Tag Archives: boot

System Panic During Boot Logging the Error “NOTICE: zfs_parse_bootfs: error 19”

Today while migrating SAN i face this issue, hope it will help others too…

The system panic during boot logging the error:

{0} ok boot 56024-disk
Boot device: /virtual-devices@100/channel-devices@200/disk@1 File and args:
SunOS Release 5.10 Version Generic_147440-01 64-bit
Copyright (c) 1983, 2011, Oracle and/or its affiliates. All rights reserved.
NOTICE: zfs_parse_bootfs: error 19
Cannot mount root on rpool/68 fstype zfs
panic[cpu0]/thread=180e000: vfs_mountroot: cannot mount root

Changes

This issue usually occurs when system is trying to boot a ZFS rpool and the path to the disk changed, or customer is trying to boot the system from a cloned disk (that means the disk is a copy of another boot disks)

Cause

The issue is caused by a mismatch between the current path of the disk you are trying to boot from and the path stored in the ZFS label of the same disk:

ok boot 56024-disk
Boot device: /virtual-devices@100/channel-devices@200/disk@1 File and args:

 

# zdb -l /dev/rdsk/c0d1s0
——————————————–
LABEL 0
——————————————–
version: 29
name: ‘rpool’
state: 0
txg: 1906
pool_guid: 3917355013518575342
hostid: 2231083589
hostname: ”
top_guid: 3457717657893349899
guid: 3457717657893349899
vdev_children: 1
vdev_tree:
type: ‘disk’
id: 0
guid: 3457717657893349899
path: ‘/dev/dsk/c0d0s0
devid: ‘id1,vdc@f85a3722e4e96b600000e056e0049/a’
phys_path: ‘/virtual-devices@100/channel-devices@200/disk@0:a
whole_disk: 0
metaslab_array: 31
metaslab_shift: 27
ashift: 9
asize: 21361065984
is_log: 0
create_txg: 4

As you can see we are trying to boot the path disk@1 but in the ZFS label the path is disk@0.

Solution

To fix the issue you have to boot the system in failsafe mode or from cdrom and import the rpool on that disk to force ZFS to correct the path:

# zpool import -R /mnt rpool
cannot mount ‘/mnt/export’: failed to create mountpoint
cannot mount ‘/mnt/export/home’: failed to create mountpoint
cannot mount ‘/mnt/rpool’: failed to create mountpoint

# zdb -l /dev/rdsk/c0d1s0
——————————————–
LABEL 0
——————————————–
version: 29
name: ‘rpool’
state: 0
txg: 1923
pool_guid: 3917355013518575342
hostid: 2230848911
hostname: ”
top_guid: 3457717657893349899
guid: 3457717657893349899
vdev_children: 1
vdev_tree:
type: ‘disk’
id: 0
guid: 3457717657893349899
path: ‘/dev/dsk/c0d1s0
devid: ‘id1,vdc@f85a3722e4e96b600000e056e0049/a’
phys_path: ‘/virtual-devices@100/channel-devices@200/disk@1:a
whole_disk: 0
metaslab_array: 31
metaslab_shift: 27
ashift: 9
asize: 21361065984
is_log: 0
create_txg: 4

As you can see the path has been corrected, however you have also to remove the zpool.cache file otherwise after boot the ZFS command will still show the disk as c0d0:

# zfs list
NAME USED AVAIL REFER MOUNTPOINT
rpool 5.86G 13.7G 106K /mnt/rpool
rpool/ROOT 4.35G 13.7G 31K legacy
rpool/ROOT/s10s_u10wos_17b 4.35G 13.7G 4.35G /mnt
rpool/dump 1.00G 13.7G 1.00G –
rpool/export 63K 13.7G 32K /mnt/export
rpool/export/home 31K 13.7G 31K /mnt/export/home
rpool/swap 528M 14.1G 114M –

# zfs mount rpool/ROOT/s10s_u10wos_17b
# cd /mnt/etc/zfs
# rm zpool.cache

Boot the IBM Aix system into Service mode

This document describes how to boot the system into Service mode (also known as Maintenance mode) to install the machine, restore an operating system backup, or perform maintenance on the rootvg volume group.

The information in this document applies to AIX Versions 3.x, 4.x and 5.x.

