| Disaster
Recovery Procedures (ATNF - Narrabri) |
This document contains procedures that
are to be followed in the case of a disaster affecting computer/network
components crucial to the operation of the telescope (offline and online).
Routine administration procedures vital to maintaining a reliable system
such as configuring disks and maintaining well defined tape backups are
also outlined here.
A copy of this document is to be kept in the
system administrators room in the case where this information is inacessible
due to the web server being unavailable.
There are a number of possible outcomes
that are discussed below. In each the appropriate staff and affected users
are to be notified.
|
Screen Room destroyed |
 |
If this was ever to occur we would surely
be up the creek without a paddle. Critical systems in the system administration
domain are indicated below:
-
The following hardware devices currently exist
in the screen room
-
Assess the damage and set priorities as to
which hardware devices need immediate attention
-
Call the relevant hardware vendors to arrange
for replacement parts (i.e. if none are currently available)
-
In the case of kaputar place a service call
with COMPAQ (1300 788 990) quoting the serial number (AY45700625) and customer
number (3632500)
-
Contact relevant technical staff to organise
installation of communication, ethernet cables, power outlets and other
devices
-
Connect relevant hardware devices to the ethernet
(SCSI disks and tape drives)
-
Reinstall the relevant operating systems on
kaputar, ningadhun and leon
-
Restore all data on kaputar, ningadhun and
leon
-
Check that all the relevant services and applications
are successfully installed
|
Control Room Destroyed |
|
-
The following hardware devices currently exist
in the control room
-
noel
-
desk2
-
nemesis
-
vladimir
-
Kodak CD Rom Archiving Unit
-
Exabyte EXB8500 8mm tape drive unit
-
Call the relevant hardware vendors to arrange
for replacement parts/hardware
-
In the case of Digital hardware check the
DEC maintenance contract to see if the hardware is covered under the maintenance
agreement. Place a service call quoting the serial numbers of the damaged
hardware components.
-
Contact relevant technical staff to check
ethernet cables and power outlets.
-
Connect relevant hardware devices to the ethernet
(SCSI disks and tape drives)
-
Restore operating systems and data on the
relevant host machines
-
Ensure that the functionality of the replace
host machines has been restored
|
kaputar |
|
Current Specifications/Configuration
-
DEC Alpha Server 1000A 5/400
-
Runs a number of services: web, mail, offline
computing applications for observer data processing
-
640Mb RAM, ~54Gb disk space (see details on
RAID disks)
-
Attached devices:
-
DLT 7000 Tape drive unit
-
External 2Gb SCSI swap disk
-
Exabyte EXB-8700 Tape drive
-
Paralan SCSI extender
The following manuals are located in the system
administrators office. The documentation provided in these manuals can
be most useful during a disaster recovery:
-
Digital UNIX system Administration v4.0B+
-
DEC Unix Network Administration
-
Networker DEC UNIX v4.2A
-
UNIX System Administration Handbook, 2nd Ed.,
Nemeth,
Snyder, Seebass & Hein
IMPORTANT NOTE:
If kaputar requires a full emergency (cold) shutdown it is critical that
the external RAID disk rack is powered down AFTER the server is powered
down. The reverse order applies when cold booting. Otherwise the RAID set
may be toggle to an NOT READY/OFFLINE status which will prevent kaputar
from booting up next time. To rectify this one needs to get an ARC console
and run the RAID Array 200 Software for AXP Systems (see the section
on Configuring RAID groups below).
-
Determine which hardware components have been
damaged and which can be recovered
-
In the case of damaged disks (in the RAID
set), replace with new disks (can be hot swapped on the fly)
-
In the case of a other damaged components
such as the motherboard or powersupply, place a service call with COMPAQ
(DEC) providing details of the problem. Arrangements should be made to
replace the damaged part(s)
-
Once the hardware has been restored boot kaputar
and check the configuration settings at the console prompt. For more information
see AlphaServer 1000A Owner's Guide.
>>> show config
Digital Equipment Coorporation
Alpha Server 1000A 5/400
Firmware
SRM Console: V5.0-101
ARC Console: v5.34
PAL Code: VMS PALCode V1.19-4, OSF PALcode V1.21-6
Serial Rom: v1.4
Processor
DECchip (tm) 21164A-2 400Mhz
Memory
640 Meg of System Memory
Bank 0 = 256 Mbytes (64MB Per SIM) Starting at 0x00000000
Bank 1 = 256 Mbytes (64MB Per SIM) Starting at 0x10000000
Bank 2 = 128 Mbytes (32MB Per SIM) Starting at 0x20000000
Bank 3 = No memory detected
Slot Option Hose 0, Bus 0, PCI
7 Intel 82375 Bridge to Bus 1, EISA
8 DECchip 21050-AA Bridge to Bus 2, PCI
11 DECchip 21050-AA Bridge to Bus 3, PCI
13 Mylex DAC960 dra.0.0.13.0
dra.0.0.13.0 2 Member RAID 1
dra.1.0.13.0 4 Member RAID 5
dra.2.0.13.0 4 Member RAID 5
dra.3.0.13.0 3 Member RAID 5
0 QLogic ISP 1020 pka0.7.0.2000.0 SCSI Bus ID 7
dka.400.4.0.2000 RRD45
dka.500.5.0.2000 RRD43
dka.600.6.0.2000 QUANTUM DLT7000
Slot Option Hose 0, Bus 0, PCI
0 DECchip 21050-AA ewa0.0.3000.0 00-00-F8-20-6B-04
1 QLogic ISP1020 pkb0.7.3001.0 SCSI Bus ID 7
dkb0.0.0.0.3001 RRD43
dkb100.1.0.3001 RRD43
mkb400.4.0.3001.0
mkb401.4.0.3001.0
mkb402.4.0.3001.0
mkb403.4.0.3001.0
mkb404.4.0.3001.0
mkb405.4.0.3001.0
mkb406.4.0.3001.0
mkb407.4.0.3001.0
bus 2, slot 0 -- pka -- QLogic ISP 1020
bus 3, slot 0 -- ewa -- DECchip 21040-AA
bus 3, slot 1 -- pkb -- QLogic ISP 1020
bus 0, slot 13 -- dra -- Mylex DAC960
Restore the operating system and check the
integrity of the filesystems. The steps are briefly outlined below:
-
At the console prompt boot into single user
mode using the command
boot -fl s
and check the integrity of the root filesystem,
i.e use the fsck command.
Restore the root filesystem using a tape containing
a full backup labelled vdump. Such a backup tape would have been produced
in single user mode.
IMPORTANT: When logging into kaputar in single
user mode, the root-file system is mounted as read-only. Use the following
command to change it to a read/write filesystem:
mount -u /
Mount the filesystem containing the networker
server (backup/restore) i.e. /usr/bin
Restore all filesystems from the last full
backup and all subsequent incremental backups until the date of the crash
Check the integrity of the restored filesystems
using the fsck command.
|
kaputar's RAID disks |
|
Specifications on kaputar's RAID disks
can be obtained by running the StorageWorks RAID Array 200 Management
Utility v1.1.1 (/usr/bin/swxcrmgr).
Specifications on RAID disks (on kaputar)
Channel Vendor Model Rev: Size (Mb)
----------------------------------------------------------------------------
A-0 DEC RZ29B 0016 4091
A-1 DEC RZ29B 0014 4091
B-0 DEC RZ29B 0014 4091
B-1 DEC RZ29B 0016 4091
B-2 DEC RZ29B 0016 4091
B-3 DEC RZ29B 0016 4091
C-0 Quantum XP34300W L915 4101
C-1 Quantum XP34300W L915 4101
C-2 Quantum XP34300W L915 4101
C-3 Quantum XP34300W L915 4101
D-0 DEC RZ1DB-CA LYJ0 8678
D-1 DEC RZ1DF-CB 0372 8678
D-2 DEC RZ1DB-CA LYJ0 8678
D-3 DEC RZ1DB-CA LYJ0 8678
----------------------------------------------------------------------------
Logical Drive Table (RAID array)
-----------------------------------------------------------------------
Drive RAID Size Status
Group Level (Mb)
-----------------------------------------------------------------------
A 1 4091 Optimal
B 5 12273 Optimal
C 5 12303 Optimal
D 5 26034 Optimal
-----------------------------------------------------------------------
RAID Levels and redundancy
-
RAID Level 1: the amount of storage data available
for unique data is 50% (i.e. one disk is a mirror image of the other).
