Bug 155521

Created attachment 113453 [details]
normal log file

This is what my log file looks like following a normal successful reboot.

Created attachment 113454 [details]
2.6.11-1.1251_FC4 lsmod output

output from lsmod command.

I have seen it too with FC devel.

Right before the line

ERROR: /sbin/udevstart exited abnormally!

There are two lines

mknod: Failed to create /dev/null
mknod: Failed to create /dev/console

Not sure if they are related.  Those error messages are not found in
any log file though, only appear on screen.

hmm, sounds like the initrd mknod has problems

problem occurs with 1253 build

is selinux turned on?
is /dev mounted with a tmpfs? (cat /proc/mounts)

selinux is on

[doug@lambda ~]$ cat /proc/mounts
rootfs / rootfs rw 0 0
none /dev tmpfs rw 0 0
/dev/root / ext3 rw 0 0
none /selinux selinuxfs rw 0 0
/proc /proc proc rw,nodiratime 0 0
/proc/bus/usb /proc/bus/usb usbfs rw 0 0
/sys /sys sysfs rw 0 0
none /dev/pts devpts rw 0 0
/dev/hda2 /boot ext3 rw 0 0
none /dev/shm tmpfs rw 0 0
none /proc/sys/fs/binfmt_misc binfmt_misc rw 0 0
sunrpc /var/lib/nfs/rpc_pipefs rpc_pipefs rw 0 0

I upgraded this system from a fully updated FC3 with the FC4test2 CDs.

Updated hardware=all since I am seeing this on an Opteron (x86_64).

I did a full update from rawhide today and I too am still seeing the above
errors.  Selinux is on. Do you need any further info?

After installing all updates, including 
kernel 2.6.11-1.1258_FC4 
udev-057-2

the problem is still present. Messages are

ERROR: /sbin/udevstart exited abnormally! (pid 407)
ERROR: /sbin/udevstart exited abnormally! (pid 426)

also there are two other messages which appear on the console but are not logged

unmounting old /proc
unmounting old /sys

I'm seeing this with selinux=0

i get this too with selinux disabled.

This is still happening with rawhide from today. I get the same errors as above
and mknod says it has problems creating /dev/null, /dev/console and /dev/zero

I've had such problems, but they appear to be somewhat random.  After I updated
to today's rawhide, the kernel wouldn't boot: after a number of failed mknods,
switchroot failed.  Another reboot still had some failed mknods, but it booted.

Try the mkinitrd and udev from rawhide as of 4/27 .  Hopefully fixed.

If it's not, please change this back to "ASSIGNED".

I don't get the udevstart errors any longer after today's yum update, but I
still see it failing to create /dev/zero and /dev/null

The error messages have disappeared with kernel 2.6.11-1.1268_FC4.

It no longer exits with a non-zero status, but it still randomly displays
messages about failure to mknod devices such as /dev/zero.  It's not reliably
reproducible, and the actual set of mknod errors varies.  This is with a fresh
install of FC4-re0503.0.

Alexandre, can you tell me what you see before and after those lines?  I'm
wondering if this morning's udev would fix it, but I'm not seeing this problem
on any of my machines.

This is what I got on FC4-re0505.0 (x86_64):

audit (...): initialized
Red Hat nash v. ...  starting
mknod: ... /dev/zero: 17
[messages about sda and sdb that I get when sbp2 and usb-storage are loaded]
Reading all physical volumes...

The odd thing is that the messages are random; sometimes there are none,
sometimes there are messages for all 3 mknod commands in the init script in
initrd, in the right order.  This points to nash's mknod or the kernel as the
prime suspects, I think.

then please open a new bugzilla for the problem

Done.  For reference, it's now Bug 158427.