02-12-2013
Quote:
Originally Posted by
Praveen_218
When you edit grub.conf to add another boot option, you will have to re-build the stage-2 GRUB loader; which is why is a necessary step.
The GRUB stage-1 is just a 512 bytes system boot loader which is written at MBR (which is located in most PC based machine at track-0, sector-1 of their boot disk) and by the common disk architecture it can only be limited to 512 bytes (Till you actually get the newer 4KB sector disks).
GRUB stage-1 is what which actually let your machine to boot after POST (Power On Self Test done by BIOS) and subsequently loads the stage-2 which could be of different sizes (based on the total number of boot-menu displayed) and is actually responsible to rendered the boot menu options; Which you would need to create to get your new boot option menu displayed.
This needs to be built when you run 'grub-install'. This is specially true if you have GRUB version below GRUB-2 (under GRUB-2, there is a intermediate stage of stage1.5, loaded by stage-1, instead of stage-2 directly. Stahe-1.5 is basically enabler of basic file system reading capability to GRUB). Hence under GRUB-2 one can just modify the boot configuration and it reads the configuration to be rendered and displayed under stage -2 for you to boot your machine.
But there are issues:
1) If the disk file system is a custom one (i.e. not of common FS like UFS, ext2, ext3..4, reiserFS, etc.. or anything drives out of lineage) then the GRUB can't directly access the disk to access the boot.conf; hence won't give you the menu.
2) Incase, stage-1.5 is not present the still stage-2 gets loaded by stage-1 and in this case too the boot menu would not be rendered out of the latest configuration you saved into your grub.conf.
In both these cases (at least) you will have to perform the grub-install to create the grub images (both Stage-1 and Stage-2). It creates Stage-2 first and then the Stage-1 and inserts the starting sector number of the location of Stage-2 inside Stage-1 image. Hence Stage-1 can access and transfer control while boot strapping.
Hence in majority of the cases one should do the 'grub-install' except a few when your boot disk is formatted with most common Linux FS.
Hope that I answered your query.
Ok..Great Theory on the Grub Thing.
please guide me on the technical part...thanks
10 More Discussions You Might Find Interesting
1. UNIX for Dummies Questions & Answers
I have installed Linux 7.2 on a IBM xseries 235 server with 5 scsi hard drives and ServerRaid-5i Ultra320 scsi controller.
I started my installation by configuring the raid controller. I created 2 arrays, raid 1 and raid5. I then proceeded to install Linux and load the drivers.
After the... (2 Replies)
Discussion started by: cstovall
2 Replies
2. Red Hat
my redhat 9 will not boot. We had a power failure and when the power came back, my redhat linux will not boot.
The machine come up to grub prompt.
I tried the following from grub prompt
root (hd0, then press tab key
partition num:0 filesystem type unknown, partition type 0x83... (7 Replies)
Discussion started by: hassan2
7 Replies
3. UNIX for Dummies Questions & Answers
I am new to UNIX and need help in cloning a HPUX 10.2 Ace 5, can anybody please guide me in making a full system backup.
Real Chess (0 Replies)
Discussion started by: real-chess
0 Replies
4. Linux
Please help. I am new to linux. I wrote a script to run the backup on lunix machine but the job gave me an error. I am using Linux 2.6.14.3. Below is the sample of my script can anyone tell me where went wrong? Thanks in advance.
#!/bin/sh
dat=$(date +%d%b%y)... (5 Replies)
Discussion started by: clement
5 Replies
5. UNIX for Dummies Questions & Answers
hi,
Recently, I had receive one system. it's run on the unix tru64 server.
I looking some log files when i know don't work few months age system backup to tape.
Below error:
INFO:Tape backups to tape tape0 starting
for backup list:
slot2:/disk4
Backup Command Variable... (0 Replies)
Discussion started by: Tlg13team
0 Replies
6. Solaris
I have unix server with OS 5.8 ,,,
I tried ufsdump 0ua -f /dev/rmt/0 / to perform full system backup on tape but I failed could any one give a procedure for full system backup on solaris machine using tapes??? (1 Reply)
Discussion started by: mm00123
1 Replies
7. UNIX for Dummies Questions & Answers
I would like to know if I can do a full system back
up on my Unix Sco openserver 5.0.5 Machine. If so, What
is the syntax to do this or where can I find this
information at?
Also, is it possible to make this tape bootable so
that I can easily do a full system restore? Any
information on... (2 Replies)
Discussion started by: nellenodrog
2 Replies
8. HP-UX
Hello,
I'm still new in HP-UX backup and I want to do a FULL BACKUP of HP-UX server to TAPE device. After reading on several forums and posts, i have list down several steps on how to do a full backup on HP-UX with fbackup. I would like the gurus here to comment/advise on the steps below
1)... (4 Replies)
Discussion started by: miskin
4 Replies
9. UNIX for Advanced & Expert Users
Hi,
In order to have a sand box machine that I could use to test some system changes before going to production state, I'd like to duplicate a working system to a virtual one. Ideally, I'd like to manage to do it this way :
- Make a full system backup excluding the user file system (this... (7 Replies)
Discussion started by: pagaille
7 Replies
10. Solaris
Hello i am new on Solaris, and i need to migrate my old AIX 5.3 to Solaris 11.2
Now i have all apps working fine but i have the backup cause i was reading and i have not idea about what method must i choose.
