kernel BUG at lib/kernel_lock.c:83! - 2.6.19-1.2895.fc6
by agraham
Hello,
Kernel 2.6.19.x seems much more stable in my setup than the previous
versions 2.6.18/17.x which used to lock-up 5 times a day.
I did get the following crash last night (forgot exactly what I was
doing - killed rsync server), I just saved the error and rebooted.
I found a similar error here:
http://lkml.org/lkml/2005/8/19/70
but this was posted more than a year ago ?
Albert.
BUGging on (task->lock_depth < 0)
------------[ cut here ]------------
kernel BUG at lib/kernel_lock.c:83!
invalid opcode: 0000 [#1]
SMP
last sysfs file: /block/loop3/range
Modules linked in: iptable_nat xt_state ip6_tables iptable_filter
ip_tables x_tables ip_nat_tftp ip_nat_snmp_basic
ip_nat_sip ip_nat_pptp ip_nat_irc ip_nat_h323 ip_nat_ftp ip_nat_amanda
ip_conntrack_tftp ip_conntrack_sip
ip_conntrack_proto_sctp ip_conntrack_pptp ip_conntrack_netlink ip_nat
ip_conntrack_netbios_ns ip_conntrack_irc
ip_conntrack_h323 ip_conntrack_ftp ts_kmp ip_conntrack_amanda
ip_conntrack nfnetlink nls_utf8 loop autofs4 hidp l2cap
bluetooth dm_multipath sha256 aes cbc blkcipher dm_crypt video sbs
i2c_ec button battery asus_acpi ac lp snd_emu10k1_synth
snd_emux_synth snd_seq_virmidi snd_seq_midi_emul snd_emu10k1 snd_rawmidi
snd_ac97_codec snd_ac97_bus snd_seq_dummy tuner
snd_seq_oss snd_seq_midi_event snd_seq snd_pcm_oss snd_mixer_oss saa7134
video_buf nvidia(P)(U) floppy snd_pcm
compat_ioctl32 ir_kbd_i2c ir_common snd_seq_device snd_timer
snd_page_alloc i82875p_edac iTCO_wdt snd_util_mem videodev
edac_mc snd_hwdep 8139cp snd sg v4l1_compat ohci1394 i2c_i801
v4l2_common e1000 8139too ieee1394 i2c_core soundcore pcspkr
mii parport_pc parport serio_raw dm_snapshot dm_zero dm_mirror dm_mod
ata_piix sata_promise libata aic7xxx
scsi_transport_spi sd_mod scsi_mod ext3 jbd ehci_hcd ohci_hcd uhci_hcd
CPU: 3
EIP: 0060:[<c0625bb7>] Tainted: P VLI
EFLAGS: 00210282 (2.6.19-1.2895.fc6 #1)
EIP is at unlock_kernel+0x27/0x52
eax: 00000025 ebx: f6cee170 ecx: c0697ed0 edx: 00200086
esi: f779b680 edi: bfa86228 ebp: 0000541b esp: f591ef60
ds: 007b es: 007b ss: 0068
Process kicker (pid: 4896, ti=f591e000 task=f6cee170 task.ti=f591e000)
Stack: c063e591 c0657e5c 00000000 c048025f bfa86228 00000000 f779b680
bfa86228
00000006 c04804b8 00000000 f725d180 00000000 f779b680 bfa86228
f591e000
c0480516 bfa86228 00000000 00000006 08b231c0 08dd86b8 c040404b
00000006
Call Trace:
[<c048025f>] do_ioctl+0x53/0x62
[<c04804b8>] vfs_ioctl+0x24a/0x25c
[<c0480516>] sys_ioctl+0x4c/0x66
[<c040404b>] syscall_call+0x7/0xb
[<00542402>] 0x542402
=======================
Code: 89 d8 5b c3 53 83 ec 08 89 e0 25 00 f0 ff ff 8b 18 83 7b 10 00 79
1c c7 44 24 04 5c 7e 65 c0 c7 04 24 91 e5 63 c0 e8
1a 1f e0 ff <0f> 0b 53 00 71 7e 65 c0 8b 43 10 48 85 c0 89 43 10 79 14 f0 ff
EIP: [<c0625bb7>] unlock_kernel+0x27/0x52 SS:ESP 0068:f591ef60
BUGging on (task->lock_depth < 0)
------------[ cut here ]------------
kernel BUG at lib/kernel_lock.c:83!
invalid opcode: 0000 [#2]
SMP
last sysfs file: /block/loop3/range
Modules linked in: iptable_nat xt_state ip6_tables iptable_filter
ip_tables x_tables ip_nat_tftp ip_nat_snmp_basic
ip_nat_sip ip_nat_pptp ip_nat_irc ip_nat_h323 ip_nat_ftp ip_nat_amanda
ip_conntrack_tftp ip_conntrack_sip
ip_conntrack_proto_sctp ip_conntrack_pptp ip_conntrack_netlink ip_nat
ip_conntrack_netbios_ns ip_conntrack_irc
ip_conntrack_h323 ip_conntrack_ftp ts_kmp ip_conntrack_amanda
ip_conntrack nfnetlink nls_utf8 loop autofs4 hidp l2cap
bluetooth dm_multipath sha256 aes cbc blkcipher dm_crypt video sbs
i2c_ec button battery asus_acpi ac lp snd_emu10k1_synth
snd_emux_synth snd_seq_virmidi snd_seq_midi_emul snd_emu10k1 snd_rawmidi
snd_ac97_codec snd_ac97_bus snd_seq_dummy tuner
snd_seq_oss snd_seq_midi_event snd_seq snd_pcm_oss snd_mixer_oss saa7134
video_buf nvidia(P)(U) floppy snd_pcm
compat_ioctl32 ir_kbd_i2c ir_common snd_seq_device snd_timer
snd_page_alloc i82875p_edac iTCO_wdt snd_util_mem videodev
edac_mc snd_hwdep 8139cp snd sg v4l1_compat ohci1394 i2c_i801
v4l2_common e1000 8139too ieee1394 i2c_core soundcore pcspkr
mii parport_pc parport serio_raw dm_snapshot dm_zero dm_mirror dm_mod
ata_piix sata_promise libata aic7xxx
scsi_transport_spi sd_mod scsi_mod ext3 jbd ehci_hcd ohci_hcd uhci_hcd
CPU: 3
EIP: 0060:[<c0625bb7>] Tainted: P VLI
EFLAGS: 00210296 (2.6.19-1.2895.fc6 #1)
EIP is at unlock_kernel+0x27/0x52
eax: 00000022 ebx: f6cee170 ecx: ffffffff edx: 00200046
esi: f73d9208 edi: f73d9208 ebp: f7120d80 esp: f591ed58
ds: 007b es: 007b ss: 0068
Process kicker (pid: 4896, ti=f591e000 task=f6cee170 task.ti=f591e000)
Stack: c063e591 c0657e5c f7e44ec0 c04a5940 f73d9208 c04a58e7 c0486df4
f73d9208
c04ee452 f73d9208 c069ff20 c0486407 f73d9208 f511fe98 f511fe98
f511fea0
c04851cc 00000010 f73d9208 c04767fa 00000000 00000000 f511fe98
c2b76740
Call Trace:
[<c04a5940>] proc_delete_inode+0x59/0x65
[<c0486df4>] generic_delete_inode+0xc3/0x145
[<c0486407>] iput+0x7c/0x81
[<c04851cc>] dput+0xe6/0xfe
[<c04767fa>] __fput+0x166/0x190
[<c04740f6>] filp_close+0x52/0x59
[<c0428fdc>] put_files_struct+0x65/0xa7
[<c042a1bf>] do_exit+0x246/0x787
[<c0405719>] die+0x2c3/0x2e8
[<c0405c5a>] do_invalid_op+0xa2/0xab
[<c0625ce1>] error_code+0x39/0x40
[<c0625bb7>] unlock_kernel+0x27/0x52
[<c048025f>] do_ioctl+0x53/0x62
[<c04804b8>] vfs_ioctl+0x24a/0x25c
[<c0480516>] sys_ioctl+0x4c/0x66
[<c040404b>] syscall_call+0x7/0xb
[<00542402>] 0x542402
=======================
Code: 89 d8 5b c3 53 83 ec 08 89 e0 25 00 f0 ff ff 8b 18 83 7b 10 00 79
1c c7 44 24 04 5c 7e 65 c0 c7 04 24 91 e5 63 c0 e8
1a 1f e0 ff <0f> 0b 53 00 71 7e 65 c0 8b 43 10 48 85 c0 89 43 10 79 14 f0 ff
EIP: [<c0625bb7>] unlock_kernel+0x27/0x52 SS:ESP 0068:f591ed58
<1>Fixing recursive fault but reboot is needed!
