you.