-
RAID Level 5: the amount of storage data available
for unique data is 75% (i.e. for a set of 4 disks one of them is redundant).
-
Only one disk is permitted to fail without
the drive group completely failing.
-
Upon failure one may add an identical disk
to replace the faulty one on the fly. One would then need to rebuild the
drive group using the StorageWorks RAID Array 200 Management Utility
v1.1.1 (/usr/bin/swxcrmgr).
Configuring RAID groups
-
In order to increase the disk capacity of
a current RAID group one needs to do the following:
-
Shutdown kaputar
-
Remove the RAID set and replace with new disks
-
At the halt prompt start an ARC console by
typing arc.
-
Insert the floppy RAID Array 200 Software
for AXP Systems and type A:swxcrmgr
-
Ensure that the new RAID set can be detected
and then proceed to format the drive group (see the RAID Array 200 User's
Guide)
-
Boot kaputar into multi-user mode and run
disklabel in order to label the new RAID set
-
Create the appropriate file domains and filesets
and then restore data onto the new disks.
|
ningadhun |
|
Current Specifications/Configuration
-
DELL PowerEdge 2300 Server: Pentium III-450
-
256Mb ECC SDRAM
-
18Gb (7200rpm) internal Western Digital Ultra-2/LVD
SCSI disk
-
Video Card: Integrated ATI-Rage Pro 3D, 2Mb
SDRAM, 1024x768x256 resolution
-
Manuals (located in the screen room next to
Monitor):
-
DELL PowerEdge 2300 - Installation and TroubleShooting
Guide
-
DELL PowerEdge 2300 - User's Guide
-
DELL OpenManage Server Assistant v4.1.1 (useful
tool for setting up diagnostic diskettes)
-
Start of 3 year lease: June 1999
-
S/N: TK2Q7
Determine and identify damaged components.
If the server has been completely destroyed (e.g. a fire) then goto section
1 otherwise proceed to section 2. A special case is section 3 where one
is expecting a certain failure and has time to move the data to another
server.
-
Establishing a new PC server
-
ningadhun is partially damaged
-
Promoting a BDC to a PDC
-
Establishing a new PC server
This section is relevant for the scenario
where the data on ningadun is not accessible due to a major hardware failure.
-
ningadhun is partially damaged
This section refers to some component
of ningadhun (DELL PowerEdge 2300) being damaged to such an extent that
the continuation of normal user services is prevented, thus requiring urgent
attention..
-
Faulty motherboard or dead power supply
-
Place a service call with DELL Customer Service
quoting the serial number and the problem. A job number will be given and
should be noted for reference when chasing up the status of a call.A serviceman
will be scheduled to come out to site with the relevant replacement parts.
-
System Disk crash
-
If the internal (18Gb) SCSI disk has been
damaged for some reason which is preventing the system from booting up
a complete recovery may be required.
-
If the disk cannot be detected as indicated
from the bootup console output (with SCSI ID set to 0), then a service
call is most likely required in order to arrange for a replacement disk.
Screen Output:
Adaptec AIC-7890 SCSI BIOS v2.01 Dell 001
(c) 1998 Adaptec, Inc. All Rights Reserved.
Press for SCSISelect(TM) Utility
SCSI ID: 0 WDIGTL WDE18300 ULTRA2 ULTRA2-LVD - Hard Disk 0
SCSI ID: 6 DELL 1x6 U2W SCSI BP ULTRA2-LVD
SCSI BIOS Installed Successfully !
Adaptec AIC-7860 SCSI BIOS v2.01 Dell 001
(c) 1998 Adaptec, Inc. All Rights Reserved.
Press for SCSISelect(TM) Utility
SCSI ID: 5 NEC CD-ROM DRIVE:465
The second device indicated is the SCSI BackPlane
board allowing up to 5 additional SCSI disks to be mounted into the front
disk rack.
-
Re-installing Windows NT4 onto the DELL
PowerEdge 2300
-
One cannot install NT Server unless the SCSI
controller card is installed correctly.
-
Boot the server with the Windows NT Server
4 CD.
-
During the initial boot phase (i.e when blue
screen appears) press F6 to interupt and specify other SCSI devices to
be installed.
-
Select: Adaptec AHA-294xU2/295xU2/AIC-789x
PCI Ultra2 SCSI Controller(NT 4.0)
-
The SCSI driver is located on a DELL floppy
disk labelled NT 4 Video and SCSI controller drivers.
-
Other options/considerations:
-
Install as a PDC (Primary Domain Controller)
-
Network drivers are for: Intel Pro 100+
-
Licencing Mode - per seat
-
Video drivers: ATI Technologies Inc. 3D RAGE
PRO
-
Install service pack 5 (PCAPPS CD in system
administrators office). Do not replace the file EB100.SYS during installation.
-
Promoting a BDC to a PDC
This procedure is relevant if one suspects
an imminent failure of ningadhun and there is still time to move the data
to another server.
-
BDC - Backup Domain Controller
-
PDC - Primary Domain Controller
-
Make arrangements for a backup server which
satisfies the minimum requirements (see section 1).
-
Boot off the Windows NT Server 4 CD.
-
Install as a BDC.
-
Licensing Mode - per seat.
-
Enter all relevant network information. Use
an hostname and IP address not currently used. This will be switched to
the correct name and IP address at the end of this section.
-
Install service pack 5 from the PCAPPS CD
in the system administrator's office.
-
Locate the scopy utility with the /o /s options.
This utility is available on the Windows NT Server Resource Kit
CD located in the system administrator's office. This utility is particularly
useful in copying security information which is not possible with the standard
copy command in Windows NT.
-
Use the scopy utility to copy the following
directories (and sub-directories) in order of importance:
-
D:\profiles
-
D:\users
-
D:\pcapps
-
D:\dist
-
D:\Cad
-
Synchronise the BDC with the PDC, i.e. goto
Start/Administrative
Tools/ Server Manager/Computer/Synchronise Entire Domain.
-
Promote the BDC to a PDC - in the same menu
as the previous step goto Promote to Primary Domain Controller.
Note that this will also automatically demote the current PDC to a BDC.
-
Shutdown (faulty) ningadhun.
-
Rename the new PDC to ningadhun and update
the IP address in the TCP/IP properties section.
-
Reboot the new ningadhun
-
Check user logins and other relevant tests
as in step 1.
|
noel |
|
A number of outcomes are discussed below.
-
System disk damaged (VMSSYS0)
-
Find a replacement SCSI disk and ensure that
a non-conflicting SCSI address have been selected
-
Make sure that all disks have been properly
dismounted
-
Shutdown the VMS cluster and attach the replacement
disk.
-
Attach the MKA600 drive and check that both
the replacement disk and tape drive have been successfully detected.
-
Restore the files from the tape.
-
Restore the VMS Cluster by booting the VMS
members in the correct sequence.
-
Record the event in the observer's logbook.
-
Data disk damaged ($DISK3)
-
Find a replacement SCSI disk and ensure that
a non-conflicting SCSI address have been selected
-
Shutdown the VMS cluster and attach the replacement
disk.
-
Attach the MKA600 drive and check that both
the replacement disk and tape drive have been successfully detected.
-
Locate and restore files from the last full
backup including all subsequent incremental backups to the date of the
disk crash.
-
Reboot the cluster and check for a successful
login.
-
Record the event in the observer's logbook.
-
Faulty monitor(s)
-
If a fuse has blown in either one of monitors
try and find an appropriate fuse and replace accordingly
-
If the monitor cannot be repaired easily,
contact COMPAQ (Dec) quoting the serial number of the faulty monitor (from
the Service contract).
-
Locate and install a temporary replacement
monitor from another VMS cluster member.
-
Does not power up
-
First check to see if the fuse of the power
supply has been blown. Get a qualified electronics technician to quickly
check but ensure that they do not attempt a more complex operation since
it would violate warranty conditions. If the fuse has been blown ensure
that the fuse has been correctly chosen !!
-
If there is still no power, place a service
call with COMPAQ Customer Services quoting the serial number and other
details if required. The responce is 24 hours. If this is unreasonable
another VMS cluster member such as leon would have to be sacrificed temporarily.
|
leon |
|
A number of outcomes are discussed below.
-
System disk damaged (VMSSYS1)
-
Find a replacement SCSI disk and ensure that
a non-conflicting SCSI address have been selected
-
Make sure that all disks have been properly
dismounted
-
Shutdown the VMS cluster and attach the replacement
disk.
