ts-7000
[Top] [All Lists]

Re: [ts-7000] Re: shell script from Kyaw Kyaw

To:
Subject: Re: [ts-7000] Re: shell script from Kyaw Kyaw
From:
Date: Mon, 26 Mar 2007 20:34:04 -0500
>Worse, or better, he might have replaced the whole of the busybox
>executable or only the softlink. /bin/sh is a hard- or soft-link to
>busybox and i cant remember what happens when you overwrite a hardlink.

In unix-like systems, a "hard link" is just another name for the same
file.  If you open the file by that name and write to it, the file is
modified.

Of course, the same happens if it is a symbolic link (what people
sometimes call "soft link", which is really a special file that contains
another file name.)  If you open a symbolic link, you actually open
the file named in the link.  Again, if you open and write, the file is
modified.

So it comes down to whether he used 
        cp my_script /bin/sh    (opens and writes to sh)
or
        mv my_script /bin/sh    (removes sh and renames my_script)

In my machine, /bin/sh was a symlink to busybox.  Assume his system has
the same filesystem that mine was delievered with:  If he used cp,
most of the system is hosed because the busybox executable was destroyed.
If he used mv, only the name /bin/sh was destroyed.

I note that /bin/ash is also a symlink to busybox.  If he used mv, then
it would work to boot the system with "init=/bin/ash".  I would try this
before anything more complicated.  If it works, you have enough access
to the system to fix everything, though you would need to mount the
root read/write before you re-create the symlink for /bin/sh.

>One sollution would be to use his USB stick with a debian distro on it
>right? Then he would have to reformat his USB stick (and risking
>destroying it if it's one of those special ones stores the firmware on
>disk.)

The kernel shipped with my 7260 was not able to use the USB stick for
a root filesystem without modification.  The USB disk device does not
exist until after the USB initialization completes, but that is shortly
AFTER the kernel tries to mount the root filesystem.  

The kernel mods to make it possible to use a USB stick for a root
are fairly simple to implement -- just if it fails to mount the root
filesystem, wait a few seconds and try again.  But I would not wish
the task of building a hacked kernel on somebody who is still trying to
learn shell programming.

>Another solution that I am about to try, that would take less effort to
>explain would be the following.
>* Put the default linux-distro of 8Mb on a http-server
>* load it into ram by executing "load -h <server IP> -r -b <memory address>"
>* "exec -r 0x0080_0000 -b 0x00218000 -c "console=ttyAM0,115200
>root=/dev/ram0"

Somebody else was looking into ram disks, and I think we determined that
an initrd must fit entirely within a _single_ _physical_ memory bank.
On at least some of the TS machines, there are 4 banks of 8 mb in a 32
mb machine.



 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/ts-7000/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/ts-7000/join
    (Yahoo! ID required)

<*> To change settings via email:
     
    

<*> To unsubscribe from this group, send an email to:
    

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/
 

<Prev in Thread] Current Thread [Next in Thread>
Admin

Disclaimer: Neither Andrew Taylor nor the University of NSW School of Computer and Engineering take any responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the birding-aus mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU