What exactly do I need to do to prepare our Tru64 box for a disaster? Do I run btcreate? Or do I run vdump/vrestore? Do I use both? What would the restore steps be? I am not a Tru64 person but am being held responsible. Any help would be GREATLY appreciated.
This document provides details of a generic procedure to perform a disaster recovery of a Tru64 V5.X AlphaServer called alpha.
Please note, it is assumed that although an exact duplicate of the existing system is not required, the system should be of as similar a type as possible and include sufficient disk space to restore the complete Operating System as well as all data filesystems.
Contents
ABSTRACT 1
1. ALPHA DISASTER RECOVERY 3
1.1 OPERATING SYSTEM RECOVERY 3
1.2 DATA FILESYSTEM RECOVERY 13
· ALPHA Disaster recovery
· Operating System Recovery
The following is a procedure to rebuild a generic Tru64 V5.X AlphaServer system on a replacement system, using a TRU64 Unix Operating System CD to boot to system management mode.
The disk storage used in this example was provided via an HSZ80 dual redundant controller. The procedure is the same regardless of whether the disk storage is provided via SAN, local RAID controllers (KZPCC) or local SCSI disks.
Please Note : It is highly recommended that for Disaster Recovery purposes, regular sys_checks be run and the output saved to one or more remote systems for easy recovery & reference when required.
· Power up the system to firmware level ( >>> prompt )
· Verify the firmware is at an appropriate level, if not the run a firmware upgrade.
P00>>>show version
version v6.1-3 Oct 15 2001 14:00:54
· At the >>> prompt verify the disks available and the CD drive, enter show dev which will display all devices detectable by the firmware. In the case of the example disaster recovery, this displayed the six hard disks at dkc0 upto dkc300, the cd device at dqa0, and a TZ90 40/80 DLT tape device at mkb400, as shown below:
P00>>>show dev
dkc0.0.0.2.0 DKC0 HSZ80 V83Z
dkc1.0.0.2.0 DKC1 HSZ80 V83Z
dkc2.0.0.2.0 DKC2 HSZ80 V83Z
dkc200.2.0.2.0 DKC200 HSZ80 V83Z
dkc3.0.0.2.0 DKC3 HSZ80 V83Z
dkc300.3.0.2.0 DKC300 HSZ80 V83Z
dqa0.0.0.15.0 DQA0 Compaq CRD-8402B 1.03
dva0.0.0.1000.0 DVA0
mkb400.4.0.1.0 MKB400 COMPAQ SDT-10000 1.12
eia0.0.0.6.1 EIA0 00-02-A5-42-B6-EF
pka0.7.0.1.1 PKA0 SCSI Bus ID 7
pkb0.7.0.1.0 PKB0 SCSI Bus ID 7 5.57
pkc0.7.0.2.0 PKC0 SCSI Bus ID 7 5.57
P00>>>
· Ensure the AlphaServer firmware settings are correct enter the following commands
P00>>> set os_type unix
P00>>> set auto_action halt
P00>>> set bootdef_dev “”
P00>>> set boot_osflags 0
P00>>> set eia0_mode fastfd
P00>>> init
· Insert the TRU64 Unix V5.1A cd and boot up, enter boot dqa0. If a graphical console is connected, then the TRU64 Unix installation GUI will be displayed. Select the Unix Shell radio button at the bottom of the screen. This will return you to the # prompt in Unix system maintenance mode. Otherwise, if a dumb terminal is connected as the console you will be presented with the following menu:
The following options are available:
o The "U.S. English Installation" installs the base operating system
software.
o The "Installation with Worldwide Language Support" (WLS) lets you
internationalize your system. This option installs the base operating
system software as well as WLS software. The additional software subsets
provide support for various countries and their native languages.
o The "Exit Installation" option stops the installation and puts your
system in single-user mode with superuser privileges. This option
is intended for experienced UNIX system administrators who want to
perform file system or disk maintenance tasks before the installation.
This option may also be used for disaster recovery on a previously
installed system.
Remember, you can always get extra information by typing help.
1) U.S. English Installation
2) Installation with Worldwide Language Support
3) Exit Installation
Enter your choice: 3
Select option 3 to enter Unix system maintenance mode.
· At the # prompt, verify that you can locate all disk devices & their location.
# hwmgr -view devices
HWID: Device Name Mfg Model Location
------------------------------------------------------------------------------
6: (unknown)
7: (unknown) (unknown) (unknown)
8: (unknown)
47: /dev/disk/floppy0c 3.5in floppy fdi0-unit-0
52: /dev/disk/dsk0c DEC HSZ80 bus-2-targ-0-lun-0
53: /dev/disk/dsk1c DEC HSZ80 bus-2-targ-0-lun-1
54: /dev/disk/dsk2c DEC HSZ80 bus-2-targ-0-lun-2
55: /dev/disk/dsk3c DEC HSZ80 bus-2-targ-0-lun-3
56: /dev/disk/dsk4c DEC HSZ80 bus-2-targ-2-lun-0
57: /dev/disk/dsk5c DEC HSZ80 bus-2-targ-3-lun-0
58: /dev/disk/cdrom0c COMPAQ CRD-8402B bus-3-targ-0-lun-0
59: /dev/ntape/tape0 COMPAQ SDT-10000 bus-1-targ-4-lun-0
· Ensure the system disk is labelled with the correct partition sizes.
# disklabel -z dsk0
# disklabel -wr -t advfs dsk0 hsz80
#
# TERM=vt100
# EDITOR=vi
# export TERM EDITOR
#
# disklabel -e dsk0 edit & write label similar to that below
vrestore: Date of the vdump save-set: Mon Sep 30 10:27:02 2002
vrestore: Save-set source directory : /
vrestore: informational: [13] posting event: sys.unix.fs.advfs.fset.backup.lock
If running in single user mode, EVM is not running.
Please ignore this posting.
vrestore: informational: [13] posting event: sys.unix.fs.advfs.fset.backup.unlock
If running in single user mode, EVM is not running.
Please ignore this posting.
· Mount usr fileset & restore usr filesystem
# mount -t advfs usr_domain#usr /mnt/usr
# vrestore -xf /dev/ntape/tape0 -D /mnt/usr
vrestore: Date of the vdump save-set: Mon Sep 30 10:29:47 2002
vrestore: Save-set source directory : /usr
vrestore: informational: [13] posting event: sys.unix.fs.advfs.fset.backup.lock
If running in single user mode, EVM is not running.
Please ignore this posting.
vrestore: informational: [13] posting event: sys.unix.fs.advfs.fset.backup.unlock
If running in single user mode, EVM is not running.
Please ignore this posting.
· Mount usr fileset & restore usr filesystem
# mount -t advfs var_domain#var /mnt/var
# vrestore -xf /dev/ntape/tape0 -D /mnt/var
vrestore: Date of the vdump save-set: Mon Sep 30 10:43:47 2002
vrestore: Save-set source directory : /var
vrestore: informational: [13] posting event: sys.unix.fs.advfs.fset.backup.lock
If running in single user mode, EVM is not running.
Please ignore this posting.
vrestore: informational: [13] posting event: sys.unix.fs.advfs.fset.backup.unlock
If running in single user mode, EVM is not running.
Please ignore this posting.
· If necessary ensure /mnt/etc/inetd.local file has appropriate lan card definitions.
# cat /mnt/etc/inet.local
#!/bin/sh
#
# *****************************************************************
# * *
# * Copyright (c) Digital Equipment Corporation, 1991, 1999 *
# * *
# * All Rights Reserved. Unpublished rights reserved under *
# * the copyright laws of the United States. *
# * *
# * The software contained on this media is proprietary to *
# * and embodies the confidential technology of Digital *
# * Equipment Corporation. Possession, use, duplication or *
# * dissemination of the software and media is authorized only *
# * pursuant to a valid written license from Digital Equipment *
# * Corporation. *
# * *
# * RESTRICTED RIGHTS LEGEND Use, duplication, or disclosure *
# * by the U.S. Government is subject to restrictions as set *
# * forth in Subparagraph (c)(1)(ii) of DFARS 252.227-7013, *
# * or in FAR 52.227-19, as applicable. *
# * *
# *****************************************************************
#
# HISTORY
#
# @(#)$RCSfile: inet.local,v $ $Revision: 1.1.2.2 $ (DEC) $Date: 1995/12/04 22:36:21 $
#
# Repository for local network daemons and startup commands
#
# This script is called by /sbin/init.d/inet after all the
# network interfaces are up and running, but before any
# other network services have been started.
#
# Examples of things to put in here would be pfconfig,
# adding aliases for existing IP addresses, etc.
#
# echo "Starting local Internet services."
lan_config -i ee0 -a 0 -s 100 -x 1
· Remove details of all hardware & SCSI device databases
· Umount any remaining filesystems and halt system
# cd /
# umount -a
# halt
syncing disks... done
CPU 0: Halting... (transferring to monitor)
halted CPU 0
halt code = 5
HALT instruction executed
PC = ffffffff002c5f30
resetting all I/O buses
P00>>>
· This will return you to the firmware level (with the >>> prompt). You can now boot up from your newly recovered alpha system. However, please remember to boot to single-user mode only from the generic kernel, as there will be minor hardware differences between the new and old AlphaServer models. For example, the network card may be installed in a different PCI slot. This means that the kernel configuration file will need to be recreated to take any possible differences into account. Once the kernel configuration file has been recreated, a kernel rebuild will be required, along with a system reboot.
P00>>>boot -fi genvmunix -fl s dkc0
(boot dkc0.0.0.2.0 -file genvmunix -flags s)
block 0 of dkc0.0.0.2.0 is a valid boot block
reading 19 blocks from dkc0.0.0.2.0
bootstrap code read in
base = 200000, image_start = 0, image_bytes = 2600
initializing HWRPB at 2000
initializing page table at 3ff54000
initializing machine state
setting affinity to the primary CPU
jumping to bootstrap code
UNIX boot - Thursday August 24, 2000
Loading genvmunix ...
Loading at 0xffffffff00000000
Sizes:
text = 7671552
data = 1953968
bss = 2460480
Starting at 0xffffffff0000fbf0
set_pmap_memdsc_state: start 0x0 end 0x47c4 cl 0xffffffffffd2cd90
Alpha boot: available memory from 0x8f88000 to 0x1bffe8000
Compaq Tru64 UNIX P5.1 (Rev. 725); Wed Apr 10 15:41:45 EDT 2002
physical memory = 7168.00 megabytes.
available memory = 7023.69 megabytes.
using 27464 buffers containing 214.56 megabytes of memory
Master cpu at slot 0
Starting secondary cpu 1
Starting secondary cpu 2
Starting secondary cpu 3
Firmware revision: 6.1-3
PALcode: UNIX version 1.88-101
Compaq AlphaServer ES40
pci1 (primary bus:1) at nexus
itpsa0 at pci1 slot 1
IntraServer ROM Version V2.0 (c)1998
scsi0 at itpsa0 slot 0 rad 0
comet0: Card type 'Elsa GLoria' with 8MB framebuffer memory.
# bcheckrc
Checking device naming:
Passed.
Checking local filesystems
Mounting local filesystems
exec: /sbin/mount_advfs -F 0x14000 root_domain#root /
root_domain#root on / type advfs (rw)
/proc on /proc type procfs (rw)
exec: /sbin/mount_advfs -F 0x4000 usr_domain#usr /usr
usr_domain#usr on /usr type advfs (rw)
exec: /sbin/mount_advfs -F 0x4000 var_domain#var /var
var_domain#var on /var type advfs (rw)
· Create new kernel configuration file and rebuild kernel. If required, edit kernel configuration file to insert specific customizations.
# TERM=vt100
# export TERM
# cd /sys/conf
#
# doconfig
*** KERNEL CONFIGURATION AND BUILD PROCEDURE ***
Enter a name for the kernel configuration file. [ALPHA]:
A configuration file with the name 'ALPHA' already exists.
Do you want to replace it? (y/n) [n]: y
Saving /sys/conf/ALPHA as /sys/conf/ALPHA.bck
*** KERNEL OPTION SELECTION ***
Selection Kernel Option
--------------------------------------------------------------
1 System V Devices
2 NTP V3 Kernel Phase Lock Loop (NTP_TIME)
3 Kernel Breakpoint Debugger (KDEBUG)
4 Packetfilter driver (PACKETFILTER)
5 IP-in-IP Tunneling (IPTUNNEL)
6 IP Version 6 (IPV6)
7 Point-to-Point Protocol (PPP)
8 STREAMS pckt module (PCKT)
9 X/Open Transport Interface (XTISO, TIMOD, TIRDWR)
10 Digital Versatile Disk File System (DVDFS)
11 ISO 9660 Compact Disc File System (CDFS)
12 Audit Subsystem
13 Logical Storage Manager (LSM)
14 ATM UNI 3.0/3.1 ILMI (ATMILMI3X)
15 IP Switching over ATM (ATMIFMP)
16 LAN Emulation over ATM (LANE)
17 Classical IP over ATM (ATMIP)
--- MORE TO FOLLOW ---
Enter your choices or press <Return>
to display the next screen.
Choices (for example, 1 2 4-6):
18 ATM UNI 3.0/3.1 Signalling for SVCs (UNI3X)
19 Asynchronous Transfer Mode (ATM)
20 All of the above
21 None of the above
22 Help
23 Display all options again
--------------------------------------------------------------
Enter your choices.
Choices (for example, 1 2 4-6) [21]: 20
You selected the following kernel options:
System V Devices
NTP V3 Kernel Phase Lock Loop (NTP_TIME)
Kernel Breakpoint Debugger (KDEBUG)
Packetfilter driver (PACKETFILTER)
IP-in-IP Tunneling (IPTUNNEL)
IP Version 6 (IPV6)
Point-to-Point Protocol (PPP)
STREAMS pckt module (PCKT)
X/Open Transport Interface (XTISO, TIMOD, TIRDWR)
Digital Versatile Disk File System (DVDFS)
ISO 9660 Compact Disc File System (CDFS)
Audit Subsystem
Logical Storage Manager (LSM)
ATM UNI 3.0/3.1 ILMI (ATMILMI3X)
IP Switching over ATM (ATMIFMP)
LAN Emulation over ATM (LANE)
Classical IP over ATM (ATMIP)
ATM UNI 3.0/3.1 Signalling for SVCs (UNI3X)
Asynchronous Transfer Mode (ATM)
Is that correct? (y/n) [y]: y
Do you want to edit the configuration file? (y/n) [n]: n
*** PERFORMING KERNEL BUILD ***
A log file listing special device files is located in /dev/MAKEDEV.log
Working....Wed Oct 2 13:51:55 BST 2002
The new kernel is /sys/ALPHA/vmunix
# mv /sys/ALPHA/vmunix /vmunix
· Reboot system
# cd /
# umount -a
# sync
# sync
# sync
# halt
:
P00>>>
P00>>> set auto_action boot
P00>>> set bootos_flags a
P00>>> set bootdef_dev dkc0
P00>>> boot
The ALPHA system should boot to multi-user mode, with all OS services running. This can be verified by logging on to the system and running the df -k command to check /, /usr & /var filesystems are mounted.
· Restore data to apps & data filesystems from backup tapes.
· Reboot system to Multi-user mode
# shutdown -r now
The ALPHA system should now reboot to multi-user mode, with all OS and database services operational. This can be verified by logging on to the system and running whatever database applications are necessary.
The above advice is excellent for getting the O/S up and running, but I notice that your question also involves Sybase.
Sybase, like most database servers, has "hot" files (i.e. always open for writing) that generally don't backup & restore correctly using conventional O/S backups.
In fact, it is common for Sybase to use raw devices such as "/dev/rvol/group2/<volumename>" to store data rather than files on a mounted file system.
So yes, you do want to follow a procedure such as that listed above, but also be prepared to dump all of your databases to tape on a regular basis using a script that calls the sybase "dump database" command, and load the databases back in afterwards.
To make it easier to manage, you might consider using a third-party software tool such as Veritas or Legato to perform daily tape backups and restores of your O/S, and make sure it includes an add-on "agent" that directly queries your Sybase servers and dumps the databases to tape.
If you are unsure of how to back up and recover Sybase databases, I would definitely recommend involving your DBA's (if you have any) or bring in outside help to ensure that you are backing up everything required.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.