-
Attach the MKA600 drive and check that both
the replacement disk and tape drive have been successfully detected.
-
Restore the files from the most recent full
backup and subsequent incremental backups.
-
Restore the VMS Cluster by booting the VMS
members in the correct sequence.
-
Record the event in the observer's logbook.
-
Data disk damaged ($DISK0)
-
Find a replacement SCSI disk and ensure that
a non-conflicting SCSI address have been selected
-
Shutdown the VMS cluster and attach the replacement
disk.
-
Attach the MKA600 drive and check that both
the replacement disk and tape drive have been successfully detected.
-
Locate and restore files from the last full
backup including all subsequent incremental backups to the date of the
disk crash.
-
Reboot the cluster and check for a successful
login.
-
Record the event in the observer's logbook.
-
Faulty VT320 terminal
Simply replace with a spare
terminal.
-
Does not power up
-
First check to see if the fuse of the power
supply has been blown. Get a qualified electronics technician to quickly
check but ensure that they do not attempt a more complex operation since
it would violate warranty conditions. If the fuse has been blown ensure
that the fuse has been correctly chosen !!
-
If there is still no power, place a service
call with COMPAQ Customer Services quoting the serial number and other
details if required. The responce is 24 hours. If this is unreasonable
another VMS cluster member such as delphi would have to be sacrificed temporarily.
|
System Backup |
|
Unix and WinNT machines
-
Tape backups are performed daily using a DLT
7000 Tape Drive connected to kaputar.
-
Networker v4.4 (/usr/bin/networker) is the
software used to drive the backups. For more information see the manual
Networker
DEC Unix v4.2A in the system administrator's office.
-
The process daemons for the Networker server
on kaputar can be stopped and restarted via /sbin/init.d/nsrd.
-
The backup for kaputar commenses daily at
23:30
-
A snapshot of the current filesystem disk
usage (Nov-1999) is shown in the following table
-------------------------------------------------------------------------
Filesystem Total Used Capacity
(Mb) (Mb) Mount point
-------------------------------------------------------------------------
/dev/re0a 129 105 91% /
/dev/re0g 705 464 73% /usr
/dev/re0h 3114 2662 95% /x
/dev/rz5c 629 529 84% /syscdrom1
kaputar#opt 12567 453 88% /opt
kaputar#usrlocal 12567 668 91% /usr/local
kaputar#atapplic 12567 1172 95% /atapplic
kaputar#applic 12567 491 89% /applic
kaputar#AIPS 12567 1011 94% /AIPS
kaputar#narusers 12567 2510 98% /narusers
kaputar#source 12567 892 93% /source
kaputar#SOLARIS2local 12567 1629 96% /export/SOLARIS2local
kaputar#SOLARIS2opt 12567 1782 97% /export/SOLARIS2opt
kaputar#aips++ 12567 64 51% /aips++
kaputar#ATOMS 12567 1411 96% /ATOMS
kaputar#www 12567 138 69% /www
data#visitors 39257 14591 64% /data/KAPUTAR_1
data#students 39257 565 6% /data/KAPUTAR_2
data#localdata 39257 15666 65% /data/KAPUTAR_3
-------------------------------------------------------------------------
The backup for ningadhun commenses daily at
02:00
-
The following directories are backed up
| Drive |
Capacity (Gb) |
Description |
| C: |
2.5 |
Operating System, System files, Registry |
| D: |
14.5 |
User Data files, User profiles |
-
A full backup is performed every 14 days.
Level 9 incremental backups are performed
every other day.
-
The naming convention for DLT tape backups
is CULGOORA_###, where ### denotes a 3 digit number
-
All DLT tapes are currently stored in the
computer room. The tape just before the currently loaded tape (in the DLT
tape drive) is to be stored offsite (at the system admin's home). This
is to ensure that:
-
The latest (mounted) tape is on site to provide
quick recovery for the latest files. This however is at the expence of
possibly losing approximately one month's backup of data (assuming that
the screeb room is completely destroyed) since currently mounted tapes
are not duplicated.
-
The offsite tape offsite provides security
and data integrity. The probability of both the computer room and the offsite
location being destroyed is extremely small.
Contents of the file /etc/fstab.
