[comp.virus] Boot Sector/Partition Table Protection

padgett%tccslr.dnet@uvs1.orl.mmc.com (Padgett Peterson) (02/05/91)

>>From:    gt1546c@prism.gatech.edu (Gatliff, William A.)

>>To help combat this, what would be the possibility of 'delibrately'
>>infecting ones boot-sector with a piece of code that would display
>>some kind of 'ok' message if it hadn't been tampered with?

Exactly what I was talking about in issue 17 except the "partition
table" sector (absolute sector one) should be used, not the boot
sector. More, such code can be used to prevent any tampering with
itself, the real partition table, or the active boot sector. At one
extreme I have tried on a system with C: & D: drives was to put all
executables on the C: drive and prohibit ANY writes or formats to that
drive (except with a special maintenance program). The D: drive just
has its low area protected and contains mutable programs and data.

A university or corporate environment might allow writing only to
floppies or bernoullis, protecting the hard disk. While such software
techniques alone cannot prevent an infected boot from occurring from a
floppy - only hardware can do this - they do allow such intrusion to
be detected prior to the load of the OS and can block any such
infection thereafter.

I hope that this will stimulate some activity on the part of the
vendors to provide such protection - it is not difficult to write, but
for me, I would no longer consider any product complete unless some
such form of low level protection was included.

					Padgett

ps: This is my hobby - you should see my job.

PHYS169@csc.canterbury.ac.nz (Mark Aitchison, U of Canty; Physics) (02/07/91)

padgett%tccslr.dnet@uvs1.orl.mmc.com (Padgett Peterson) writes:
>>>... what would be the possibility of 'delibrately' infecting ones boot-secto
r
with a piece of code ...
..
> allow such intrusion to be detected prior to the load of the OS and can block

> any such infection thereafter...

If anybody's interested, there is such a program avaliable, i.e. stops
hard disk boot viruses early in the start-up sequence. If anyone is
interested, I can e-mail further details. It's a companion product to
an automatic diskette boot sector scanner.

Mark Aitchison, Physics, University of Canterbury, New Zealand.

frisk@rhi.hi.is (Fridrik Skulason) (02/12/91)

Regarding the subject of automatically detecting infections by boot
sector viruses, I just wanted to point out that F-DRIVER.SYS (a part
of my F-PROT package) will detect all known boot sector viruses, and
is also designed to detect new/unknown boot sector and partition table
viruses.  I will, however include an option in version 1.15 to disable
this check, as it may cause problems on machines with network boot
ROMs.

- -frisk

71435.1777@CompuServe.COM (Bob Bosen) (02/28/91)

Referring to the idea of inserting viral detection code very early in
the bootstrap sequence by modifying the partition table, Padgett
Peterson writes:

>I hope that this will stimulate some activity on the part of the
>vendors to provide such protection -- it is not difficult to write,
>but for me, I would no longer consider any product complete unless
>some such form of low level protection was included.

I'm sorry, but it would just be too easy to fake the "all clear"
message generated by any such technique.

I agree that some form of low level protection is necessary but I fear
that defensive code hiding in partition tables will be much more
vulnerable to attack than MY preferred method: periodically
bootstrapping from a "sterile" boot diskette that is kept isolated from
every other usage. If I never use that boot diskette in any machine
executing any code that didn't COME from that diskette, then it CAN't
be corrupted. Period. End of discussion. That's the ultimate low-level
protection.

Bob Bosen
Enigma Logic Inc. (Producers of SafeWord VIRUS-Safe [Now Shareware])
2151 Salvio Street #301
Concord, CA  94520
USA

Tel: (415) 827-5707
FAX: (415) 827-2593
Internet: 715435.1777@COMPUSERVE.COM