Booting microchannel systems into Service mode Booting  PCI-based systems into Service mode PCI machine-specific information Accessing rootvg and mounting file systems
Related documentation

——————————————————————————–

Booting microchannel systems into Service mode

To boot microchannel systems into Service mode, turn the key to the Maintenance  position and press the yellow reset button twice. You must boot from bootable  media, such as an installation CD-ROM, installation tape, or a bootable backup  tape made via the mksysb command or the Sysback product of the correct level for  this  machine.

For AIX Version 3.2, you may use bootable bosboot diskettes. To boot from these,  insert the first bosboot diskette into the diskette drive. When you see LED c07,  insert the next diskette, which is usually the display extensions diskette.  After this diskette is read, you should receive a menu prompting you for the  installation diskette.

For information on accessing your rootvg volume group, see the section entitled  “Accessing rootvg and mounting file systems”.

The preceding discussion assumes that the Service mode bootlist has not been  modified from the default bootlist. If the bootlist has been modified, it must  be reset such that one of the boot media types from the preceding selections is  before the standard boot media, such asthe hard disk.

If the machine is an SMP model (7012-Gxx, 7013-Jxx, and 7015-Rxx) and the  Autoservice IPL flag is disabled, then a menu like the following will display  when it is booting in Service mode:

MAINTENANCE MENU (Rev. 04.03)
0> DISPLAY CONFIGURATION
1> DISPLAY BUMP ERROR LOG
2> ENABLE SERVICE CONSOLE
3> DISABLE SERVICE CONSOLE
4> RESET
5> POWER OFF
6> SYSTEM BOOT
7> OFF-LINE TESTS
8> SET PARAMETERS
9> SET NATIONAL LANGUAGE
SELECT:

You can boot these machines into Service mode or even Normal mode with the Fast  IPL Flag set. If you do not, the machine can take anywhere from 30 to 60 minutes  to boot up. There are a few ways to set the Fast IPL Flag for these machines.

NOTE: The console must be an ASCII type and connected to the S1 port of the  system. Graphic monitors will not work.

Use the following instructions to boot SMP machines into service with Fast IPL set.

Insert the bootable media of the same OS Level. Use the mksysb/cd-rom command.
Turn off the machine by pressing the white button on front.
Turn the key to the Wrench or Service position.
The LCD should read STAND-BY.
Press the Enter key on the console.
A greater-than prompt (>) should display on the monitor.
Type in sbb followed by the Enter key.
The menu Stand By Menu should now display.
Select 1 Set Flags. This will take you to another set of menus.
Select 6 Fast IPL. This should change to enable after it is selected.
Enter x to exit the second set of menus.
Enter x to exit the first menu.
At a blank screen, press the Enter key to obtain the greater-than prompt (>).
Type in the word power followed by the Enter key.
Turn the machine back on. It should start to boot up. A prompt may display asking
if you want to update the firmware. Do not respond; let it continue.
Now you may be at the Maintenance Menu with 10 options displayed, 0 through 9. If
that is the case, select option 6, System Boot. This will take you to another
menu. Select option 0, Boot from the list.
The Standard Maintenance Menu should display. System recovery and maintenance
can be completed from here.
After system recovery and maintenance has been performed, the system is ready to
be rebooted into Normal mode. Enter the command mpcfg -cf 11 1 at the command
line prompt, then press Enter. This will set the Fast IPL Flag. The system is
ready to reboot.
Turn the key back to the OK/Normal position.
Enter shutdown -Fr, followed by the Enter key.


——————————————————————————–

Booting PCI-based systems into Service mode

When booting a PowerPC into Service mode, cd0 or rmt0 must be before the hdisk in the bootlist. If not, change the bootlist at boot time. On some models, you can set the machine to use a default bootlist that includes both cd0 and rmt0. If a bootable CD or tape is in the CD-ROM or tape drive, the machine will boot from this device.

For most of the newer PCI-based models, selecting the default bootlist, with a bootable tape or CD loaded in the machine, causes the system to automatically boot from that device. Generally, the next menu on the screen asks the administrator to define the system console.

For all machines discussed here, if you are using a graphical terminal, you will use a function key such as F5. If you are using an ASCII terminal, use an equivalent number key such as 5. Use the numbers across the top of the keyboard, not the numbers on the numeric keypad. On ASCII terminals, the icons may not be displayed on the screen; the number can be pressed between the second and third beeps, the second beep being a series of three clicks.