The following details are shown from left
to right respectively: Name of filesystem, mount point on kaputar, type
of filesystem and read/write details. All the filesystems indicated below
are backed up (except for swap1, swap2, /syscdrom0, /syscdrom1,
/data/KAPUTAR_2
and /data/KAPUTAR_3).
=====================================================================================
Filesystem 1024-blocks Used Available Capacity Mounted on
=====================================================================================
/dev/re0b swap1 ufs sw 0 2
/dev/rz1b swap2 ufs sw 0 2
/dev/re0a / ufs rw 1 1
/dev/re0g /usr ufs rw 1 2
/dev/re0h /x ufs rw 1 2
/proc /proc procfs rw 0 0
/dev/rz4c /syscdrom0 ufs ro 0 0
/dev/rz5c /syscdrom1 ufs ro 0 0
#/dev/fd0c /fd ufs rw 0 0
kaputar#opt /opt advfs rw,userquota,groupquota 0 2
kaputar#usrlocal /usr/local advfs rw,userquota,groupquota 0 2
kaputar#atapplic /atapplic advfs rw,userquota,groupquota 0 2
kaputar#applic /applic advfs rw,userquota,groupquota 0 2
kaputar#AIPS /AIPS advfs rw,userquota,groupquota 0 2
kaputar#narusers /narusers advfs rw,userquota,groupquota 0 2
kaputar#source /source advfs rw,userquota,groupquota 0 2
kaputar#SOLARIS2local /export/SOLARIS2local advfs rw,userquota,groupquota 0 2
kaputar#SOLARIS2opt /export/SOLARIS2opt advfs rw,userquota,groupquota 0 2
kaputar#aips++ /aips++ advfs rw,userquota,groupquota 0 2
kaputar#ATOMS /ATOMS advfs rw,userquota,groupquota 0 2
kaputar#www /www advfs rw,userquota,groupquota 0 2
data#visitors /data/KAPUTAR_1 advfs rw,userquota,groupquota 0 2
data#students /data/KAPUTAR_2 advfs rw,userquota,groupquota 0 2
data#localdata /data/KAPUTAR_3 advfs rw,userquota,groupquota 0 2
#ningadhun:/T/users /nt/users nfs rw,nfsv2
#ningadhun:/T/cad /nt/cad nfs rw,nfsv2
#ningadhun:/S/pcapps /nt/apps nfs ro,nfsv2
#%aips2.nrao.edu:/export/aips++/master /aips++_master nfs rw,grpid,hard,intr,retrans=20,tim
eo=60
=====================================================================================
|
VMS Cluster |
|
-
The strategy followed at present involves
a full backup of the cluster every 8 weeks, with an incremental backup
being performed on all other weeks. A full backup takes approximately 2
days to complete, whilst an incremental may take rom half a day to one
day and a half, depending on the proximity of a full backup. The current
trend is to start a backup on Thursday morning, so that if it takes a second
day to complete it will not run into the weekend. See
VMS
Cluster Backup
Current status of cluster mounted devices:
$ show dev/m
Device Device Error Volume Free Trans Mnt
Name Status Count Label Blocks Count Cnt
$1$DKA0: (NOEL) Mounted 0 VMSSYS0 793821 5 4
$1$DKA100: (NOEL) Mounted 0 $DISK3 106104 58 4
$2$DKA0: (LEON) Mounted 0 VMSSYS1 639485 514 4
$2$DKA100: (LEON) Mounted 0 $DISK0 222690 26 4
$2$DKA400: (LEON) Mounted 0 PAGEDISK 9699 2 1
$12$DKA0: (DELPHI) Mounted wrtlck 0 VAXDOCMAR951 369165 1 4
$12$DKA200: (DELPHI) Mounted 0 $DISK1 1267092 7 4
$12$DKA300: (DELPHI) Mounted 0 DELPHI_1037 47295 1 4
$15$DKA0: (DESK2) Mounted 0 KOALA 1344708 1 4
$15$DKA300: (DESK2) Mounted 0 DESK2_1255 41412 1 4
Device Device Error
Name Status Count
LTA0: Offline mounted 0
RTA1: Mounted 0
RTA2: Mounted 0
Shutting down the VMS Cluster
The VMS cluster need to be shutdown in
the following order.
To re-establish the cluster boot the machines
in the reverse order.
|
Fileservers and critical workstations |
|
| Host name |
Description |
Operating System |
Location |
Purpose |
Serial Number |
| kaputar |
DEC Alpha 1000A 5/400 |
Digital Unix v 4.0B |
Computer Room |
UNIX, Web, Mail server,etc... |
AY45700625 |
| ningadhun |
DELL PowerEdge 2300 |
WinNT Server v4.0 SP5 |
Computer Room |
PC server - user data/applications |
TK2Q7 |
| leon |
MicroVax 3100-80 |
Open VMS v 6.1 |
Computer Room |
VAX server |
KA224R6060 |
| noel |
Vaxstation 4000-60 |
Open VMS v 6.1 |
Control Room |
Compact Array control computer |
AB22202U7T |
| atria |
Digital PC |
Red Hat Linux v5.2 |
Correlator Room |
Correlator data acquisition computer |
- |
|
SUN workstations |
|
All the currently used SUN workstations
are listed below. In the case of hardware problems one may place a service
call with SUN Service Centre quoting the serial number and describing the
problem to the customer service representative. The serial numbers of the
last two workstations are not indicate since they are no longer covered
under service warranty.
| Host name |
Description |
Op. System |
Location |
Purpose |
Serial Number |
| achilles |
SUN Ultra 10 |
Solaris 2.5.6 |
Computer Room |
Solaris Server/Workstation |
HW82004508 |
| medea |
SUN Ultra 10 |
Solaris 2.5.6 |
Computer Room |
SUN workstation |
HW82004506 |
| ambrosia |
SUN Ultra 10 |
Solaris 2.5.6 |
Computer Room |
SUN workstation |
HW82004503 |
| orpheus |
SUN Ultra 10 |
Solaris 2.5.6 |
Control Room |
Online imaging |
FW84750472 |
| poseidon |
SUN Ultra 10 |
Solaris 2.5.6 |
Mark Wieringa's office |
SUN workstation |
FW84750488 |
| argos |
SUN Ultra 10 |
Solaris 2.5.6 |
Dave McConnell's office |
SUN workstation |
FW84750478 |
| molen |
SparcStation 5 |
Solaris 2.5.6 |
Observer's Area |
SUN workstation |
425F5931 |
| corvus |
Sparc ULTRA 10 |
Solaris 2.5.6 |
Observer's Area |
SUN workstation |
FW93230158 |
| vladimir |
SparcStation 5 |
SUN OS v4.1.3 |
Control Room |
SUN workstation |
- |
|
Contact Names |
|
| Company Name |
Phone |
Fax |
Contact Name(s) and details |
| Arrow Direct P/L |
(03) 9763 8433 |
(03) 9763 8823 |
Geoff Bull or John Taylor |
| ComNet Solutions |
(02) 9899 5700 |
(02) 9634 1432 |
Brian Denley |
| Connections P/L |
(02) 9552 3088 |
(02) 9552 3258 |
David Simmons |
| COMPAQ Customer Serivce |
1300 368 369 |
- |
- |
| COMPAQ Serivce Calls |
1300 788 990 |
- |
Customer ID: 363250 |
| DELL Sales Representative |
1800 803 385 |
1800 818 341 |
Abe Khamis or Kevin Keheo |
| DELL Credit Manager |
(02) 9930 3355 |
- |
Kylee Mace |
| DELL Technical Support |
1800 808 378 |
- |
- |
| DELL Delivery Enquiries |
1800 819 339 |
- |
- |
| Epson - Technical Support |
(02) 9903-9040 |
- |
- |
| HP Direct |
131 047 |
- |
- |
| Hunter Digital |
(02) 4968 4455 |
- |
Marty Wilson |
| Software Spectrum |
(02) 9418 3811 |
- |
Michael van Zoggel |
| Sun Microsystems Aust. P/L |
(02) 9466 9466 |
(02) 9466 9410 |
Robert Drake |
| Sun Service Centre |
1800 555 786 |
- |
- |
|
Service Contract |
|
| Location |
Licenses Box file (System
Admin's Office) |
| Account Number |
0195340 |
|
| Agreement Number |
5908200600B |
| Administrator |
Paul Cruz |
Created: John Giovannis (6-Aug-1998)
Modified: John Giovannis (9-Nov-1999)