Btw on AIX i had a mksysb backup to restore all system and obviously another backups to... (4 Replies)
Discussion started by: thorin666
4 Replies
LEARN ABOUT REDHAT
amrestore
AMRESTORE(8) System Manager's Manual AMRESTORE(8)
NAME
amrestore - extract backup images from an Amanda tape
SYNOPSIS
amrestore [ -r | -c | -C ] [ -b blocksize ] [ -p ] [ -h ] tapedevice | holdingfile [ hostname [ diskname [ datestamp [ hostname [ diskname
[ datestamp ... ]]]]]]
DESCRIPTION
Amrestore extracts backup images from the tape mounted on tapedevice or from the holding disk file holdingfile that match hostname,
diskname and datestamp patterns given on the command line. The tape or holding file must be in a format written by the amdump or amflush
program.
If diskname is not specified, all backups on the tape for the previous hostname are candidates. If datestamp is not specified, all backups
on the tape for the previous hostname and diskname are candidates. If no hostname, diskname or datestamp are specified, every backup on
the tape is a candidate.
Hostname and diskname are special expression descibe in the "HOST & DISK EXPRESSION" section of amanda(8). Datestamp are special expres-
sion describe in the "DATESTAMP EXPRESSION" section of amanda(8). For example, if diskname is "rz[23]a", it would match disks rz2a and
rz3a.
Datestamp is useful if amflush writes multiple backup runs to a single tape.
Unless -p is used, candidate backup images are extracted to files in the current directory named:
hostname.diskname.datestamp.dumplevel
OPTIONS
-b Set the blocksize used to read the tape or holding file. All holding files must be read with a blocksize of 32 KBytes. Amrestore
should normally be able to determine the blocksize for tapes on its own and not need this parameter.
The default is 32 KBytes.
-p Pipe output. The first matching backup image is sent to standard output, which is normally a pipe to restore or tar, then amrestore
quits. It may be run again to continue selecting backups to process. Make sure you specify the no-rewind tapedevice when doing
this.
Note: restore may report "short read" errors when reading from a pipe. Most versions of restore support a blocking factor option to
let you set the read block size, and you should set it to 2. See the example below.
-c Compress output using the fastest method the compression program provides. Amrestore normally writes output files in a format
understood by restore or tar, even if the backups on the tape are compressed. With the -c or -C option, amrestore writes all files
in compressed format, even if the backups on the tape are not compressed. Output file names will have a .Z or .gz extension depend-
ing on whether compress or gzip is the preferred compression program. This option is useful when the current directory disk is
small.
-C Compress output using the best method the compression program provides (may be very CPU intensive). See the notes above about the
-c option.
-r Raw output. Backup images are output exactly as they are on the tape, including the amdump headers. Output file names will have a
.RAW extension. This option is only useful for debugging and other strange circumstances.
-h Header output. The tape header block is output at the beginning of each file. This is like -r except -c or -C may also be used to
compress the result. Amrecover uses the header to determine the restore program to use.
If a header is written (-r or -h), only 32 KBytes are output regardless of the tape blocksize. This makes the resulting image usable as a
holding file.
EXAMPLES
The following does an interactive restore of disk rz3g from host seine, to restore particular files. Note the use of the b option to
restore, which causes it to read in units of two 512-byte blocks (1 Kbyte) at a time. This helps keep it from complaining about short
reads.
% amrestore -p /dev/nrmt9 seine rz3g | restore -ivbf 2 -
The next example extracts all backup images for host seine. This is the usual way to extract all data for a host after a disk crash.
% amrestore /dev/nrmt9 seine
If the backup datestamp in the above example is 19910125 and seine has level 0 backups of disks rz1a and rz1g on the tape, these files will
be created in the current directory:
seine.rz1a.19910125.0
seine.rz1g.19910125.0
You may also use amrestore to extract a backup image from a holding disk file that has not yet been flushed to tape:
% amrestore -p /amanda/20001119/seine.rz1a.2 | restore -ivbf 2 -
Amrestore may be used to generate a listing of images on a tape:
% mt -f /dev/nrmt9 rewind
% amrestore -p /dev/nrmt9 no-such-host > /dev/null
This asks amrestore to find images for host no-such-host. It will not find any entries that match, but along the way will report each
image it skips.
CAVEATS
GNU tar must be used to restore files from backup images created with the GNUTAR dumptype. Vendor tar programs sometimes fail to read GNU
tar images.
AUTHOR
James da Silva <jds@cs.umd.edu>
University of Maryland, College Park
SEE ALSO
amanda(8), amdump(8), amflush(8), tar(1) restore(8)
AMRESTORE(8)