Highlighted
Outstanding Contributor.
Outstanding Contributor.
243 views

OES 2015 -> 2018 grub boot failure

Hi all,

Upgraded one of our customers from OES2015 to OES2018 SP1 using the disk upgrade method and after few hiccups managed to get everything up and running fine with one exception, booting the server from disk!

If we just let the server boot from the drive we drop out at the grub rescue prompt, the exact error is:

--
error disk 'hd0,msdos1' not found
entering rescue mode
--

I'm guessing the migration from grub to grub2 went awry and doesn't know what disk naming is in place. We can get the server to start correctly by using the Boot/Upgrade DVD and then choosing "Boot from Disk" I think it was on the second page of the start up menu on the DVD.

The server is currently up and running fine (until it needs to be restarted and then the manual menu process needs to be run), all I need to find is how to reconfigure grub2 with the configuration that the DVD is happily using to start the server.

TIA,
Mark.

Labels (2)
Tags (3)
0 Likes
6 Replies
Highlighted
Knowledge Partner
Knowledge Partner

Re: OES 2015 -> 2018 grub boot failure

As your server is running, I would first try to reinstall GRUB2 from Yast2 > System > Boot Loader under the running server. In your case, it seems as if just the MBR is missing and needs reinstall (otherwise the boot from the DVD wouldn't work either).

CU,
--
Massimo Rosen
Micro Focus Knowledge Partner
No emails please!
http://www.cfc-it.de
0 Likes
Highlighted
Knowledge Partner
Knowledge Partner

Re: OES 2015 -> 2018 grub boot failure

And doublecheck the content of

/boot/grub2/device.map

 

0 Likes
Highlighted
Outstanding Contributor.
Outstanding Contributor.

Re: OES 2015 -> 2018 grub boot failure

Definitely something not right in the boot loader setup. In the bootcode code options screen of yast none of the Boot Loader Locations are checked, nor any of the other options below. Compared this to a different OES2018 server and that has checked: Boot from Root Partition, Set active Flag in Partition Table for Boot Partition, and Write generic Boot Code to MBR.

The only other difference I can see is the fully working servers, device.map file uses old style /dev/sda for the drive mapping, the non-self-booting server uses the newer /dev/disk/by-id/... method and has multiple disks listed for hd0, hd1 and hd2. The hd0 id does link back to /dev/sda though.

If I look at /etc/fstab I see this:
UUID=e6d8137a-8f8d-4ebc-9abc-e210ad3cf4d8 swap swap defaults 0 0
UUID=d41270ad-70ed-4e29-a377-a80fd30b9875 / ext3 acl,user_xattr 1 1
UUID=e6d2234d-6d34-4a71-9c1f-09c8e52cc201 /boot ext3 acl,user_xattr 1 2
SHARED /media/nss/shared nssvol noauto,rw,name=SHARED,ns=Unix 0 0
GRPWISE /media/nss/grpwise nssvol noauto,rw,name=GRPWISE,noatime,ns=Unix 0 0

UUIDs seem fine:
ls -ltr /dev/disk/by-uuid/
total 0
lrwxrwxrwx 1 root root 10 Feb 10 17:16 e6d2234d-6d34-4a71-9c1f-09c8e52cc201 -> ../../sda1
lrwxrwxrwx 1 root root 10 Feb 10 17:16 d41270ad-70ed-4e29-a377-a80fd30b9875 -> ../../sda3
lrwxrwxrwx 1 root root 10 Feb 10 17:16 e6d8137a-8f8d-4ebc-9abc-e210ad3cf4d8 -> ../../sda2
lrwxrwxrwx 1 root root 10 Feb 10 17:16 fa3849fe-2c06-e201-8000-3b511be8be3b -> ../../dm-0

Here's the filesystem list:
Filesystem Size Used Avail Use% Mounted on
devtmpfs 3.9G 8.0K 3.9G 1% /dev
tmpfs 3.9G 120K 3.9G 1% /dev/shm
tmpfs 3.9G 9.6M 3.9G 1% /run
tmpfs 3.9G 0 3.9G 0% /sys/fs/cgroup
/dev/sda3 259G 20G 227G 8% /
/dev/sda1 65M 53M 8.3M 87% /boot
tmpfs 791M 0 791M 0% /run/user/111
admin 4.0M 0 4.0M 0% /_admin
/dev/pool/DATA 201G 123G 78G 62% /opt/novell/nss/mnt/.pools/DATA
SHARED 100G 54G 47G 54% /media/nss/shared
GRPWISE 201G 69G 78G 47% /media/nss/grpwise
tmpfs 791M 16K 791M 1% /run/user/0

and finally the device.map file:
(hd2) /dev/disk/by-id/usb-iDRAC_OEMDRV_20111109-0:0
(hd1) /dev/disk/by-id/scsi-3690b11c003259c0017ef3d0937be010f
(hd0) /dev/disk/by-id/scsi-3690b11c003259c0017ef3ad81643c970

Mark

0 Likes
Highlighted
Knowledge Partner
Knowledge Partner

Re: OES 2015 -> 2018 grub boot failure


@MarkDissington wrote:

Definitely something not right in the boot loader setup. In the bootcode code options screen of yast none of the Boot Loader Locations are checked, nor any of the other options below. Compared this to a different OES2018 server and that has checked: Boot from Root Partition, Set active Flag in Partition Table for Boot Partition, and Write generic Boot Code to MBR.

Well, so go ahead and check them. 😉

CU,
--
Massimo Rosen
Micro Focus Knowledge Partner
No emails please!
http://www.cfc-it.de
0 Likes
Highlighted
Knowledge Partner
Knowledge Partner

Re: OES 2015 -> 2018 grub boot failure

...and take note of your (separate) boot partition's disk usage. Close to full...

 

Highlighted
Knowledge Partner
Knowledge Partner

Re: OES 2015 -> 2018 grub boot failure

Good catch. Had the same issue on one server that had 4 or 5 leftover copies of SLES11 kernels hanging around there. It then failed to install the new kernel during patching.

CU,
--
Massimo Rosen
Micro Focus Knowledge Partner
No emails please!
http://www.cfc-it.de
0 Likes
The opinions expressed above are the personal opinions of the authors, not of Micro Focus. By using this site, you accept the Terms of Use and Rules of Participation. Certain versions of content ("Material") accessible here may contain branding from Hewlett-Packard Company (now HP Inc.) and Hewlett Packard Enterprise Company. As of September 1, 2017, the Material is now offered by Micro Focus, a separately owned and operated company. Any reference to the HP and Hewlett Packard Enterprise/HPE marks is historical in nature, and the HP and Hewlett Packard Enterprise/HPE marks are the property of their respective owners.