Bug#615906: installation-reports: SunFire V120 Unable to handle kernel NULL pointer dereference
Package: installation-reports
Severity: normal
-- Package-specific info:
Boot method: network
Image version: squeeze official netboot.
Date: 20/Feb/2011
Machine: SunFire V120
Partitions: n/a
Base System Installation Checklist:
[O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it
Initial boot: [O]
Detect network card: [ ]
Configure network: [ ]
Detect CD: [ ]
Load installer modules: [ ]
Detect hard drives: [ ]
Partition hard drives: [ ]
Install base system: [ ]
Clock/timezone setup: [ ]
User/password setup: [ ]
Install tasks: [ ]
Install boot loader: [ ]
Overall install: [E]
Comments/Problems:
I get the following fatal ooops just before setting up the network in the
installer during a netboot install.
[ 225.950970] Unable to handle kernel NULL pointer dereference
[ 226.025376] tsk->{mm,active_mm}->context = 0000000000000588
[ 226.098688] tsk->{mm,active_mm}->pgd = fffff800ddb9c000
[ 226.167644] Kernel panic - not syncing: Aiee, killing interrupt handler!
[ 226.256067] Press Stop-A (L1-A) to return to the boot prom
I've actually reproduced it a few times. Unfortunately it is proving difficult
to get the full Oops stack trace out of it because klogd is altering the
/proc/sys/kernel/printk console log level and is stopping getting the
information to the console.
I've managed to change the logging level by booting using BOOT_DEBUG=3 and
applying the changes below but that seems to change the timing and the
problem doesn't occur. If you are able to rebuild the netboot image with
/sbin/klogd -c 7
in /lib/debian-installer-startup.d/S10syslog (normally it uses -c 2)
then that would mean that I could test it without having to use BOOT_DEBUG=3
Procedure for booting and manually tweaking S10syslog is as follows:
boot using
boot net BOOT_DEBUG=3
then use "exit" to leave the "just booted" debug shell
then in the "before init" debug shell run the following
cat > /lib/debian-installer-startup.d/S10syslog <<EOF
echo -n "Starting system log daemon: "
/sbin/syslogd -m 0 -O /var/log/syslog -S
echo -n "syslogd, "
/sbin/klogd -c 7
echo "klogd."
EOF
exit
I then reply with "English" as language and "United Kingdom". The crash
happens for me shortly after this point but before it starts mentioning
network cards.
On occasions where the kernel does not panic at this point I can perform
a full install successfully.
--
Please make sure that the hardware-summary log file, and any other
installation logs that you think would be useful are attached to this
report. Please compress large files using gzip.
Once you have filled out this report, mail it to submit@bugs.debian.org.
-- System Information:
Debian Release: 6.0
APT prefers stable
APT policy: (990, 'stable'), (1, 'experimental')
Architecture: sparc (sparc64)
Kernel: Linux 2.6.38-rc6-sparc64
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash
Reply to: