pozar@hoptoad.uucp (Tim Pozar) (07/17/87)
Volume 4, Number 26 13 July 1987
+---------------------------------------------------------------+
| _ |
| / \ |
| /|oo \ |
| - FidoNews - (_| /_) |
| _`@/_ \ _ |
| International | | \ \\ |
| FidoNet Association | (*) | \ )) |
| Newsletter ______ |__U__| / \// |
| / FIDO \ _//|| _\ / |
| (________) (_/(_|(____/ |
| (jm) |
+---------------------------------------------------------------+
Editor in Chief: Thom Henderson
Chief Procrastinator Emeritus: Tom Jennings
FidoNews is published weekly by the International FidoNet
Association as its official newsletter. You are encouraged to
submit articles for publication in FidoNews. Article submission
standards are contained in the file ARTSPEC.DOC, available from
node 1:1/1.
Copyright 1987 by the International FidoNet Association. All
rights reserved. Duplication and/or distribution permitted for
noncommercial purposes only. For use in other circumstances,
please contact IFNA.
Five Weeks to FidoCon!
Table of Contents
1. ARTICLES ................................................. 1
Hotline! (1) ............................................. 1
2400 Baud At The Right Price ............................. 4
Talking ASSEMBLER (Number #3) ............................ 5
THE DIRTY DOZEN -- An Uploaded Program Alert List ........ 11
Running 2 BBS's, or CHOOSEr.exe to have FUN! ............. 20
International Vietnam Veterans EchoConference Anniversa .. 23
2. COLUMNS .................................................. 25
Borland's Turbo C: Review, part 2 ........................ 25
From the Broadcasters Booth - routing .................... 29
The Regular Irregular Column ............................. 32
3. WANTED ................................................... 37
INFORMATION "FEEDS" WANTED ............................... 37
4. NOTICES .................................................. 38
The Interrupt Stack ...................................... 38
Latest Software Versions ................................. 38
International FidoNet Conference Registration Form ....... 39
IFNA Board of Directors Ballot ........................... 40
FidoNews 4-26 Page 1 13 Jul 1987
=================================================================
ARTICLES
=================================================================
James Zachary
Fido 115/537
Hotline! (1)
(c) 1987
James Zachary
On any given day, at any given moment, the phone at the water and
sewage department can ring with a crisis call.
Southeast Treatment Plant, this is Zack.
"Ahem, err, why are ... uhhh are you adding ... uhhhmmm, why are
you putting ammonia in our drinking water ...?"
Pardon me?
"Uhhh, in my drinking water ... in my fish-tank ..."
You drink the water from your fish-tank?
"Uhhhh, errr, nooooo, I uh ..."
Sir, we have a terrible connection, sounds like you're talking
into a garbage can ...
"Hold on a sec ... THERE! IS THAT BETTER? I WAS ON MY NEW
SPEAKER PHONE!"
What else are you on?
"'SCUSE ME?"
Never mind. Don't shout, I can hear ya fine now. You said your
fish-tank tastes funny?
"Uhhhh, nooooo. All my fish died so I tested the water and it
has ammonia in it."
What's your point?
"The water in my fish-tank came from our faucet ... it's regular
drinking water from you. Your ammonia killed my fish!"
We don't add ammonia to our water. Some of the large systems do,
to form chloramines so they can carry a chlorine residual for
great distances, but we don't do that.
"Now wait a minute! I tested the water, both in the fish tank
and from the faucet and it has ammonia in it!"
I see. How much ammonia?.
FidoNews 4-26 Page 2 13 Jul 1987
"Five."
Five ...? Five what? Five parts per million, parts per billion,
parts per trillion...?
"Uhhhhh, it just says five."
What kind of equipment are you using?
"Well, I dunno but it cost me PLENTY! I spent $12 on it to find
out YOU killed my fish! It has test tubes and a color chart! I
went to college, you know!"
Uh huh, I'm sure your mother is proud. Look, friend, let me
assure you our lab, as well as the EPA lab, cost a tad more than
$12. Both labs are certified for technique and accuracy and
their results show the drinking water has barely enough ammonia
to measure.
"You mean I wasted $12?"
Looks that way.
"But my fish are all dead! The tank even smells like ammonia!"
How big was the tank and how many fish were in it?
"It was a 10 gallon tank and I had 50 black mollies in it."
Wonderful. You had 50 fish in a 10 gallon tank?
"Sure! The book that came with the tank said ..."
Whoa! Listen, ammonia may have played a part in bumping your
fish off but the ammonia came from their own waste.
"Their own waste? I don't understand!"
Waste ... excrement ... in college terms, your fish made wee wee
in the water...
"Oooooh......."
... and they made big poo poo ...
"Ahhhh, but my filter removes all that!"
Right. When was the last time you cleaned your filter?
"Why, NEVER! This filter turns the waste into air by rotifer
reaction so it never needs cleaning. It worked fine for a
month!"
Sir, have you ever considered changing hobbies to something other
than tropical fish?
FidoNews 4-26 Page 3 13 Jul 1987
"Well, I USED to raise tropical plants until YOUR water killed
them!"
How often did you water them?
"At least four times a day ..."
Maybe you'd consider raising hydroponic pet rocks.
"C L I C K ! ! "
Sometimes it doesn't pay to be helpful ...
-----------------------------------------------------------------
FidoNews 4-26 Page 4 13 Jul 1987
Jean Coppola 107/201
2400 Baud At The Right Price
Recently I was looking for a new 2400 baud modem and decided to
look around a bit. After seeing the prices on Hayes and Courier,
I knew I had to go with a clone if I wanted 2400 baud at this
time!
After searching around I found the MAXUM 2400 baud modem at a
very reasonable $199.95 from a rather large discount store in the
area. (47th Street Photo)
So, I plunked down the green and took this little joy home to see
if I had been ripped off or not. To my surprise, it turned out to
be a rather good purchase!
It claimed to be Hayes compatible, but many of us have seen that
before, and I for one was a little wary. But I was wrong to
worry! I opened the box, plugged it in, and set Opus for Hayes
2400 and away we went. Both Opus and SEAdog recognized it as a
Hayes 2400, and all the command sequences built into both pieces
of software work fine with it!
I have tested it now with both normal communications and file
transfers with all the popular protocols, and have not found any
increase in errors over 1200 baud transfers. But of course, it is
much quicker, especially when you use the SEAlink transfer
protocol and even faster when using some of the newer protocols.
