How do i use the rescue boot option

How do i use the rescue boot option

Post by Ingvar Tunel » Mon, 21 Jul 1997 04:00:00



Hi, i need to get access to my hd with its file to change a link so
that ld.so is found. I cant figure out how to access it from the
minimal system it sets up
I also wonder what rpm package that contains ldconfig since it seems to
be gone after i updated to ld.so2.0.4 with rpm

 
 
 

How do i use the rescue boot option

Post by James William » Tue, 22 Jul 1997 04:00:00


: Hi, i need to get access to my hd with its file to change a link so
: that ld.so is found. I cant figure out how to access it from the
: minimal system it sets up
: I also wonder what rpm package that contains ldconfig since it seems to
: be gone after i updated to ld.so2.0.4 with rpm

I agree with you that the Redhat rescue option doesn't make a lot of
sense.  It boots you into a shell where you don't have any tools.  I
couldn't even mount my hard drives!  What I recommend doing is one of
two things, depending on your level of expertise, and how much time
and effort you want to put into this.  The first option is to grab
the boot and root disk images from a Slackware ftp site, and use those
to get access to your hard drive.  They're not perfect, but if you
know what you're doing, they should allow you to get in and fix your
system.

The other option is to go in like you were going to upgrade your
Redhat install using your current version.  When it asks if you want
to do an install or upgrade, choose upgrade.  When it asks if you want
to customize your installation, choose yes.  From there, remove all
packages except the base packages (which I don't think are even listed).
Then let the upgrade continue.  It will restore your base operating
system.

For your second question, I'm not sure which package has ldconfig, but
you can use rpm to get a list of files in a package with:

rpm -qlp <package_name>

There's a way you can search for a specific file in all packages, but
I can't remember the syntax.

Good luck restoring your system.

--
           /   AIX     | James Williams

PA_RISC __\ \  Sco     |
  Sparc \ \\// OSF     |
PowerPC  \///  Linux   |
 RS6000  ///\  FreeBSD |
  Alpha //\\_\ Ultrix  |
   MIPS  \ \   Xenix   |
MC68000   \/   Solaris |
          /    HPUX    |

 
 
 

1. Odd Hex numbers after boot using Slackware 3.5 rescue.gz

First, a little background:
I've been running 1.3.20 from Slackware 3.0 on my 386 w/6MB RAM and got
DNS, sendmail, Apache 1.1.1, telnet, ftp, etc running OK.  I then
decided I wanted to use IPX and found that the kernel code was buggy and
that I'd have to upgrade.  Given that I was going to have to change
kernels, I decided to go to 2.0.34 (among other things, I wanted the
anti-teardrop patch).  I upgraded gcc to egcs-something and libc to a
recent version.  However, I was still getting the "Numeric constant with
no digits" msg when attempting to compile anything.  I read through
Dejanews and found references to upgrading ld.so.  Being a complete
moron, I removed the old ld.so (with pkgtool) before trying to install
the new one.  Those of you who aren't morons will realize that I
rendered the system more or less unusable (anything that spawned a new
process would fail to work correctly).

Now, the problem:
I figured that I'd need to make boot and root disks to work around
this.  OK, I download RAWRITE, net.i, and rescue.gz.  Good enough,
right?  I successfully made a net.i bootdisk (second try) and then tried
about ten diskettes for a rescue.gz root disk.  None of them worked; I
even checked w/a surface scan (Win95 scandskw) and most came up clean.
When I switch to the root disk, I get a "loading ramdisk" message and
then a message about finding the info it needs at sector 0 or whatnot.
This is followed by numerous numbers being written to my screen (ex:
[<001a0018>]; they are all bracketed like that) and scrolling by at an
extremely high rate.  Then either:
(a) the system just apparently hangs with many of those numbers still
onscreen
(b) it crashes with a message about a NULL kernel pointer and gives
debug info (eg register status, CS, DS, EIP, etc)
(c) it stops at a line like:
code 39 8a 00 01 00 00 75 1b c7 42 60 11 00 000 00 a1 7c 71 21 00

I have gotten similar results using color.gz in place of rescue.gz
If instead of switching disks as I should per directions, I insert a
DOS-formatted diskette with color.gz on it, I get a message about being
unable to find an image at 0 and am then prompted again for a root
disk.  If I leave the DOS disk in, I get an "unable to open virtual
console" message.  If I switch for one of the other root disks I tried
to make w/RAWRITE, I get UMSDOS messages followed by the usual VFS:
Kernel Panic stuff.

Any and all help appreciated greatly; please cc by email if
possible--this NNTP server can be flaky.

--
                                                --Sparty

web: [down]
* Ski Fast, Live Forever

2. boot up problem

3. Using a Dos boot disk to boot Linux

4. FreeBSD documentation

5. How to set DOS as the default boot option?

6. Multithread Prog Problem : RW Lock

7. Kernel uncorrectly forces cruft option when some options of mkisofs are used

8. is there an xnetload like xload type program?

9. Compaq Evo N800v has no option to boot from USB device option on its BIOS

10. Using options BRIDGE and options IPFIREWALL together?

11. How to copy files from Linux fs to DOS in rescue mode ?!!!

12. Help: can't boot dos from /dev/hdb1 using lilo

13. OS/2 and DOS, now want NETBSD on 2nd HD using boot manager