——————————————————————————–

PCI machine-specific information
The following systems all use the F5 or 5 key to read from the default boot list, which is written into the system firmware:

MODEL       7017       7024       7025       7026       7043       7137
——-   ——-    ——-    ——-    ——-    ——-    ——-
TYPE      S70        E20        F30        H10        43P-140    F3L
S7A        E30        F40        H50        43P-150
S80                   F50        H70        43P-240
B80        43P-260

On these machines, use 5 (on the keyboard, not the keypad) if you are using an ASCII terminal. On a locally attached graphics console, use the F5 function key. The F5 or 5 key must be pressed just after the keyboard icon or message is displayed on the console. If you have either a 7026-M80, 7026-H80 or a 7025-F80, then the 5 key will be the default whether you have an ascii or graphics console.

The following systems use the F1 key to enter System Management Services mode (SMS):

MODEL       6040       7042       7247       7249
——-   ——-    ——-    ——-    ——-
TYPE        620        850        82x        860

You should be in an Easy-Setup menu. Select the Start Up menu. Clear the current bootlist settings and then select the CD-ROM for choice 1 and hdd (the hard disk) for choice 2. Select OK. Insert the CD-ROM and select the EXIT icon. The machine should now boot from the CD-ROM.

The following systems use the F2 key to enter SMS:

MODEL         6015       6050       6070       7020       7248
——-     ——-    ——-    ——-    ——-    ——-
TYPE          440        830        850        40P        43P

Select Select Boot Device from the initial menu on the screen, and then select Restore Default Settings from the list. Press the Esc key to exit all the menus, and then reboot the machine. The system should boot from your bootable media.

For information on accessing the rootvg volume group, see the next section in this document.

——————————————————————————–

Accessing rootvg and mounting file systems
For AIX Version 3, choose the limited function maintenance shell (option 5 for AIX 3.1, option 4 for AIX 3.2).

If you only have one disk on the system, then hdisk0 will be used in the execution of the getrootfs or /etc/continue commands, which follow. If you have more than one disk, determine which disk contains the boot logical volume in this manner:

AIX 3.2.4 or AIX 3.2.5:

Run getrootfs; the output will indicate which disk contains the hd5 logical volume.

AIX 3.1 to AIX 3.2.3e:

Run lqueryvg -Ltp hdisk# for each hdisk. You can obtain a listing of these with the command lsdev -Cc disk. Repeat this command until you get output similar to the following:

00005264feb3631c.2  hd5  1

If more than one disk contains this output, use any disk when running getrootfs.
Now, access the rootvg volume group by running one of the following commands, using the disk you obtained in the preceding step:

AIX 3.1:                     /etc/continue hdisk#
AIX 3.2.0-3.2.3e:            getrootfs -f hdisk#
AIX 3.2.4-3.2.5:             getrootfs hdisk#

NOTE: If you want to leave the primary OS file systems (/, /usr, /tmp, and /var) unmounted after this command has completed, to run fsck, for instance, place a space and the letters sh after the hdisk in the preceding command. For example:

getrootfs hdisk0 sh

For AIX Versions 4 and 5, choose Start Maintenance Mode for System Recovery , option 3. The next screen will be called Maintenance; select option 1, Access a Root Volume Group. At the next screen, type 0 to continue, and select the appropriate volume group by typing the number next to it. A screen like the following will display.
Example:

Access a Root Volume Group

Type the number for a volume group to display the logical volume information and press Enter.

1)  Volume Group 0073656f2608e46a contains these disks:
hdisk0  2063 04-C0-00-4,0

Once a volume group has been selected, information will be displayed about that volume group.

Example:

Volume Group Information
——————————————————————————
Volume Group ID 0073656f2608e46a includes the following logical volumes:
hd6         hd5         hd8         hd4         hd2      hd9var
hd3         hd1
——————————————————————————

Type the number of your choice and press Enter.

1) Access this Volume Group and start a shell
2) Access this Volume Group and start a shell before mounting filesystems
99) Previous Menu

If the logical volumes listed do not include logical volumes like hd4, hd2, hd3, and so on, you may have selected the wrong volume group. Press 99 to back up one screen and select again.