In all fairness, there was one problem that I could not solve by
normal methods. For some unexplained reason, no matter what
command sequence I fed it, I could not shut off the speaker, or
lower the volume. Unlike most other external modems, there is no
outside volume control. So a quick call to the dealer who was as
stumped as I was.
Being handy with tools, the speaker is now shut off!
All in all, a very good purchase for someone looking for an
inexpensive 2400 baud external modem.
-----------------------------------------------------------------
FidoNews 4-26 Page 5 13 Jul 1987
TALKING ASSEMBLER 3
by Ned Sturzer
Chattanooga, Tennessee
OpusNODE 362/1
The machine instructions are the soul of Assembler and we will
examine them in great detail. Each instruction may have two
operands, one operand, or even no operands. Most instructions
refer to memory locations, registers, port addresses, or
immediate data. The manner in which these locations are referred
to are known as addressing modes. Some instructions allow for
implied operands and the operand is not indicated. For example,
the instruction
cbw
takes the high-order bit in the AL register and places the same
value in all the bits of AH. In effect, cbw extends the sign of
AL to AH. In this instruction the operands are in the implied
addressing mode.
Immediate data is allowed in the operands of many instructions
as in
int 9 or in mov pointer,417H
where pointer is a label for a memory address. In direct memory
addressing the operand is in the form
segment:offset
an example being
test di,ss:4ee7H
Based operands refer to the memory location computed by adding a
displacement, disp, to the value in DS:BX, CS:BX, SS:BX, or
ES:BX. You may also use DS:BP, CS:BP, SS:BP, or ES:BP. DS:BX may
be written simply as BX, while SS:BP may be expressed as BP. If
the displacement resolves to 0 it is not written. Among valid
ways of writing based operands are
disp.[BX]
ES:[BP]+disp
[disp][BX]
Now consider the instruction xchg [bx-3],di where bx contains the
value a06b. Our instruction will exchange the contents of the di
register with whatever value is in the two bytes of memory
ds:a069 and ds:a06a. If the displacement is 0 it need not be
indicated.
Indexed operands are treated similarly to based operands except
the si and di registers are used. With either si or di the
default segment register is ds. An exception to this is that if
di is used in string instructions such as stos or lodsw then the
FidoNews 4-26 Page 6 13 Jul 1987
default segment register used with di is es. In this case es
cannot be overriden.
Based indexed operands use a displacement in conjunction with one
register from either bp or bx and one register from either si or
di. As you would expect by now the default segment register if bp
used is ss and otherwise the default segment register is ds. Also
if based indexed operands are used in string instructions then di
forces the segment register to be es. As an illustration consider
the instruction
cmp byte ptr es:[bp+6][di],0feH
where es:bp has the value b800:0147 and ds:di has the value
0000:0200. This instruction then compares the contents of the
memory location b800:034d with fe.
If based indexed operands seem complicated - they are. However,
they allow you to do some fancy footwork through memory. Before
moving on to examine the individual instructions I need to tie up
a loose end. Whenever register IP is used to refer to a memory
location the reference is always to location cs:ip. Similarly if
register SP is used to refer to memory the location ss:sp is
intended.
I'll group the machine op codes according to their function,
indicate the flags affected after the operation is performed,
give the nature and size of the operands, and give examples.
I. Logical Operators
The logical operators AND, NOT, OR, and XOR (exclusive or)
perform the bit-by-bit logical operations their names imply. The
result is stored in the first operand. They are used to gain
access to the individual bits of a byte or word.
A. AND ax/al,immed
Flags:C=0, P, A=?, Z, S, O=0
B. AND r/m,immed
8 or 16 bits
Flags:C=0, P, A=?, Z, S, O=0
C. AND r,r
8 or 16 bits
Flags:C=0, P, A=?, Z, S, O=0
D. AND m,r
8 or 16 bits
Flags:C=0, P, A=?, Z, S, O=0
E. AND r,m
8 or 16 bits
Flags:C=0, P, A=?, Z, S, O=0
To explain the shorthand I'll use: In the operand area r is any
register other than a segment register, m is any memory location,
and immed stands for immediate data. Should a specific register
be used it will be indicated as in form A. If the size of the
operands (8 or 16 bits) is not implied by the form of the
FidoNews 4-26 Page 7 13 Jul 1987
instruction I have indicated it. The flags are affected as
indicated where =? means the flag may be changed arbitrarily.
Consider AND al,01101000B where ax contains 01000011B. After the
instruction is performed al will contain 01000000B, C=0, P=0,
Z=0, S=0, O=0, and A is arbitrary. Form A is distinguished from
form B since it requires one less byte in memory to store the
instruction. AND is used to clear specific bits to 0 while
keeping the other bits intact.
F. NOT r/m
8 or 16 bits
Flags: none
NOT simply interchanges all the bits of the operand. Thus in NOT
word ptr [bp+di] if bp contains 0050, di contains 0017, ss
contains 0040, and 0040:0067 to 0040:0068 contain the word a51c,
then after the operation this word will contain 5ae3. Note that
it is here necessary to tell the Assembler that we are
considering a word since there is nothing in the instruction
which otherwise indicates the length of the operand.
G. OR ax/al,immed
Flags:C=0, P, A=?, Z, S, O=0
H. OR r/m,immed
8 or 16 bits
Flags:C=0, P, A=?, Z, S, O=0
I. OR r,r
8 or 16 bits
Flags:C=0, P, A=?, Z, S, O=0
J. OR m,r
8 or 16 bits
Flags:C=0, P, A=?, Z, S, O=0
K. OR r,m
8 or 16 bits
Flags:C=0, P, A=?, Z, S, O=0
L. XOR ax/al,immed
Flags:C=0, P, A=?, Z, S, O=0
M. XOR r/m,immed
8 or 16 bits
Flags:C=0, P, A=?, Z, S, O=0
N. XOR r,r
8 or 16 bits
Flags:C=0, P, A=?, Z, S, O=0
O. XOR m,r
8 or 16 bits
Flags:C=0, P, A=?, Z, S, O=0
P. XOR r,m
8 or 16 bits
Flags:C=0, P, A=?, Z, S, O=0
The OR instruction allows you to set specific bits to 1 while the
other bits are unchanged. XOR is used to reverse specific bits
keeping the others as they were. Also the XOR instruction
provides an efficient way of setting a register or memory
location to 0. For example XOR bx,bx sets bx to 0.
FidoNews 4-26 Page 8 13 Jul 1987
II. Other Bit Manipulators
The TEST instruction and the various shift and rotate commands
are akin to to logical operators in that they perform bit
manipulation.
