Waiting for device... to appear: !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!not found -- exiting to /bin/sh

SUSE Linux Enterprise Server SLES Install-Boot
1

Hi,

after applying the latest kernel update (might be coincidence) one of our
SLES 11 SP3 server did not boot any more. It stops because the root device
is not found and exits into a shell. If I leave the shell with ‘exit’, the
device gets mounted and the server continues to boot without any problems:

Boot logging started on /dev/tty1(/dev/console) at Fri Apr 25 13:13:37 2014

FATAL: Error inserting mgag200 (/lib/modules/3.0.101-0.21-
default/kernel/drivers/gpu/drm/mgag200/mgag200.ko): Invalid argument
Setup multipath devices: ok.
Trying manual resume from /dev/disk/by-
id/scsi-360050760580b25501a2b5baa1bc1de9a-part1
Invoking userspace resume from /dev/disk/by-
id/scsi-360050760580b25501a2b5baa1bc1de9a-part1
resume: libgcrypt version: 1.5.0
Trying manual resume from /dev/disk/by-
id/scsi-360050760580b25501a2b5baa1bc1de9a-part1
Invoking in-kernel resume from /dev/disk/by-
id/scsi-360050760580b25501a2b5baa1bc1de9a-part1
Waiting for device /dev/disk/by-id/scsi-360050760580b25501a2b5baa1bc1de9a-
part2 to appear: !!!Could not find /dev/disk/by-
id/scsi-360050760580b25501a2b5baa1bc1de9a-part2.
Want me to fall back to /dev/disk/by-
id/scsi-360050760580b25501a2b5baa1bc1de9a-part2? (Y/n)
Waiting for device /dev/disk/by-id/scsi-360050760580b25501a2b5baa1bc1de9a-
part2 to appear: !!!not found – exiting to
/bin/sh
$ exit
exit
fsck from util-linux 2.19.1
[/sbin/fsck.ext3 (1) – /] fsck.ext3 -a -C0 /dev/sda2
/dev/sda2: clean, 269222/14024704 files, 6787395/56097024 blocks
fsck succeeded. Mounting root device read-write.
Mounting root /dev/disk/by-id/scsi-360050760580b25501a2b5baa1bc1de9a-part2
mount -o rw,acl,user_xattr -t ext3 /dev/disk/by-
id/scsi-360050760580b25501a2b5baa1b0cde9a-part2 /root

I tried to downgrade the kernel, but the problem stays the same.

Boot and root partitions are on local RAID 1 disks.

Any ideas?

Thanks,
Mirko

2

Hi Mirko,

strange co-incidence: I’m booting of a (FC-SAN) disk that, according to the output of initrd, comes only only after a few seconds - the LVM is already started (or rather - not) at that moment and the failure is similar to yours. In my case, I need to (re-)activate the VG before exiting, but that’s because this server uses the device as a physical volume and yours doesn’t. This is a fresh install of SLES11SP3, including latest updates, on a clean machine.

I haven’t had time to open a service request yet - if you do, please drop me a personal message so I can reference it with my SR, it’ll probably speed up things if there’s more than one customer affected.

Regards,
Jens