Public bug reported:

Sorry about the long bug report, I'm trying to be as detailed as
possible.

After using it on other machines without problem, I've recently converted a 
server running Redhat 9 (old I know) to Hardy Server. The installation went 
fine and I was using it without problem for a few weeks, however the other day 
when I tried to SCP a large file (to it I started getting odd errors:
BUG: scheduling while atomic: swapper/0/0x0000010
repeated over and over and over on the console, and

 Pid: 0, comm: swapper Not tainted 2.6.24-18-server #1
  [jbd:schedule+0x582/0x6b0] schedule+0x582/0x630
  [ktime_get+0x18/0x40] ktime_get+0x18/0x40
  [rcu_pending+0x37/0x50] rcu_pending+0x37/0x50
  [cpu_idle+0x8b/0xd0] cpu_idle+0x8b/0xd0
  =======================

logged to the syslog each time, before eventually panicing.

I searched google for information about this and found a similar bug
(https://bugs.launchpad.net/ubuntu/+bug/235889) which suggested trying a
kernel from kernel.org or trying the intrepid kernel from the Kernel-PPA

Firstly I tried to use the PPA, but it appears to be failing to build

Then I tried to follow https://wiki.ubuntu.com/KernelGitGuide but whilst
in the middle of building the console started spaming errors so fast I
couldn't read them, and the machine became unresponsive any time it
tried to access the ahrd drive (squid stoped working, ssh stopped
working, but other machines on the network which use this machine as a
gateway were able to use the itnernet without problems)

I followed the "Master Kernel Thread"
(http://ubuntuforums.org/showthread.php?t=311158) guide to compile a
replacement kernel (2.6.25.5), and after 2 hours it seemed to compile
fine but when it got to actually packaging the kernel it paniced.

(for the "make oldconfig" step I just accepted all the defaults)

I restarted and tried to build the package again however make-kpgk
wanted to start again, so I let it and once again it paniced at the same
place.

So now I can reproduce the panic easily by just running: "cd
/usr/src/linux ; dpkg --build /usr/src/linux/debian/linux-
image-2.6.25.5-df .." as root

Unfortunatly the default frame bffer didn't allow for the full panic to
be seen so I followed
https://help.ubuntu.com/community/ConsoleFramebuffer and enabled  large
frame buffer and set vga=0x31A in my grub config, rebooted and tried to
build the package again, once again it paniced and I managed to get a
picture of the panic (attached) altho I've made it panic since and it
paniced for a different reason (it mentioned HIGHMEM, but I didn't have
a camera handy at the time).

I ran memtest to rule out a memory problem and it ran fine.

So as I have another hardy machine (my desktop) whilch I rebuilt the
kernel on earlier (non-server kernel, using CheckKernel from the master
kernel thread, so its kernel 2.6.25.6) I copied it across and rebooted
into it and ran the "dpkg --build" command again and this time is
started spamming

NOHZ: Local_softirq_pending 152
BUG: scheduling while atomic: swapper/0/0x00000100
NOHZ: Local_softirq_pending 1de

a few times in the console

followed by lots of just
BUG: scheduling while atomic: swapper/0/0x00000100

with this in the syslog:
Jun 11 12:41:44 neo kernel: [ 1610.917763] BUG: scheduling while atomic: 
swapper/0/0x00000100
Jun 11 12:41:44 neo kernel: [ 1610.917909] Pid: 0, comm: swapper Not tainted 
2.6.25.6 #1
Jun 11 12:41:44 neo kernel: [ 1610.917909]  [lp:schedule+0x59e/0x680] 
schedule+0x59e/0x630
Jun 11 12:41:44 neo kernel: [ 1610.917909]  [ktime_get_ts+0x1e/0x60] 
ktime_get_ts+0x1e/0x60
Jun 11 12:41:44 neo kernel: [ 1610.917909]  [default_idle+0x0/0xa0] 
default_idle+0x0/0xa0
Jun 11 12:41:44 neo last message repeated 2 times
Jun 11 12:41:44 neo kernel: [ 1610.917909]  
[tick_nohz_restart_sched_tick+0x2b/0x140] 
tick_nohz_restart_sched_tick+0x2b/0x140
Jun 11 12:41:44 neo kernel: [ 1610.917909]  
[touch_softlockup_watchdog+0x21/0x30] touch_softlockup_watchdog+0x21/0x30
Jun 11 12:41:44 neo kernel: [ 1610.917909]  [default_idle+0x0/0xa0] 
default_idle+0x0/0xa0
Jun 11 12:41:44 neo kernel: [ 1610.917909]  [cpu_idle+0x95/0xd0] 
cpu_idle+0x95/0xd0
Jun 11 12:41:44 neo kernel: [ 1610.917909]  =======================

before hanging intermittedly (stopping network traffic and any commands
over ssh) and then hanging completely (but without panicing)

I have copied the .config over from the failed builds on the server to
my desktop and built a 2.6.25.5 server kernel on here, and copied that
over to the server and this time it did panic (I Have a picture of this
aswell if needed)

This happens on the following tested kernels:
 * 2.6.24-16-server
 * 2.6.24-17-server
 * 2.6.24-18-server
 * 2.6.25.6 (using desktop .config)
 * 2.6.25.5 (using server .config)

as Per https://wiki.ubuntu.com/KernelTeamBugPolicies I am also attaching
the output of various things.

** Affects: linux (Ubuntu)
     Importance: Undecided
         Status: New

-- 
Kernel panic on dpkg --build
https://bugs.launchpad.net/bugs/239155
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to