17 years, 3 months
Installing Fedora Core 6 from USB
by Rakhesh Sasidharan
Hi there!
Has anybody tried installing Fedora Core 6 from USB? I understand one
has to download the diskboot.img file, dd it to the USB disk (dd
if=diskboot.img of=/dev/sda), and that's it. I did these steps, but am
unable to boot from the USB! My machine just bypasses the USB stick as
if there was nothing bootable on it.
I know my USB stick is fine. And I know the BIOS is checking it for
bootsectors too. Coz, the first time I tried, I dd'ed the bootsectors
to /dev/sda1 (the first partition) and so that time the BIOS gave me
an error saying it was unable to boot the USB stick.
Any suggestions on what could be the problem?
I'm trying to boot from USB coz I can't boot from the CD. I get some
isolinux error messages (I've posted in another thread) and I haven't
figured yet why.
Thanks in advance.
Regards,
Rakhesh
17 years, 3 months
Re: Certain screensavers lock my machine up
by David G. Miller (aka DaveAtFraud)
Ed Gurski <ed(a)gurski.com> wrote:
> I've run into the same problem and with a little bit of digging I found
> the Kaleidoscope was the screen save locking up my machine. I have
> unselected it and have now been running for 24 hours without a
> lock-up...
>
> HTH...
>
>
I just tried the Kaleidoscope preview for about a minute before replying
to your post and it worked just fine. Could be there are a swarm of
separate bugs that are dependent on things like the specific video card,
nv driver vs. Nvidia binary driver, amount of installed memory, speed of
the box, phase of the moon, etc. I'll try running just Kaleidoscope for
a while and see if I get a lock up.
Dave
--
Politics, n. Strife of interests masquerading as a contest of principles.
-- Ambrose Bierce
17 years, 3 months
RPM Updates
by Ashley M. Kirchner
Would someone please tell me how's this done?
Vanilla FC6 installs
elfutils-libelf-0.123-1.fc6.i386.rpm
elfutils-libelf-devel-0.123-1.fc6.i386.rp
However, running rpm updates, it wants:
elfutils-libelf-0.125-1.fc6.i386.rpm
elfutils-libelf-devel-0.125-1.fc6.i386.rpm
elfutils-libelf-devel-static-0.125-1.fc6.i386.rpm
...notice the -static one in there. Apparently -devel wants the
-static one as well. Which is fine, however I can't just install
-static with -ivh because it wants the -devel at the same time. And I
can't update the -devel one through -Fvh because -static isn't installed.
Someone explain to me why this "chicken and egg" scenario exists in
various packages (I'm takling of previous versinos of FC here)? It has
been like that for years with some package or another. And chances are
a bunch of you are going to sing it in chorus to me: use yum. Well
fine, however that still does not solve the underlying problem of why
this is happening in the first place nor how to successfully solve it
with rpm, instead of switching to yum.
--
W | It's not a bug - it's an undocumented feature.