Now you may select one of two options: Access this volume group and start a shell , option 1, or Access this volume group and start a shell before mounting file systems , option 2. Option 2 allows you to perform file system maintenance on /, /usr, /tmp, and /var before mounting them.

NOTE: If you intend to use SMIT or vi, set your terminal type in preparation for editing the file. xx stands for a terminal type such as lft, ibm3151, or vt100.

TERM=<xx>
export TERM

Errors from these steps may indicate failed or corrupt disks in rootvg. These problems should
be corrected. For additional assistance, contact your vendor, your local branch office, or your AIX support center.

——————————————————————————–

Related documentation
For more in-depth coverage of this subject, the following IBM publication is recommended:
AIX Version 4.3 System Management Guide: Operating System and Devices
AIX Version 5.1 System Management Guide: Operating System and Devices

IBM documentation can also be accessed online through the following URL:
http://www.rs6000.ibm.com/resource/aix_resource/Pubs/index.html

Similar documents can be accessed through the following URL:
http://techsupport.services.ibm.com/server/support?view=pSeries

How to configure a VMware ESX / ESXi host with a QLogic HBA to boot from SAN in a CLARiiON environment

How to configure a VMware ESX / ESXi host with a QLogic HBA to boot from SAN in a CLARiiON environment

Solution:

This procedure explains how to configure the QLogic HBA to boot ESX/ESXi from SAN. The procedure involves enabling the QLogic HBA BIOS, enabling the selectable boot, and selecting the boot LUN.

Procedure:

1. While booting the server, press Ctrl+Q to enter the Fast!UTIL configuration utility.

2. Perform the appropriate action depending on the number of HBAs.

Option Description
One HBA If you have only one host bus adapter (HBA), the Fast!UTIL Options page appears. Skip to Step 3.
Multiple HBAs If you have more than one HBA, select the HBA manually.

1. In the Select Host Adapter page, use the arrow keys to position the cursor on the appropriate HBA.

2. Press Enter.

3. In the Fast!UTIL Options page, select Configuration Settings and press Enter.

4. In the Configuration Settings page, select Adapter Settings and press Enter.

5. Set the BIOS to search for SCSI devices.

a. In the Host Adapter Settings page, select Host Adapter BIOS.

b. Press Enter to toggle the value to Enabled.

c. Press Esc to exit.

6. Enable the selectable boot.

a. Select Selectable Boot Settings and press Enter.

b. In the Selectable Boot Settings page, select Selectable Boot.

c. Press Enter to toggle the value to Enabled.

7. Use the cursor keys to select the Boot Port Name entry in the list of storage processors (SPs) and press Enter to open the Select Fibre Channel Device screen.

8. Use the cursor keys to select the specific SP and press Enter.

If you are using an active-passive storage array, the selected SP must be on the preferred (active) path to the boot LUN. If you are not sure which SP is on the active path, use your storage array management software to find out. The target IDs are created by the BIOS and might change with each reboot.

9. Perform the appropriate action depending on the number of LUNs attached to the SP.

Option Description
One LUN The LUN is selected as the boot LUN. You do not need to enter the Select LUN screen.
Multiple LUNs Select LUN screen opens. Use the cursor to select the boot LUN, then press Enter.

To configure a CLARiiON array to serve as a boot device for a Solaris server, follow these steps:

Note: Check the EMC Support Matrix or E-Lab Navigator for the versions of Solaris and arrays that support using the array as a boot device.

1. Partition your LUN on your CLARiiON array so you have the slices of required sizes.

2. Run the newfs command to make a filesystem on the slices you need.

3. Make a mount point for the slice that you are going to copy to the LUN.

4. Mount the slice at the mount point.

5. Use the cd command to change your current directory to the mounted slice.

6. Run the following command to copy the slice to the array:

# ufsdump 0f – /dev/dsk/cxtxdxsx | ufsrestore rf –

Where x= controller,target,LUN, slice where the OS currently resides.

7. Run the command to copy a boot block to the LUN:

# /usr/sbin/installboot /usr/platform/sun4u/lib/fs/ufs/bootblk /dev/rdsk/cxtxdxsx

Where x= controller, target, LUN, and slice of array LUN.

8. Change /etc/vfstab to the new slices.

9. If ATF is to be part of this configuration, you must install it after setting up the boot partition.