Baslinux mailing list

[baslinux AT lists.ibiblio.org]

>I think we should not be discussing MBRs for so long on the BL3 list.
>My email address can be found (by people, not bots) at my website address
>(which I posted for downloading a kernel and a program recently).  Please
>write me directly.

Ok.

> I use EXT3 on Kurumin, a Debian-based distro (in Portuguese).
> I have no way to compare those FSs, since BL3 and Kurumin are very
different
> each other.
>You can copy Kurumin to a partition that was formatted ext2 and compare
>sizes.

>> 2kb is big enough for making a lot of things in Assembly or machine
>> language.
>> In the past I've programmed Burroughs  mini-computers and all I had were
512
>> "memory words", what means 512 bytes.

>I programmed a very small computer that took one card.  My partner made a
>4K DOS text editor in assembler language that works under dosemu.

If the HD is still recognized by BIOS, we can say it's a good start.
If everything goes definitely wrong you can still boot from a floppy or any
other media which points to the bootable partition. That media needs a
healthy MBR.
A friend has told me to check if the font has power enough to feed the PC
conveniently, and also to check the BIOS in order to remove a "reading
ahead" setting ( for some reason some HDs do not accept that setting ).

>>> So why won't LILO boot floppy work either?  Does the floppy disk version
>>> look at the MBR or other parts of track 0?
>>
>> I hope not.
>> But it certainly looks for the bootable partition, which is determined by
>> the user.

>I will try LILO again from floppy boot disk and maybe from MBR.

>> I believe it just looks for a bootable partition, since it can be hda1,
>> hda2, etc.
>> since Linux users are not obliged to install the system oon MBR.

>Is this the linux partition, or the partition where I put the kernel and
>boot.b?   I could put them in hda2 instead of hda1.

>>> Are they entirely in track 0?
>>
>> Well, 1kb is big enough to handle a lot of information.
>> These viruses alter the MBR table and point to other locations on HD
where
>> the malware really is.
>> This is to say that malware codes can be placed out of track 0, but since
>> the pointers are zeroed these files can't be executed anymore.

>The rest of the drive is also zeroed.

So, there's no danger of a remaining malware activity.
However, except for the master boot record, those viruses would not affect
Linux, but just DOS and Windows.

>> Teixeira

>Sindi