A. TEST ax/al,immed
Flags:C=0, P, A=?, Z, S, O=0
B. TEST r/m,immed
8 or 16 bits
Flags:C=0, P, A=?, Z, S, O=0
C. TEST r,r
8 or 16 bits
Flags:C=0, P, A=?, Z, S, O=0
D. TEST m,r
8 or 16 bits
Flags:C=0, P, A=?, Z, S, O=0
E. TEST r,m
8 or 16 bits
Flags:C=0, P, A=?, Z, S, O=0
TEST is similar to AND in that it performs an AND but does not
alter either operand. Only the flags are adjusted. TEST may be
used to determine the bit settings of specific bits.
F. SHL r/m,1
8 or 16 bits
Flags:C, O
G. SHL r/m,cl
8 or 16 bits
Flags:C, O
The SHL command shifts the contents of the first operand either
one bit left or by the number of bits indicated in the cl
register. The low order bits vacated by this operation are
replaced by zeros. Should cl contain a value greater than or
equal to the length of the first operand then the operand will
become 0. As long as no ones bits are shifted out then SHL
multiplies by powers of two. For example,
mov cl,3
shl byte ptr ds:0143
multiplies the contents of the byte at ds:0143 by 8 if ds:0143
originally has zeros in its high order three bits. Be aware that
the Macro Assembler, MASM, accepts the mnemonic SAL which
performs the same function as SHL. Oddly, SYMDEB and DEBUG do not
recognize SAL in the -a command. The convention used by the 8088
chip is that if an integer is regarded as signed then the high
order bit is the sign bit. A 1 in the sign bit means a negative
number and a 0 means a positive number. With this in mind we can
understand the rules used to set the flag bits in shift and
rotate operations. The carry flag is set to the last bit pushed
out of the end of the operand. The overflow flag is undefined if
the second operand is a cl and cl has a value other than 1. if
the second operand is 1 or if it is cl with a value of 1, then
FidoNews 4-26 Page 9 13 Jul 1987
the overflow flag is set to 1 if the sign bit changes. Otherwise
the overflow flag is cleared to 0.
H. SHR r/m,1
8 or 16 bits
Flags:C, O
I. SHR r/m,cl
8 or 16 bits
Flags:C, O
This is a shift right with the vacated bits replaced by zeros.
SHR acts as an integer divide by powers of two as long as no ones
bits are shifted out.
J. SAR r/m,1
8 or 16 bits
Flags:C, P, A=?, Z, S, O
K. SAR r/m,cl
8 or 16 bits
Flags:C, P, A=?, Z, S, O
This is an arithmetic shift right instruction which functions
similarly to SHR. The difference, besides the effect on the
flags, is that the vacated bits are filled by the sign bit. Thus
the result of the sequence
mov al,8f
mov cl,3
sar al,cl
is that al will contain f1.
L. ROL r/m,1
8 or 16 bits
Flags:C, O
M. ROL r/m,cl
8 or 16 bits
Flags:C, O
This rotate left instruction acts like the SHL command except
that the value in the high order bit of the first operand
replaces the value vacated in bit 0. Therefore if dh contains 7e
and cl contains 2 then ROL dh,cl will leave dh with f9.
N. ROR r/m,1
8 or 16 bits
Flags:C, O
O. ROR r/m,cl
8 or 16 bits
Flags:C, O
The rotate right instruction, ROR, is the analog of ROL except
that the value in bit 0 replaces the value vacated in the high
order bit of the first operand.
P. RCR r/m,1
FidoNews 4-26 Page 10 13 Jul 1987
8 or 16 bits
Flags:C, O
Q. RCR r/m,cl
8 or 16 bits
Flags:C, O
Rotate right through carry, RCR, is similar to ROR but here the
value in bit 0 is placed in the carry flag and the value in the
carry flag is placed in the high order bit of the first operand.
R. RCL r/m,1
8 or 16 bits
Flags:C, O
S. RCL r/m,cl
8 or 16 bits
Flags:C, O
RCL is the left rotating version of RCR.
-----------------------------------------------------------------
IFNA, FIDONEWS, FIDO AND OPUS SYSOPS ARE GRANTED A NON-EXCLUSIVE
LICENSE TO COPY, PROCESS, AND DISTRIBUTE THIS ARTICLE IN ANY
MATTER THEY SEE FIT. COMMERCIAL RIGHTS GRANTED AT NO COST UPON
WRITTEN REQUEST TO THE AUTHOR. REQUESTS MY BE SENT BY FIDO MAIL
TO 362/1, CHATTANOOGA, TENNESSEE OR BY LOGGING ON AT
(615) 892-7773.
-----------------------------------------------------------------
FidoNews 4-26 Page 11 13 Jul 1987
THE DIRTY DOZEN -- An Uploaded Program Alert List
Issue #7 Compiled by Eric Newhouse
Recently, many unlawfully copied or modified programs have
appeared on various IBM PC bulletin boards across the country.
THE DIRTY DOZEN is a list of known examples.
There are four major categories of bad software: commercial
pirate jobs, unauthorized copies of otherwise legitimate freeware
programs, malicious "TROJAN" programs which damage your system,
and miscellaneous illegal software. Please look in the
definitions section of this document for a more detailed
explanation of these terms.
SysOps: Please be careful with the files you post in your
download libraries! An professional quality uploaded game or
disk utility should arouse your suspicions, especially if it
doesn't include the author's name, address, and distribution
policy. Such programs are probably NOT public domain! The BBS
community is already under legislative threat at the State and
Federal level. We cannot fight this trend effectively while our
directories sit stocked with cracked Sega games, wargames
dialers, and malicious "trojan horses!" Let's demonstrate a
little social responsibility by cleaning up our download
libraries.
If you as a SysOp have any of these files on your system, please
delete them and post "blocking" dummy file entries like this one:
ZAXXON.COM DELETED!! NOT PUBLIC DOMAIN!!
If everyone works together to fight this new brand of software,
the growing numbers of piraters and trojan horse writers may well
be put 'out of business!'
The idea behind THE DIRTY DOZEN is to bring this important issue
to the attention of more SysOps and users - to act as an
information "clearing -house" for the latest known examples of
"bogusware," so that an educated public can fight effectively for
safe downloadable files.
The Dirty Dozen is a big project, and it needs your help to
succeed! Please call in any updates of bad software that you
know of, but DO NOT modify this article yourself. If everyone
who discovers a pirated program starts modifying the DD, there
would be hundreds of issues in circulation.