+--------------------------------------------------------------------
Ashley M. Kirchner <mailto:ashley@pcraft.com> . 303.442.6410 x130
IT Director / SysAdmin / Websmith . 800.441.3873 x130
Photo Craft Imaging . 3550 Arapahoe Ave. #6
http://www.pcraft.com ..... . . . Boulder, CO 80303, U.S.A.
17 years, 3 months
kernel-2.6.19-1.2895.fc6 dies. do_IRQ
by Naoki
This locks my IBM x3550 :
"do_IRQ: 0.177 No irq handler for vector"
I need a power cycle after that.
Messages file reveals.. Not much really :
Jan 30 20:32:10 localhost kernel: Bluetooth: HIDP (Human Interface
Emulation) ver 1.1
Jan 30 20:32:10 localhost kernel: SELinux: initialized (dev autofs, type
autofs), uses genfs_contexts
Jan 30 20:32:10 localhost last message repeated 2 times
Jan 30 20:32:10 localhost kernel: do_IRQ: 0.177 No irq handler for
vector
Looking around I find Mr Andrew Morton on the case, which is a relief.
http://lkml.org/lkml/2007/1/22/121
It's being seen on 16-core systems, I'm running only a four core box
though.
Doesn't look like a patch is out yet so my original plan of trying
Rawhide probably will not help.
I have not (yet?) had the problem on 2.6.18-1.2798.fc6. It only seems
to be in the 2.6.19 kernel.
17 years, 3 months
Howto open xterm to another X virtual monitor
by Dario Lesca
Hi! ... from a gnome-terminal if I run "xterm &" the xterm is open on
current desktop; it's possible open it on another desktop?
Many thanks
--
Dario Lesca <d.lesca(a)solinos.it>
17 years, 3 months
Need assistance compiling package
by awrobinson-ml@nc.rr.com
I'm trying to build palm-db-tools. The build used to work on FC4, but
fails as you see on FC6. Is there a flag or something that would get me
around this particular problem?
Thanks!
Andrew Robinson
[root@proteus palm-db-tools]# make
make -C libsupport; make -C libpalm; make -C libflatfile; make -C
flatfile;
make[1]: Entering directory
`/data/Downloads/Pilot-DB/palm-db-tools/libsupport'
g++ -ansi -pedantic -g -O2 -Wall -Werror -Wno-deprecated -I. -I..
-DHAVE_CONFIG_H -c csvfile.cpp
cc1plus: warnings being treated as errors
../libsupport/infofile.h:33: warning: ‘class DataFile::InfoFile::Parser’
has virtual functions but non-virtual destructor
../libsupport/infofile.h:38: warning: ‘class
DataFile::InfoFile::ConfigParser’ has virtual functions but non-virtual
destructor
../libsupport/infofile.h:48: warning: ‘class
DataFile::InfoFile::DatabaseParser’ has virtual functions but
non-virtual destructor
../libsupport/infofile.h:58: warning: ‘class
DataFile::InfoFile::TypeParser’ has virtual functions but non-virtual
destructor
../libsupport/infofile.h:68: warning: ‘class
DataFile::InfoFile::PDBPathParser’ has virtual functions but non-virtual
destructor
make[1]: *** [csvfile.o] Error 1
make[1]: Leaving directory
`/data/Downloads/Pilot-DB/palm-db-tools/libsupport'
make[1]: Entering directory `/data/Downloads/Pilot-DB/palm-db-tools/libpalm'
g++ -ansi -pedantic -g -O2 -Wall -Werror -Wno-deprecated -I. -I..
-DHAVE_CONFIG_H -c File.cpp
cc1plus: warnings being treated as errors
../libsupport/infofile.h:33: warning: ‘class DataFile::InfoFile::Parser’
has virtual functions but non-virtual destructor
../libsupport/infofile.h:38: warning: ‘class
DataFile::InfoFile::ConfigParser’ has virtual functions but non-virtual
destructor
../libsupport/infofile.h:48: warning: ‘class
DataFile::InfoFile::DatabaseParser’ has virtual functions but
non-virtual destructor
../libsupport/infofile.h:58: warning: ‘class
DataFile::InfoFile::TypeParser’ has virtual functions but non-virtual
destructor
../libsupport/infofile.h:68: warning: ‘class
DataFile::InfoFile::PDBPathParser’ has virtual functions but non-virtual
destructor
make[1]: *** [File.o] Error 1
make[1]: Leaving directory `/data/Downloads/Pilot-DB/palm-db-tools/libpalm'
make[1]: Entering directory
`/data/Downloads/Pilot-DB/palm-db-tools/libflatfile'
g++ -ansi -pedantic -g -O2 -Wall -Werror -Wno-deprecated -I. -I..
-DHAVE_CONFIG_H -c Factory.cpp
cc1plus: warnings being treated as errors
../libsupport/infofile.h:33: warning: ‘class DataFile::InfoFile::Parser’
has virtual functions but non-virtual destructor
../libsupport/infofile.h:38: warning: ‘class
DataFile::InfoFile::ConfigParser’ has virtual functions but non-virtual
destructor
../libsupport/infofile.h:48: warning: ‘class
DataFile::InfoFile::DatabaseParser’ has virtual functions but
non-virtual destructor
../libsupport/infofile.h:58: warning: ‘class
DataFile::InfoFile::TypeParser’ has virtual functions but non-virtual
destructor
../libsupport/infofile.h:68: warning: ‘class
DataFile::InfoFile::PDBPathParser’ has virtual functions but non-virtual
destructor
make[1]: *** [Factory.o] Error 1
make[1]: Leaving directory
`/data/Downloads/Pilot-DB/palm-db-tools/libflatfile'
make[1]: Entering directory
`/data/Downloads/Pilot-DB/palm-db-tools/flatfile'
g++ -ansi -pedantic -g -O2 -Wall -Werror -Wno-deprecated -I. -I..
-DHAVE_CONFIG_H -c pdbtools.cpp
cc1plus: warnings being treated as errors
../libsupport/infofile.h:33: warning: ‘class DataFile::InfoFile::Parser’
has virtual functions but non-virtual destructor
../libsupport/infofile.h:38: warning: ‘class
DataFile::InfoFile::ConfigParser’ has virtual functions but non-virtual
destructor
../libsupport/infofile.h:48: warning: ‘class
DataFile::InfoFile::DatabaseParser’ has virtual functions but
non-virtual destructor
../libsupport/infofile.h:58: warning: ‘class
DataFile::InfoFile::TypeParser’ has virtual functions but non-virtual
destructor
../libsupport/infofile.h:68: warning: ‘class
DataFile::InfoFile::PDBPathParser’ has virtual functions but non-virtual
destructor
make[1]: *** [pdbtools.o] Error 1
make[1]: Leaving directory `/data/Downloads/Pilot-DB/palm-db-tools/flatfile'
make: *** [all] Error 2
17 years, 3 months
Xorg is killing the box
by Evan Panagiotopoulos
I checked with the command top and the command Xorg has 98% of the cpu.
When I use the command prompt it takes a long time to refresh the screen.
Any ideas?
Evan
17 years, 3 months
Re: Bash globbing files only?
by David G. Miller (aka DaveAtFraud)
"David G. Miller" <dave(a)davenjudy.org> wrote:
> Yes, I am painfully aware of the power of zsh :) But in this case I
> > need to do my script with bash and it seems that although globbing for
> > directories is trivial ( */ ) there is no similar pattern to match
> > files. Oh, well, I guess I'll need to waste some CPU cycles and launch
> > find everytime I need to know file names in the current directory.
> >
> > Thanks.
>
> How about:
>
> ls -dF ./.[^.]*/ ./*/
> ./.gconf// ./.gnome2// ./.kde// ./.vnc//
> ./.gconfd// ./.gnome2_private// ./mail//
> ...
>
Nothing like getting the question backwards. Doh! Still seems like it
should be doable.
Cheers,
Dave
--
Politics, n. Strife of interests masquerading as a contest of principles.
-- Ambrose Bierce
17 years, 3 months