How to restore a whole Cluster to new unknown Disk's

Old cluster configuration:

dsk0 => inst. disk

dsk1 => member1 bootdisk

dsk2 => member2 bootdisk

dsk3 => cluster disk

dsk4 => quorum disk

root@conair:/> hwmgr -show scsi

SCSI DEVICE DEVICE DRIVER NUM DEVICE FIRST

HWID: DEVICEID HOSTNAME TYPE SUBTYPE OWNER PATH FILE VALID PATH

-------------------------------------------------------------------------

74: 0 conair cdrom none 0 1 cdrom0 [0/5/0]

75: 1 conair disk none 0 1 dsk0 [1/1/0]

76: 2 conair disk none 2 1 dsk1 [1/1/1]

77: 3 conair disk none 0 1 dsk2 [1/1/2]

78: 4 conair disk none 2 1 dsk3 [1/1/3]

79: 5 conair disk none 2 1 dsk4 [1/1/4]

80: 7 conair changer none 0 1 mc0 [2/3/1]

81: 6 conair tape none 0 1 tape0 [2/3/0]

hsz-configuration old/new one:

HSZ> sho disk

Name Type Port Targ Lun Used by

------------------------------------------------------------------------------

DISK300 disk 3 0 0 D100

DISK320 disk 3 2 0 D101

DISK340 disk 3 4 0 D102

DISK410 disk 4 1 0 D103

DISK430 disk 4 3 0 D104

to be shure, that all data are lost on these disk's and new disk's will become a new wwid,

i deleted all and regenerated it in a different unit, except inst. disk.

HSZ> del d104

HSZ> del d103

HSZ> del d102

HSZ> del d101

HSZ> del disk430

HSZ> del disk410

HSZ> del disk340

HSZ> del disk320

HSZ> run config

Config Local Program Invoked

Config is building its tables and determining what devices exist

on the subsystem. Please be patient.

add disk DISK320 3 2 0

add disk DISK340 3 4 0

add disk DISK410 4 1 0

add disk DISK430 4 3 0

Config - Normal Termination

HSZ> init disk320

HSZ> init disk340

HSZ> init disk410

HSZ> init disk430

HSZ> add unit d201 disk320

HSZ> add unit d202 disk340

HSZ> add unit d203 disk410

HSZ> add unit d204 disk430

HSZ> sho unit

LUN Uses

--------------------------------------------------------------

D100 DISK300

D201 DISK320

D202 DISK340

D203 DISK410

D204 DISK430

HSZ>

First we have to boot inst. disk....

P00>>> sho dev

polling ncr0 (NCR 53C810) slot 1, bus 0 PCI, hose 1 x SCSI Bus ID 7

dka500.5.0.1.1 DKA500 RRD45 0436

polling pks0 (DEC KZPSA) slot 3, bus 0 PCI, hose 1

pkb.7.0.3.1 dkb TPwr 1 Fast 1 Bus ID 7 R01 A12

dkb100.1.0.3.1 DKB100 HSZ50-AX V57Z

dkb201.2.0.3.1 DKB201 HSZ50-AX V57Z

dkb202.2.0.3.1 DKB202 HSZ50-AX V57Z

dkb203.2.0.3.1 DKB203 HSZ50-AX V57Z

dkb204.2.0.3.1 DKB204 HSZ50-AX V57Z

polling isp0 (QLogic ISP10X0) slot 0, bus 2 PCI, hose 1 SCSI Bus ID 7

P00>>>

P00>>> b dkb100

Initializing...

the booted inst. disk has now stale pathes on "old" disk's.

root@conair:/> hwmgr -show scsi

SCSI DEVICE DEVICE DRIVER NUM DEVICE FIRST

HWID: DEVICEID HOSTNAME TYPE SUBTYPE OWNER PATH FILE VALID PATH

-------------------------------------------------------------------------

74: 0 conair cdrom none 0 1 cdrom0 [0/5/0]

75: 1 conair disk none 2 1 dsk0 [1/1/0]

76: 2 conair disk none 0 1 (null)

77: 3 conair disk none 0 1 (null)

78: 4 conair disk none 0 1 (null)

79: 5 conair disk none 0 1 (null)

80: 7 conair changer none 0 1 mc0 [2/3/1]

81: 6 conair tape none 0 1 tape0 [2/3/0]

85: 8 conair disk none 0 1 dsk5 [1/2/1]

86: 9 conair disk none 0 1 dsk6 [1/2/2]

87: 10 conair disk none 0 1 dsk7 [1/2/3]

88: 11 conair disk none 0 1 dsk8 [1/2/4]

To fix this problem we will execute the cleanup-script for single system where we can download from this location:

http://www-unix.zuo.cpqcorp.net/~reto/single_v5.htm

<![endif]>

root@conair:/tmp> ./cleanup_hwdb.sh

This procedure brings down the system with a clean hardware

database, in order to be able to boot this system disk in

a system having a different architecture.

Do you want to continue (y/[n]): ? y

NOTE: The next reboot should stop in Single User Mode on genvmunix,

Before going to multiuser mode you should execute

mountroot, and fix /etc/fdmns and /etc/sysconfigtab if needed.

Then, perform a doconfig and reboot on it.

Copying /genvmunix to /vmunix...Done.

root@conair:/tmp>

INIT: New run level: 0

The system is coming down. Please wait...

Logins disabled

LAT stopped.

Unmounting NFS filesystems

Halting processes ...

The system is down.

NOTE: The next reboot should stop in Single User Mode on genvmunix,

Before going to multiuser mode you should execute

mountroot, and fix /etc/fdmns and /etc/sysconfigtab if needed.

Then, perform a doconfig and reboot on it.

/proc: Invalid argument

....Halt completed....

syncing disks... CP - SAVE_TERM routine to be called

CP - SAVE_TERM exited with hlt_req = 1, r0 = 00000000.00000000

halted CPU 0

halt code = 5

HALT instruction executed

PC = fffffc00005e43c0

P00>>> b dkb100 -fi genvmunix -fl s

Initializing...

INIT: SINGLE-USER MODE

@:$PWD> mountroot

Mounting / (root)

msfs_mount: error getting disk list for domain "root_domain"

msfs_mount: Setting root device name to "root_device" RW

user_cfg_pt: reconfigured

root_mounted_rw: reconfigured

user_cfg_pt: reconfigured

root_mounted_rw: reconfigured

Device Naming: first boot initialization . . .

etc/cfginfo -> ../cluster/members/member/etc/cfginfo

cluster/members/member0/etc/cfginfo

etc/dccd.dat

etc/dcdd.dat

etc/dfsc.dat

etc/dfsl.dat -> ../cluster/members/member/etc/dfsl.dat

cluster/members/member0/etc/dfsl.dat

ptm

disk rdisk tape ntape dmapi

user_cfg_pt: reconfigured

dsfmgr: NOTE: creating device special files for system at /

+scp_scsi +kevm +kevm.pterm +kevm.debug -kevm.debug +tty00 +tty01 +lp0 +floppy0a +floppy0a +floppy0b -floppy0b +floppy0b -floppy0b +floppy0c +floppy0c +floppy0d -floppy0d +floppy0d -floppy0d +floppy0e -floppy0e +floppy0e -floppy0e +floppy0f -floppy0f +floppy0f -floppy0f +floppy0g -floppy0g +floppy0g -floppy0g +floppy0h -floppy0h +floppy0h -floppy0h +cdrom0a +cdrom0a +cdrom0b -cdrom0b +cdrom0b -cdrom0b +cdrom0c +cdrom0c +cdrom0d -cdrom0d +cdrom0d -cdrom0d +cdrom0e -cdrom0e +cdrom0e -cdrom0e +cdrom0f -cdrom0f +cdrom0f -cdrom0f +cdrom0g -cdrom0g +cdrom0g -cdrom0g +cdrom0h -cdrom0h +cdrom0h -cdrom0hAssigning a cluster device number to root

+dsk0a +dsk0a +dsk0b +dsk0b +dsk0c +dsk0c +dsk0d +dsk0d +dsk0e +dsk0e +dsk0f +dsk0f +dsk0g +dsk0g +dsk0h +dsk0h +dsk1a +dsk1a +dsk1b +dsk1b +dsk1c +dsk1c +dsk1d +dsk1d +dsk1e +dsk1e +dsk1f +dsk1f +dsk1g +dsk1g +dsk1h +dsk1h +dsk2a +dsk2a +dsk2b +dsk2b +dsk2c +dsk2c +dsk2d +dsk2d +dsk2e +dsk2e +dsk2f +dsk2f +dsk2g +dsk2g +dsk2h +dsk2h +dsk3a +dsk3a +dsk3b +dsk3b +dsk3c +dsk3c +dsk3d +dsk3d +dsk3e +dsk3e +dsk3f +dsk3f +dsk3g +dsk3g +dsk3h +dsk3h +dsk4a +dsk4a +dsk4b +dsk4b +dsk4c +dsk4c +dsk4d +dsk4d +dsk4e +dsk4e +dsk4f +dsk4f +dsk4g +dsk4g +dsk4h +dsk4h -mc0 +mc0 +tape0 +tape0 +tape0c +tape0c +tape0_d0 +tape0_d0 +tape0_d1 +tape0_d1 +tape0_d2 +tape0_d2 +tape0_d3 +tape0_d3 +tape0_d4 +tape0_d4 +tape0_d5 +tape0_d5 +tape0_d6 +tape0_d6 +tape0_d7 +tape0_d7 -scp0 +scp0

@:$PWD>

check the new generated hardware database

@:$PWD> hwmgr -show scsi

SCSI DEVICE DEVICE DRIVER NUM DEVICE FIRST

HWID: DEVICEID HOSTNAME TYPE SUBTYPE OWNER PATH FILE VALID PATH

-------------------------------------------------------------------------

76: 0 cdrom none 0 1 cdrom0 [0/5/0]

77: 1 disk none 2 1 dsk0 [1/1/0]

78: 2 disk none 0 1 dsk1 [1/2/1]

79: 3 disk none 0 1 dsk2 [1/2/2]

80: 4 disk none 0 1 dsk3 [1/2/3]

81: 5 disk none 0 1 dsk4 [1/2/4]

82: 7 changer none 0 1 mc0 [2/3/1]

83: 6 tape none 0 1 tape0 [2/3/0]

Important!!!

when you have lost your inst. disk, then install a new single system with trucluster-software

and the same patchkit where orig. cluster was running!!

now customize your disklabels on other disks and create file-domain and file-set for cluster_root cluster_var and cluster_usr.

root@conair:/> mkfdmn /dev/disk/dsk3a cluster_root

root@conair:/> mkfset cluster_root root

root@conair:/> mkfdmn /dev/disk/dsk3g cluster_usr

root@conair:/> mkfset cluster_usr usr

root@conair:/> mkfdmn /dev/disk/dsk3h cluster_var

root@conair:/> mkfset cluster_var var

restore cluster file-system

root@conair:/> mount cluster_root#root /mnt

root@conair:/> cd /mnt

root@conair:/mnt> vrestore -xf /dev/ntape/tape0

vrestore: Date of the vdump save-set: Mon May 28 14:03:15 2001

vrestore: Save-set source directory : /

update all Links in /mnt/etc/fdmns (when needed)

root@conair:/mnt> cd

root@conair:/> umount /mnt

root@conair:/> mount cluster_var#var /mnt

root@conair:/> cd /mnt

root@conair:/mnt> vrestore -xf /dev/ntape/tape0

vrestore: Date of the vdump save-set: Mon May 28 14:04:44 2001

vrestore: Save-set source directory : /var

root@conair:/mnt> cd

root@conair:/> umount /mnt

root@conair:/> mount cluster_usr#usr /mnt

root@conair:/> cd /mnt

root@conair:/mnt> vrestore -xf /dev/ntape/tape0

vrestore: Date of the vdump save-set: Mon May 28 14:06:46 2001

vrestore: Save-set source directory : /usr

root@conair:/mnt> cd

root@conair:/> umount /mnt

create now member-boot disk's

root@conair:/> clu_bdmgr -c dsk1 1

*** Error ***

Bad disk label.

Creating AdvFS domains:

Creating AdvFS domain 'root1_domain#root' on partition

'/dev/disk/dsk1a'.

root@conair:/> clu_bdmgr -c dsk2 2

*** Error ***

Bad disk label.

Creating AdvFS domains:

Creating AdvFS domain 'root2_domain#root' on partition

'/dev/disk/dsk2a'.

restore now member1-boot disk

root@conair:/> mount root1_domain#root /mnt

root@conair:/> cd /mnt

root@conair:/mnt> vrestore -xf /dev/ntape/tape0

vrestore: Date of the vdump save-set: Mon May 28 14:04:44 2001

vrestore: Save-set source directory : /cluster/members/member1/boot_partition

get info for sysconfigtab modifications

root@conair:/mnt> file /dev/disk/dsk1h (member1 cnx partition)

/dev/disk/dsk1h: block special (19/63)

root@conair:/mnt> file /dev/disk/dsk4h (quorum cnx partition)

/dev/disk/dsk4h: block special (19/111)

edit on /mnt/etc/sysconfigtab these parameters:

swapdevice=/dev/disk/dsk1b

cluster_seqdisk_minor=63

cluster_qdisk_minor=111

restore of member2-boot disk

root@conair:/> mount root2_domain#root /mnt

root@conair:/> cd /mnt

root@conair:/mnt> vrestore -xf /dev/ntape/tape0

vrestore: Date of the vdump save-set: Mon May 28 14:02:13 2001

vrestore: Save-set source directory : /cluster/members/member2/boot_partition

get info for sysconfigtab modifications

root@conair:/mnt> file /dev/disk/dsk2h (member2 cnx partition)

/dev/disk/dsk2h: block special (19/79)

root@conair:/mnt> file /dev/disk/dsk4h (quorum cnx partition)

/dev/disk/dsk4h: block special (19/111)

edit on /mnt/etc/sysconfigtab these parameters

swapdevice=/dev/disk/dsk2b

cluster_seqdisk_minor=79

cluster_qdisk_minor=111

modify and update now members cnx partitions

If cluster_root disk has changed, you have to edit the configuration file to the new cluster_root!!

root@conair:/mnt> cat /mnt/etc/clu_bdmgr.conf

# clu_bdmgr configuration file

# DO NOT EDIT THIS FILE

::TYP:m:CFS:/dev/disk/dsk3a:: (cluster_root domain)

root@conair:/mnt>clu_bdmgr -h dsk1 /mnt/etc/clu_bdmgr.conf (for member1)

root@conair:/mnt>clu_bdmgr -h dsk2 /mnt/etc/clu_bdmgr.conf (for member2)

recover now the cluster Hardware Database

- mount (member1) to /mnt (mount root1_domain#root /mnt)

- mount cluster-root on /mnt1 (mount cluster_root#root /mnt1)

- delete files as indicated (rm)

/mnt/etc/dec*

/mnt/etc/ddr*

/mnt1/etc/dfsc*

/mnt1/etc/dec_unid_db*

/mnt1/etc/dec_hwc_cdb*

/mnt1/etc/dccd*

/mnt1/etc/dcdd*

/mnt1/cluster/members/member1/.Booted

/mnt1/cluster/members/member1/etc/dfsl*

/mnt1/cluster/members/member1/etc/cfginfo

/mnt1/cluster/members/member1/dev/tty0*

/mnt1/cluster/members/member1/dev/lp*

/mnt1/cluster/members/member1/dev/kevm*

/mnt1/cluster/members/member1/dev/scp*

/mnt1/cluster/members/member1/dev/disk

/mnt1/cluster/members/member1/dev/rdisk

/mnt1/cluster/members/member1/dev/ntape

/mnt1/cluster/members/member1/dev/tape

and these directories (rm -r)

/mnt1/devices/disk

/mnt1/devices/rdisk

/mnt1/devices/tape

/mnt1/devices/ntape

copy the following files from install disk

/etc/dfsc* to /mnt1/etc/

/etc/dec_unid_db* to /mnt1/etc/

/etc/dec_hwc_cdb* to /mnt1/etc/

/etc/dccd* to /mnt1/etc/

/etc/dcdd* to /mnt1/etc/

/etc/ddr* to /mnt/etc/

/etc/dec_devsw* to /mnt/etc/

/etc/dec_hw_db* to /mnt/etc/

/etc/dec_hwc_ldb* to /mnt/etc/

/etc/dec_scsi* to /mnt/etc/

/etc/dfsl* to /mnt1/cluster/members/member1/etc/

/etc/cfginfo to /mnt1/cluster/members/member1/etc/

shutdown install-disk and boot member1-bootdisk (where we copied files to)

to single-user mode with genvmunix and clubase options. (b -fl "ias" -fi genvmunix dkxy)

CPU 0 booting

(boot dkb201.2.0.3.1 -file genvmunix -flags ias)

block 0 of dkb201.2.0.3.1 is a valid boot block

reading 18 blocks from dkb201.2.0.3.1

bootstrap code read in

Building FRU table

base = 200000, image_start = 0, image_bytes = 2400

initializing HWRPB at 2000

initializing page table at 1f2000

initializing machine state

setting affinity to the primary CPU

jumping to bootstrap code

UNIX boot - Thursday August 24, 2000

Enter <kernel_name> [option_1 ... option_n]

Press Return to boot default kernel 'genvmunix': genvmunix clubase:cluster_expected_votes=1 clubase:cluster_qdisk_votes=0

Loading genvmunix ...

Loading at 0xffffffff00000000

Sizes:

text = 9366208

data = 2498784

bss = 4458512

Starting at 0xffffffff00010af0

Kernel argument clubase:cluster_expected_votes=1

Kernel argument clubase:cluster_qdisk_votes=0

Alpha boot: available memory from 0x286c000 to 0x3fffa000

Compaq Tru64 UNIX P5.1-10 (Rev. 388); Thu Apr 12 22:02:37 EDT 2001

physical memory = 1024.00 megabytes.

available memory = 983.54 megabytes.

using 3856 buffers containing 30.12 megabytes of memory

Firmware revision: 5.8

Checking device naming:

Checking local filesystems

Mounting / (root)

msfs_mount: error getting disk list for domain "cluster_root"

msfs_mount: Setting root device name to "root_device" RW

user_cfg_pt: reconfigured

root_mounted_rw: reconfigured

Mounting /cluster/members/member1/boot_partition (boot filesystem)

msfs_mount: error getting disk list for domain "root1_domain"

msfs_mount: Setting boot device name to "local_root" RW

user_cfg_pt: reconfigured

root_mounted_rw: reconfigured

Device Naming: first boot initialization . . .

ptm

disk rdisk tape ntape changer

user_cfg_pt: reconfigured

dsfmgr: NOTE: creating device special files for system at /

+scp_scsi +kevm +kevm.pterm +tty00 +tty01 +lp0 +dsk0a +dsk0a +dsk0b +dsk0b +dsk0c +dsk0c +dsk0d +dsk0d +dsk0e +dsk0e +dsk0f +dsk0f +dsk0g +dsk0g +dsk0h +dsk0h +dsk1a +dsk1a +dsk1b +dsk1b +dsk1c +dsk1c +dsk1d +dsk1d +dsk1e +dsk1e +dsk1f +dsk1f +dsk1g +dsk1g +dsk1h +dsk1h +dsk2a +dsk2a +dsk2b +dsk2b +dsk2c +dsk2c +dsk2d +dsk2d +dsk2e +dsk2e +dsk2f +dsk2f +dsk2g +dsk2g +dsk2h +dsk2h +dsk3a +dsk3a +dsk3b +dsk3b +dsk3c +dsk3c +dsk3d +dsk3d +dsk3e +dsk3e +dsk3f +dsk3f +dsk3g +dsk3g +dsk3h +dsk3h +dsk4a +dsk4a +dsk4b +dsk4b +dsk4c +dsk4c +dsk4d +dsk4d +dsk4e +dsk4e +dsk4f +dsk4f +dsk4g +dsk4g +dsk4h +dsk4h +mc0 +tape0 +tape0 +tape0c +tape0c +tape0_d0 +tape0_d0 +tape0_d1 +tape0_d1 +tape0_d2 +tape0_d2 +tape0_d3 +tape0_d3 +tape0_d4 +tape0_d4 +tape0_d5 +tape0_d5 +tape0_d6 +tape0_d6 +tape0_d7 +tape0_d7 +floppy0a +floppy0a +floppy0c +floppy0c +cdrom0a +cdrom0a +cdrom0c +cdrom0c

dsfmgr: verify all datum for system at /

Default File Tree:

OK.

Device Class Directory Default Database:

OK.

Device Category to Class Directory Database:

OK.

Dev directory structure:

OK.

Device Status Files:

OK.

Dev Nodes:

ERROR: wrong ldev major #(81,64): /dev/cport/scp0 : 86

ERROR: wrong ldev minor #(81,64): /dev/cport/scp0 : 448

Errors: 2

Total errors: 2

bcheckrc: Device Naming failed boot configure or verify.

Please correct the problem and continue or reboot

CNX QDISK: Successfully claimed quorum disk, adding 0 vote.

INIT: SINGLE-USER MODE

@:$PWD>

if necessery... check if all devices are correct or move them to the right location

boot now to multiuser-mode. Fux some false maj/min ID's with "dsfmgr -vF". If there is a problem, then use "mountroot"

@:$PWD> dsfmgr -vF

dsfmgr: verify with fix all datum for system at /

Default File Tree:

OK.

Device Class Directory Default Database:

OK.

Device Category to Class Directory Database:

OK.

Dev directory structure:

OK.

Device Status Files:

OK.

Dev Nodes:

WARNING: wrong ldev major #(81,64): /dev/cport/scp0 : 86

WARNING: wrong ldev minor #(81,64): /dev/cport/scp0 : 448

OK.

Total warnings: 2

@:$PWD> dsfmgr -vF

dsfmgr: verify with fix all datum for system at /

Default File Tree:

OK.

Device Class Directory Default Database:

OK.

Device Category to Class Directory Database:

OK.

Dev directory structure:

OK.

Device Status Files:

OK.

Dev Nodes:

OK.

@:$PWD> hwmgr -show scsi

SCSI DEVICE DEVICE DRIVER NUM DEVICE FIRST

HWID: DEVICEID HOSTNAME TYPE SUBTYPE OWNER PATH FILE VALID PATH

-------------------------------------------------------------------------

76: 0 cdrom none 0 1 cdrom0 [0/5/0]

77: 1 disk none 0 1 dsk0 [1/1/0]

78: 2 disk none 2 1 dsk1 [1/2/1]

79: 3 disk none 0 1 dsk2 [1/2/2]

80: 4 disk none 2 1 dsk3 [1/2/3]

81: 5 disk none 2 1 dsk4 [1/2/4]

82: 7 changer none 0 1 mc0 [2/3/1]

83: 6 tape none 0 1 tape0 [2/3/0]

@:$PWD> mount -a

@:$PWD> mount

cluster_root#root on / type advfs (rw)

root1_domain#root on /cluster/members/member1/boot_partition type advfs (rw)

cluster_usr#usr on /usr type advfs (rw)

cluster_var#var on /var type advfs (rw)

/proc on /proc type procfs (rw)

@:$PWD> init 3

INIT: New run level: 3

Checking device naming:

Passed.

Checking local filesystems

Mounting local filesystems

exec: /sbin/mount_advfs -F 0x14000 cluster_root#root /

cluster_root#root on / type advfs (rw)

exec: /sbin/mount_advfs -F 0x4000 cluster_usr#usr /usr

cluster_usr#usr on /usr: Device busy

exec: /sbin/mount_advfs -F 0x4000 cluster_var#var /var

cluster_var#var on /var: Device busy

/proc on /proc: Device busy

Subsystem hwautoconfig was successfully configured.

May 29 16:50:16 update: started

Checking for Installation Tasks...

The following STREAMS devices were created:

Name Major Minor

---- ----- -----

/dev/ptmx 32 60

/dev/streams/bba 32 61

swapon: added /dev/disk/dsk1b as swap device.

Executing Installation Tasks...

The system is coming up. Please wait...

Checking for crash dumps

Initializing paging space

Mounting Memory filesystems

evmstart: Daemon started

security configuration set to default (BASE).

File /etc/sia/matrix.conf updated successfully.

Successful SIA initialization

The following devices were DELETED:

Name Major Minor

---- ----- -----

/dev/streams/xtiso/tcp5 32 65

/dev/streams/xtiso/udp5 32 64

/dev/streams/xtiso/tcp+ 32 63

/dev/streams/xtiso/udp+ 32 62

/dev/streams/xtiso/tcp 32 61

/dev/streams/xtiso/udp 32 60

/dev/streams/dlb 32 67

/usr/sbin/autopush: Can't push requested modules on STREAM for entry 36

/usr/sbin/autopush: Device (6,-1) already configured

Streams autopushes configured

NIFF daemon started

Configuring network

hostname: conair.zuo.cpqcorp.net

Loading LMF licenses

System error logger started

Binary error logger started

add net default: gateway 16.184.36.1

gateway daemon started

Setting kernel timezone variable

Setting the current time and date with ntpdate

Tue May 29 16:50:41 MEST 2001

Ntpdate succeeded.

starting cluster alias

cluster alias subsystem enabled

enable: reconfigured

aliasd: setting up NIFF monitor for interface tu0

ONC portmap service started

CAA daemon started

NIS domain name set to rle

ypbind: Secure mode sunos 3.x servers rejected.

ypbind daemon started

NFS mount daemon started

NFS export service started

Attempting to start `cluster_lockd` on member `conair`

cluster NFS Locking:

cluster rpc.statd started

cluster rpc.lockd started

Start of `cluster_lockd` on member `conair` succeeded.

NFS IO service started

NFS Locking:

rpc.statd started

rpc.lockd started

Mounting NFS filesystems

NFS automount service started

Preserving editor files

Clearing temporary files

Unlocking ptys

SMTP Mail Service started.

Network Time Service started

Environmental Monitoring Subsystem Configured.

Using snmp service entry port 161.

Extensible SNMP master agent started

Base O/S subagent started

Server System subagent started

Server Management subagent started

CIM SNMP subagent started

Performance Management subagent started

Web Based Management Agent started

ConfigReport Management Module started

SysMan Management Module started

Threshold Management Subagent started

Environmental Monitoring Daemon started.

AdvFS daemon (advfsd) started.

Internet services provided.

Cron service started

/sbin/rc3.d/S58lat ERROR: DLB not in running kernel; cannot use LAT

Printer service started

SysMan authentication server started

CAA Applications now started

Cluster sub-agent started

cluster wall daemon started

SysMan Station Server (smsd) started

The system is ready.

Compaq Tru64 UNIX P5.1-10 (Rev. 388) (conair.zuo.cpqcorp.net) console

Password:

Last login: Mon May 28 14:01:33 MEST 2001 from faceoff.zuo.cpqcorp.net

Compaq Tru64 UNIX V5.1 (Rev. 732); Wed May 23 17:22:38 MEST 2001

root@conair:/> clu_get_info

Cluster information for cluster faceoff

Number of members configured in this cluster = 2

memberid for this member = 1

Quorum disk = dsk4h

Quorum disk votes = 0

Information on each cluster member

Cluster memberid = 1

Hostname = conair.zuo.cpqcorp.net

Cluster interconnect IP name = conair-mc0

Member state = UP

Cluster memberid = 2

Hostname = tron.zuo.cpqcorp.net

Cluster interconnect IP name = tron-mc0

Member state = DOWN

root@conair:/> clu_quorum

Cluster Quorum Data for: faceoff as of Tue May 29 16:52:50 MEST 2001

Cluster Common Quorum Data

Quorum disk: dsk4h

File: /etc/sysconfigtab.cluster

Attribute File Value

expected votes 3

qdisk votes

qdisk major

qdisk minor

Member 1 Quorum Data

Host name: conair.zuo.cpqcorp.net Status: UP

File: /cluster/members/member1/boot_partition/etc/sysconfigtab

Attribute Running Value File Value

current votes 1 N/A

quorum votes 1 N/A

expected votes 1 3

node votes 1 1

qdisk votes 0 1

qdisk major 19 19

qdisk minor 111 111

Member 2 Quorum Data

Host name: tron.zuo.cpqcorp.net Status: DOWN

File: /cluster/members/member2/boot_partition/etc/sysconfigtab

root@conair:/>

Create member2-specific HW-DB

conair # clu_get_info

Cluster information for cluster faceoff

Number of members configured in this cluster = 2

memberid for this member = 1

Quorum disk = dsk3h

Quorum disk votes = 1

Information on each cluster member

Cluster memberid = 1

Hostname = conair.zuo.cpqcorp.net

Cluster interconnect IP name = conair-mc0

Member state = UP

Cluster memberid = 2

Hostname = tron.zuo.cpqcorp.net

Cluster interconnect IP name = tron-mc0

Member state = DOWN

mount root2_domain#root /mnt

rm /mnt/etc/dec*

rm /cluster/members/member2/.Booted

rm /cluster/members/member2/etc/dfsl*

rm /cluster/members/member2/etc/cfginfo

rm -rf /cluster/members/member2/dev/[a-z]*

cd /cluster/members/member2/dev/; ./MAKEDEV std

##############################################################################

Create genesis databases

------------------------

clu_bdmgr -d dsk1 >/tmp/dsk2.bd #dsk1 = bootdisk member1 (a known good one)

/usr/sbin/cluster/clu_partmgr -mg /tmp/dsk2.bd dsk2 #dsk2 = bootdisk member2

mv /etc/dec_hwc_genesis* /mnt/etc/

##############################################################################

cd /

umount /mnt

ATTENTION!!!

You have to boot member2 with the same flags and kernel option to single-usermode

(b -fl "ias" -f genvmunix dkb202) # Will fail if you forgot umount

for kernel option see above (on member1)

mountroot

dn_setup -init

dsfmgr -K

dsfmgr -v # optionally -vF

hwmgr show scsi

Now we should have both member up and running (without quorum-disk)

you are able now to delete quorum disk with “clu_quorum -f -d remove”

and add the new quorum with “clu_quorum -f -d add dsk4 1”

have fun!

cheers Reto