Also, I think it's quite unfair, especially considering that
I've spent over a hundred hours of my time on this list, for just
anyone to put their name at the top of the list and say that they
write, or helped write, the DD. For example, someone named
Gerhard Barth added two files, both of which were already listed
in the DD, and proceeded to write "Updated by Gerhard Barth,
FidoNews 4-26 Page 12 13 Jul 1987
please send all further updates to Gerhard Barth," etc. If
everyone does this, how will anyone know which file is the latest
and TRUE Dirty Dozen? If you have an update, please see the end
of this article for information on how to reach me with new
information.
A word on TROJANS: I have been hearing more and more reports of
these "worm" programs, from all directions. While I don't doubt
their existence, do not get hysterical. Remember, a Trojan rumor
is much easier to START than it is to STOP. Some people have
accused legitimate *joke* programs, like DRAIN (which pretends to
be gurgling excess water out of your A drive) of being "killers."
If a program locks up your system, it isn't necessarily Trojan;
it might not like co-residing with Superkey, or your graphics
card. Ask around a little before you announce something as
Trojan. I would appreciate a bagged specimen of any real Trojan
program that you might have the (un)luck to find.
A word on Pirated programs: Recently many pirated programs such
as AUTODEX have been going under many different names. Although
I will try to keep all these names current in the DD, the best
way to check for piracy in a file is to run that file yourself --
checking for (C)opyright notices of commercial manufactures,
similarities in looks and operations of commercial programs, and
of course whether the name is in this list.
Finally I want to thank all BBS SysOps and users that notified
me of updates, additions, and/or corrections to DIRTYDOZ.006.
It's great to see so much support! In this issue more people
than ever called in with updates. Everyone else who reads this
list, along with myself, really appreciates the effort!
NOTE: If I do not supply a file extension, that means that the
file circulates under many different extensions. For instance,
users commonly upload with extensions of either: .EXE, .COM,
.EQE, .CQM, .LBR, .LQR, and .ARC.
TROJAN HORSE PROGRAMS:
Name Category Notes
ANTI-PCB *TROJAN* The story behind this trojan horse is
sickening. Apparently one RBBS-PC
sysop and one PC-BOARD sysop started
feuding about which BBS system is
better, and in the end the PC-BOARD
sysop wrote a trojan and uploaded it to
the rbbs SysOp under ANTI-PCB.COM. Of
course the RBBS-PC SysOp ran it, and
that led to quite a few accusations and
a big mess in general. Let's grow up!
Every SysOp has the right to run the
type of BBS that they please, and the
FidoNews 4-26 Page 13 13 Jul 1987
fact that a SysOp actually wrote a
trojan intended for another simply
blows my mind.
ARC513.EXE *TROJAN* This hacked version of arc appears
normal, so beware! It will write over
track 0 of your [hard] disk upon usage,
destroying the disk.
ARC514.COM *TROJAN* This is totally similar to arc version
5.13 in that it will overwrite track 0
(FAT Table) of your hard disk. Also, I
have yet to see an .EXE version of this
program..
BACKTALK *TROJAN* This program used to be a good PD
utility, but some one changed it to be
trojan. Now this program will
write/destroy sectors on your [hard]
disk drive. Use this with caution if
you acquire it, because it's more than
likely that you got a bad copy.
CDIR.COM *TROJAN* This program is supposed to give you a
color directory of files on disk, but
it in fact will scramble your disks FAT
table.
DANCERS.BAS *TROJAN* This trojan shows some animated dancers
in color, and then proceeds to wipe out
your [hard] disk's FAT table. There is
another perfectly good copy of
DANCERS.BAS on BBS's around the
country; apparently the idiot author in
question altered a legitimate program
to do his dirty work.
DISKSCAN.EXE *TROJAN* This was a PC-MAGAZINE program to scan
a (hard) disk for bad sectors, but then
a joker edited it to WRITE bad sectors.
Also look for this under other names
such as SCANBAD.EXE and BADDISK.EXE...
DMASTER *TROJAN* This is yet another FAT scrambler..
DOSKNOWS.EXE *TROJAN* I'm still tracking this one down --
apparently someone wrote a FAT killer
and renamed it DOSKNOWS.EXE, so it
would be confused with the real,
harmless DOSKNOWS system-status
utility. All I know for sure is that
the REAL DOSKNOWS.EXE is 5376 bytes
long. If you see something called
DOSKNOWS that isn't close to that size,
sound the alarm. More info on this one
is welcomed -- a bagged specimen
FidoNews 4-26 Page 14 13 Jul 1987
especially.
DPROTECT *TROJAN* Apparently someone tampered with the
original, legitimate version of
DPROTECT and turned it into a FAT table
eater.
DROID.EXE *TROJAN* This trojan appears under the guise of
a game. You are supposedly an
architect that controls futuristic
droids in search of relics. In fact,
PC-Board sysops, if they run this
program from C:\PCBOARD, will find that
it copies C:\PCBOARD\PCBOARD.DAT to
C:\PCBOARD\HELP\HLPX. In case you were
wondering, the file size of the .EXE
file is 54,272 bytes.
EGABTR *TROJAN* BEWARE! Description says something like
"improve your EGA display," but when
run it deletes everything in sight and
prints "Arf! Arf! Got you!"
EMMCACHE *CAREFUL* This program is not exactly a trojan,
but it may have the capability of
destroying hard disks by:
A) Scrambling every file modified
after running the program,
B) Destroying boot sectors.
This program has damaged at least two
hard disks, yet there is a base of
happily registered users. Therefore, I
advise extreme caution if you decide to
use this program.
FILER.EXE *TROJAN* One SysOp complained a while ago that
this program wiped out his 20 Megabyte
HD. I'm not so sure that he was
correct and/or telling the truth any
more. I have personally tested an
excellent file manager also named
FILER.EXE, and it worked perfectly.
Also, many other SysOp's have written
to tell me that they have like me used
a FILER.EXE with no problems. If you
get a program named FILER.EXE, it is
probably alright, but better to test it
first using some security measures.
FINANCE4.ARC *CAREFUL* This program is not a verified trojan;
there is simply a file going around
BBS's warning that it may be trojan.
In any case, execute extreme care with
it.
FidoNews 4-26 Page 15 13 Jul 1987
FUTURE.BAS *TROJAN* This "program" starts out with a very
nice color picture (of what I don't
know) and then proceeds to tell you
that you should be using your computer
for better things than games and
graphics. After making that point it
trashes your A: drive, B:, C:, D:, and
so on until it has erased all drives.
It does not go after the FAT alone, but
it also erases all of your data. As
far as I know, however, it erases only
one sub-directory tree level deep, thus
hard disk users should only be
seriously affected if they are in the
"root" directory. I'm not sure about
this on either, though.
NOTROJ.COM *TROJAN* This "program" is the most
sophisticated trojan horse that I've
seen to date. All outward appearances
indicate that the program is a useful
utility used to FIGHT other trojan
horses. Actually, it is a time bomb
that erases any hard disk FAT table
that IT can find, and at the same time
it warns: "another program is
attempting a format, can't abort!"
After erasing the FAT(s), NOTROJ then
proceeds to start a low level format.
One extra thing to note: NOTROJ only
damages FULL hard drives; if a hard
disk is under 50% filled, this program
won't touch it! If you are interested
in reading a thorough report on
NOTROJ.COM, James H. Coombes has
written an excellent text file on the
matter named NOTROJ.TXT. If you have
trouble finding it, you can get it from
my board.
TIRED *TROJAN* Another scramble the FAT trojan by Dorn
W.Stickle.
TSRMAP *TROJAN* This program does what it's supposed to
do: give a map outlining the location
(in RAM) of all TSR programs, but it
also erases the boot sector of drive
"C:".
PACKDIR *TROJAN* This utility is supposed to "pack"
(sort and optimize) the files on a
[hard] disk, but apparently it
scrambles FAT tables.
PCW271xx.ARC *TROJAN* A modified version of the popular PC-
WRITE word processor (v. 2.71) has now
FidoNews 4-26 Page 16 13 Jul 1987
scrambled at least 10 FAT tables that I
know of. If you want to download
version 2.71 of PC-WRITE be very
careful! The bogus version can be
identified by its size; it uses 98,274
bytes whereas the good version uses
98,644. For reference, version 2.7 of
PC-WRITE occupies 98,242 bytes.
QUIKREF *TROJAN* This ARChive claims that it will load
RBBS-PC's message file into memory 2
times faster than normal. What it
really does is copy RBBS-PC.DEF into an
ASCII file named HISCORES.DAT...
RCKVIDEO *TROJAN* This is another trojan that does what
it's supposed to do, then wipes out
hard disks. After showing some simple
animation of a rock star ("Madonna," I
think), the program will go to work on
erasing every file it can lay it's
hands on. After about a minute of
this, it will create 3 ascii files that
say "You are stupid to download a video
about rock stars," or something of the
like.
SECRET.BAS *TROJAN* BEWARE!! This may be posted with a note
saying it doesn't seem to work, and
would someone please try it; when you
do, it formats your disks.
SIDEWAYS.COM *TROJAN* Be careful with this trojan; there is a
perfectly legitimate version of
SIDEWAYS.EXE circulating. Both the
trojan and the good SIDEWAYS advertise
that they can print sideways, but
SIDEWAYS.COM will trash a [hard] disk's
boot sector instead. The trojan .COM
file is about 3 KB, whereas the
legitimate .EXE file is about 30 KB
large.
STAR.EXE *TROJAN* Beware RBBS-PC SysOps! This file puts
some stars on the screen while copying
RBBS-PC.DEF to another name that can be
downloaded later!
STRIPES.EXE *TROJAN* Similar to STAR.EXE, this one draws an
American flag (nice touch), while it's
busy copying your RBBS-PC.DEF to
another file (STRIPES.BQS) so Bozo can
log in later, download STRIPES.BQS, and
steal all your passwords. Nice, huh!
FidoNews 4-26 Page 17 13 Jul 1987
TOPDOS *TROJAN* This is a simple high level [hard] disk
formatter.
VDIR.COM *TROJAN* This is a disk killer that Jerry
Pournelle wrote about in BYTE Magazine.
I have never seen it, although a
responsible friend of mine has.
This is the end of the "bad files list." The rest of this
document contains instructions on what to do if YOU run a trojan
horse, an update history, a glossary, and information on how and
where to contact me with updates.
If you run a trojan horse..
While reading this, bear in mind that there is no better remedy
for a drive that has run a trojan horse than a recent backup..
The first thing to do after running what you think to be a
trojan horse is diagnose the damage. Was your [hard] drive
formatted? Did the trojan scramble your FAT table? Did every
file get erased? Did your boot sector on the [hard] drive get
erased/formatted? Odds are that the trojan incurred one of these
four disasters.. After the initial diagnosis, you are ready to
remedy the problem.
1) If the trojan low-level formatted your [hard] disk: Hope that
you have a recent backup; that's the only remedy for this
disease.
2) If the trojan high-level formatted your [hard] disk: There is
only one way out of this mess, and that is to use the MACE+
utilities by Paul Mace. MACE+ has two devices in it to
recover formatted disks, and believe me, they work! I will
talk more about the MACE+ utilities later.
3) If the trojan scrambled your FAT table: Once again, there is
nothing to do. However, there is a program called
FATBACK.COM (available on my board named as FATBCK11.ARC)
that will back up your FAT table in under a minute to floppy.
Using FATBACK, it is easy and non time consuming to back up
your FAT regularly.
4) If the trojan erased file(s), and the FAT table is undamaged:
There are many packages to undelete deleted files. Norton
Utilities, PC-tools, MACE+, and UNDEL.COM will all do the
job. I recommend the first three, but they are more
expensive than the Public Domain program UNDEL.COM. When you
are undeleting, be sure to undelete files in the order of
last time written to disk. I know that PC-tools
automatically lists undeletable files in the correct order,
but the other three may not.
5) If the boot sector on your [hard] disk gets erased/formatted:
There are four things to do if this happens, and the worst
FidoNews 4-26 Page 18 13 Jul 1987
that can happen is that you will go without a [hard] disk for
a while. To be on the safest side, back up everything before
even proceeding to step "A," although I can not see why it
would be necessary.
A) Try doing a "SYS C:" (or "SYS A:") from your original DOS
disk, and copy COMMAND.COM back onto the [hard] drive
after that. Try booting and if that doesn't work try step
B.
B) If you have the MACE+ utilities go to the "other
utilities" section and "restore boot sector." This should
do the job if you have been using MACE+ correctly.
C) If you are still stuck, BACK EVERYTHING UP and proceed to
do a low level format. Instructions on how to perform a
low-level format should come with your [hard] disk
controller card. Be sure to map out bad sectors using
either SCAV.COM by Chris Dunford or by manually entering
the locations of bad sectors into the low level format
program. After the low level format, if your have a hard
disk, run FDISK.COM (it comes with DOS) and create a DOS
partition. Refer to your DOS manual for help in using
FDISK. Then put your original DOS diskette in drive A:
and do a FORMAT <drive letter>:/S/V. Drive letter can
stand for "C" or "B" depending on whether you are
reformatting a hard disk or not. Finally you are ready to
attempt a reboot.
D) If you are still stuck, either employ some professional
computer repairmen to fix your drive, or live with a non-
bootable [hard] drive..
By now you may be saying to yourself:
"How can I get a hold of a 'MACE+' utilities package so that I
can guard against trojans? Why, MACE+ can recover a formatted
drive, undelete files, restore boot sectors, optimize a disk, and
provide a disk cache!
Anyone can obtain these marvelous utilities in one of two ways:
one is to call up the Paul Mace Software Company (tm) and order
them at a retail of $ 79.95. The other is place an order for
them at the WEST LOS ANGELES PC-STORE, which supports next day
UPS shipping! The BBS phone # for the PC-STORE is at the end of
this document.
Finally:
If you have any additions or corrections for this list, send them
to Eric Newhouse at any of the following places: (in order of
most frequented)
FidoNews 4-26 Page 19 13 Jul 1987
* The Crest RBBS (213-471-2518) (1200/2400) (80 MB)
* The West LA PC-STORE (213-559-6954)(300/1200/2400)
-----------------------------------------------------------------
FidoNews 4-26 Page 20 13 Jul 1987
Are you tired of the same old thing, are you catering to the same
old users and tired of maintaining the same old BBS? Well have
you ever thought of putting up a whole new BBS, cater to a new
group of users, but haven't wanted to lose the established crowd
you have worked so hard for? Well have I go a deal for you!
Why not leave your old BBS as it is and put up a second one and
use the same phone line for both? This was a problem presented
to me when one of the local syslops had to give up an adult board
when his parents found it.
The idea of two BBS's isn't new and I don't lay claim to it. The
OTHER Side was originally a second BBS that was invoked with the
'O' command.
This worked okay but had some definite problems. First when the
second BBS was called up it loaded one on top of the other and
used all available memory (DoubleDOS was out of the question).
It also had the tendency to screw up the on line time if you
switched back and forth.
I approached Randy Bush with the problem and after some gentle
reminding he built a little program called CHOOSER. What
CHOOSER does is sit between SeaDOG or Dutchie and the BBS.
SeaDOG answers the phone and then passes control to CHOOSER,
CHOOSER then asks the caller what he wants to do. Currently
there are 10 possibilities, depending on a reply of 0,1,,,,,9
CHOOSER passes the error level to a small batch file that then
invokes your regular BBS command line.
This has enabled me to effectively run two BBS's off of one phone
line. SeaDOG now calls up OSWEGO OPERA (Chooser) that then asks
you if you want OSWEGO OPUS (1) or OSWEGO's OTHER Side (2).
Think of the possibilities, you can run your regular BBS (OSWEGO
OPUS) and beta test another software package (FIDO V12) or maybe
run the same BBS but with two different bbs packages (Opus and
Fido, same files areas and message areas) or possibly become
demented like I have and run the same software but each side has
two different themes. (the OTHER Side is a sexually oriented BBS)
I have two sperate user.bbs and am able to keep OSWEGO OPUS at
the same level it has always been (whatever that is).
CHOOSER.ARC is file requestable from DawGone Disgusted 105/6. It
doesn't like X00 as a fossil driver and really requires the most
current version of OPUSCOMM.
I'm including my runbbs.bat and also the complete DOCs file for
CHOOSER.
so you can see just how simple the setup is.
; Chooser 0.04: Select a function via a FOSSIL driver
;
; Copyright 1987, Pacific Systems Group. All rights
FidoNews 4-26 Page 21 13 Jul 1987
; reserved.
;
; This program may be used by any member of IFNA, Inc. without
; further obligation to the author. Members of IFNA, Inc. may
; give it to other members as long as no fee is charged in any
; way.
;
; For use or distribution by or to others (e.g. non-m,embers of
; IFNA, Inc.), you must contact the author:
; Randy Bush
; Fido 105/6
; 9501 SW Westhaven Drive
; Portland, Oregon US-97225
; But, it will probably be easier to just join IFNA, Inc.
;
;
; The command line MUST look like
;
; CHOOSER <port> <prompt>
;
; where
;
; <port> is 1, 2, 3, or 4
; <prompt> is the string to be shown to the caller."~" will be
; translated to <cr><lf>.
;
; For example:
;
; CHOOSER 1 Welcome~ 1 - Randy's Fido~ 2 - Other Side~Choose
; one :
;
; will cause the caller on port 1 to be prompted as follows:
;
; Welcome
; 1 - Randy's Fido
; 2 - Other Side
; Choose one :
;
;
; A FOSSIL driver must have been installed to run this program.
;
; Returned errorlevels are:
; 12 - Bad command line or FOSSIL driver
; 11 - Carrier lost
; 10 - Response was <cr>
; 0..9 - Service 0..9 selected by caller
;
:Restart
CHOOSER %1 OSWEGO OPERA~~ 1 for OSWEGO OPUS~ 2 for OSWEGO's
OTHER Side~~~
IF ERRORLEVEL 11 GOTO End
IF ERRORLEVEL 10 GOTO Normal
IF ERRORLEVEL 2 GOTO Other
IF ERRORLEVEL 1 GOTO Normal
FidoNews 4-26 Page 22 13 Jul 1987
:Normal
OPUS DAY -p%1 -b%2 -t%3
GOTO End
:Other
cd\other
OPUS OTHER -p%1 -b%2 -t%3
cd\opus
GOTO End
:
:End
It should also be noted for any syslops not currently using a
front end like SeaDog of Dutchie that the events and maintenance
is handled by the your SEADOG/DUTCHIE.BAT Remember that the front
end answers the phone not CHOOSER or OPUS.
Bill
President and Chairman of the Board of FUN, an elite organization
for the elimination of UN_FUN.
-----------------------------------------------------------------
FidoNews 4-26 Page 23 13 Jul 1987
Todd Looney
Vietnam Veterans' Valhalla IFNA 143/27
INTERNATIONAL VIETNAM VETERANS' ECHOCONFERENCE ANNIVERSARY!!!
Well, here I am again, this makes about the fifth article I
published in this distinguished (?) newsletter over the past 12
months. Many of you probably at least recognize my name if not
already know who I am. But, for those of you who are not
familiar with my not-so-illustrious personage, I am the SysOp of
the Vietnam Veterans' Valhalla Opus BBS in San Jose, California.
I am also the founder and international coordinator of the
Vietnam Veterans' EchoConference. Without the indefatigable help
and support from the several dozen Fido/Opus/TBBS SysOps who host
the conference on their own bulletin boards making it's
intelligent and mature debates and communications available to
hundreds of readers across the continent. The conference network
has grown steadily from the one lonely BBS in San Jose,
California (then know only as the LooneyBin), to an organized
cluster of some 40 bulletin boards located from the far-off
islands of Hawaii, clear across the central United States to New
York and Canada. The ever-increasing visibility of the
International Vietnam Veterans' EchoConference offered by these
doughty SysOps contributes greatly to it's overwhelming
popularity and undeniable success.
It has been said that the Vietnam Veterans' EchoConference
has become "one of the most important and productive echomail
conferences available in the Untited States today." Frankly, my
own modesty makes me question the validity of that declaration,
but the depth and maturity of the conference message base
substantially demonstrates is the need for a support resourse
like it in todays often mis-directed, largely mis-informed
society whose level of collective familiarity with the era
surrounding the Vietnam conflict is practically non-existent.
A short while back I conducted a survey of the various
SysOps hosting the International Vietnam Veterans' EchoConference
in an attemp to gather statistical data related to their
individual caller bases. The results of my analysis indicate
upwards to 2000 people have read the conference message base, and
that approximately 40% of those are military veterans. The
remaining 60% are composed of a diverse variety of ex-protesters,
ex-supporters, ex-Vietnamese civilians and military, draft
dodgers, etc.
A significant number of vets have found solace in the
International Vietnam Veterans' EchoConference during the year
the echomail project has been in place. They have found a place
where they can "bare their souls" wihtout fear of having to deal
with someone else' reaction to what they have to say. They can
also take their time inputting their messages in the comfort and
safety of their own homes. They can also respond to sincere
questions placed in the conference by the numerous non-vets who
are interested in knowing about the Vietnam conflict and those
who served in the war. Some very meaningful communication has
FidoNews 4-26 Page 24 13 Jul 1987
taken place between the vets and non-vets, as well as every other
combination of the two. A lot of vets have found their way to
professional therapy and guidance through the VA Vietnam Era Vet
Centers. The entire continent-wide network acts like the one
massive support network it has become. We all lend comfort and
support to one another, even if our problem is not directly
related to Vietnam. We all need emotional encouragement at times
and many people have found a niche in the Vietnam Veterans'
EchoConference. And not all of them are veterans.
Last month the conference celebrated its first anniversary.
We were a year old on the 16th of June! We have grown. We are
stable. We are productive. We are a valuable service in this
country where most everything comes at a cost.
We are still trying to spread, too. If you are a sysop and
would like to host this conference, either contact me direct at
143/27, or contact one of the regional coordinators for the
Vietnam Veterans EchoConference listed below. Any one of us will
help you get aboard (with a full message base to boot!).
123/6 Memphis, TN
107/105 Staten Island, NY
17/43 Tacoma, WA
200/100 Redondo Beach, CA
113/1 Hawaii
143/27 San Jose, CA
If you are not a sysop of a Fido or Opus bulletin board,
please give us a call. You should be able to find a local sysop
carrying the conference listed in the ORIGIN statements listed
below each message, or ask your local sysop to host the
conference on his or her board!
We are looking forward to hearing from you!
-----------------------------------------------------------------
FidoNews 4-26 Page 25 13 Jul 1987
=================================================================
COLUMNS
=================================================================
Eric Ewanco, private SEAdog 130/3
Borland's Turbo C:
Review, part 2
(continued from last week)
HARDWARE INTERFACE
Turbo offers more than you could ever imagine in the way of
interfacing with assembler, other languages, interrupts (calling
and being), TSRing, and everything else. Turbo offers (via MASM)
inline ASSEMBLY source which can reference any C symbol
accessible at that point (provided it makes sense in the
context), even labels to C statements. If that's not enough, or
is too much, all registers can be accessed through "pseudo-
variables," like _AX, _BX, etc., for all the general purpose
variables, not including IP and flags (which aren't very
accessible in assembler, either). If you need more power, the
documentation for writing your own assembler programs is clear,
complete, easy to understand, and all-encompassing. It is the
best I have ever seen, and I've seen a lot. In addition, Turbo
allows you to generate interrupts, write interrupt routines,
terminate and stay resident (and Turbo C programs can be
converted into .COM files), call bios routines very easily, set
interrupt vectors, set a control break handler, set a hardware
error handler ("Abort, Retry, Ignore?"), and a million other
things. MS C style DOS calls are offered, too. One example of
these routines is bioscom(), call the BIOS communications
routines. With this call, IT IS POSSIBLE TO INTERFACE TO ANY
FOSSIL DRIVER WITH NO ASSEMBLER PROGRAM SUPPORT! I have in fact
written a program dependant only on Turbo C's libraries to use
Opus!Comm, without the need of external assembler routines. One
statement does it all. THIS is POWER. Turbo Lighting can be
called just as easily using the bioskey(). And Turbo C also
offers Pascal-style parameter passing for those "older" routines
you've written.
DOCUMENTATION
The documentation for Turbo C, contained in two voluminous
manuals, is the User's Guide (300 pages, 2 centimeters thick) and
the Reference Manual (385 pages, 2.5 centimeters). Both are the
usual paperback, and are written so that beginners and experts
alike will love it. It offers an excellent step-by-step
introduction to the integrated environment for beginners, as well
as a short two-chapter coverage of the C language. There are two
chapters for interfacing with Turbo Prolog and a thorough
comparison with Turbo Pascal, complete with side-by-side code,
and, my favorite, caveats to avoid. Borland's insight is
incredible in the two sections they devote to "Common Pitfalls of
Programmers Using C," one for programmers in general, one for
Turbo Pascal programmers. They cover the usual screw-ups, like
FidoNews 4-26 Page 26 13 Jul 1987
using assignment for comparison, the difference between strings
and arrays, zero based arrays, leaving semicolons off last block,
case sensitivity, multi-dimensional arrays, omitting required
parens on function calls, using \ in constant path names, and
forgetting to pass addresses. All utilities are thoroughly
explained in plain language. Also included is a reference to C,
similar to the appendix in the C bible, with C bible cross
reference. Also included is a figure of C constructs, again like
the tail end of the C bible. It covers the ANSI C draft
implementation, too. In general, the documentation is the best I
have ever seen, covering everything you wanted to know. It
handles beginner to operating system developer. The very best,
written in the wonderful Borland style.
ENVIRONMENT
Let's start with the environment. There are two programs on
the Turbo C set; TCC.EXE and TC.EXE. One is the usual integrated
enviroment, close to but slightly different from Turbo Prolog and
Turbo BASIC. The other is the long-awaited command line
environment that the traditional compiler is written as, an
example of Philippe's insight for the programmers of C and care
for the customers who asked for it. Although Turbo C keeps much
of Borland's magic, they seemed to have lost the compiler code
size magic:
TC EXE 231257 5-20-87 1:00a
TCC EXE 169098 5-20-87 1:00a
As you can C, Turbo C is HUGE. So is Turbo BASIC and Prolog.
Borland has moved into hard disk territory. But the features
gained well outweigh it. Both programs put together beat MS C
whose 5 or 6 different files add up to more than a disk, although
they work in overlays (something we may expect from Borland,
hopefully).
The integrated environment is much like the other new
generation Borland programs, with an interface very similar to
TB's. There are two primary windows; one for editing, and one for
messages. All functions are available anyplace in the compiler.
However, the old familiar and much-praised compile to memory is
gone. It has been replaced by a shell-to-dos-and-run scheme,
which is functionally equivalent, but takes up a lot of memory.
One of the unique features of Turbo C is the multiple error
message capability, something long awaited. The Turbo C compiler
will compile and flag as many errors as you select then proceed
to put you, in Borland style, into the source code where the
first error is. Hitting F8 will move to the next place and the
next error found, and so forth. Editing the source does not
invalidate the error positions; further errors will always put
you in the correct place.
One unique thing about Borland's compiling: as you compile
and link, it keeps track of how many warnings and errors flagged,
source lines compiled, and amount of memory left, displayed in
FidoNews 4-26 Page 27 13 Jul 1987
real time. When it finishes, it displays all errors and warnings
and steps through the source code.
The environment also offers an extensive "project" utility,
similar to make, but simpler and easier to use. (A full-fledged
UNIX make is included as a standalone.) The included MicroCalc
has an example of a control file for this project manager:
mcalc (mcalc.h)
mcparser (mcalc.h)
mcdisply (mcalc.h)
mcinput (mcalc.h)
mcommand (mcalc.h)
mcutil (mcalc.h)
mcmvsmem.obj
This indicates that the source files on the left are dependent on
the file in parens. There is also an implicit dependency that all
executable files are dependent on all object files and the
project file itself.
Also included is a small CPP C Pre-Processor utility to
preprocess to a file (strictly no-frills). A Touch utility is
included to update a file's date and time; used to force a make
to occur.
Turbo C comes with its own linker. This linker is about 1/5
the size of the MicroSoft Linker (that comes with MS C) and is
much faster:
TLINK EXE 9753 5-20-87 1:00a
LINK EXE 47896 3-14-86 3:16p
As the manual puts it, "As we said earlier, TLINK is lean and
mean; it does not have an excessive supply of options. . . . it
is not a general replacement for MS Link." Boy, for 1/5 the size,
it sure packs an awful lot. Let me stress again the Turbo C
object modules are supposed to be compatible with every IBM or
MicroSoft linker, although in one instance I ran into problems
using a non-TLINK linker. Borland is working on the problem, as
well as a bigger linker.
CODE SIZE and BENCHMARKS
I am very impressed with Turbo C's "extensive properties," a
phrase I just invented to apply to such things as compilation
speed, code size, execution speed, and the like. For the most
part Turbo C is tight in it's code generation, the most notable
exception being floating point object (but not .EXE) files. Turbo
C always optimizes at least space. You can switch that to speed
or turn on various other optimizations aforementioned. The
compiler is FAST as usual; a moderately long floating-point low-
comment program (164 lines, 4329 bytes) took 48 seconds to
compile and link in TC, but 2 minutes 9 seconds to compile in MS
C. The same program had a TC OBJ size of 4191 and EXE size 28140
emulation, 18236 8087 only; MSC OBJ was 3648, EXE 29126
FidoNews 4-26 Page 28 13 Jul 1987
emulation, 22214 8087, and 3567/24812 for altmath, which Turbo C
doesn't offer, but neither have I found a need for it. As for
runtime execution, Turbo C wins. The same program (which
generates hi-res 3-d surfaces, see 80 Micro, May 83, p. 236) took
4:22 minutes TC and 5:45 minutes MS C (for reference, it takes 15
minutes TPascal 8087, 3 hours C without 8087, 15 hours TPascal
w/o 8087, and well near 24 hours in BASICA). It used square roots
and sines.
Other benchmarks: a UNIX grep utility, no floating point,
compiled in 49 seconds TC no options, 46 seconds no warnings, 41
seconds no floating point link search, and 40 seconds no warnings
or floating point link search; code size (obj/exe) 7025/12864.
MSC took 2 minutes 50 seconds, with code size 6795/13608. My
favorite, the null program ("main (){}"), took 28 seconds in TC
and 50 seconds in MS C; code size was 160/1694 TC and 276/1986
MSC. All times are for both compile and link.
I have had only one problem in the week or two I've had Turbo
C, and that was the handling of floating point exceptions. This
is the only problem I've seen; other than that it's flawless.
In conclusion, Turbo C is 99% the functionality of MS C in
some places, 101% in others, but at 20% the price (discount;
reference: PC Connection), and besides that, it took them one
version, not three or four. It is fast in both compiling and
executing, produces tight code, offers advanced warnings, can
produce .COM files, can be used to write TSRs, has several
optimization settings, and will be very well supported. Every
programmer should have a copy. However, if you do want to get a
copy, I recommend expedience, because Borland does not guarantee
the current price beyond July 1st, and Turbo C is sure worth a
heck of a lot more than $99.
-----------------------------------------------------------------
FidoNews 4-26 Page 29 13 Jul 1987
Glen Jackson
Broadcast Software BBS
Fido/SEAdog 100/517
Echo mail routing
It's been 5 months now since I have been into Echo Mail. I run a
backbone on one echo (*plug on* - the Broadcast echos , also
including the Request Line for the Public - *plug off*), feed
another node across the country, and am down line on several
others. Because of this, I have run into several combinations of
software (SEAdog/TBBS, SEAdog/OPUS, SEAdog/FIDO, FIDO/OPUS, and
of course, FIDO by itself).
The biggest problem that most of the mail systems run into is
this:
Node A polls or sends some mail to Node B. Node B has
messages waiting for Node A, but when Node A calls, he isn't
given any mail packets.
Let's start with the route files. Here's how we want our route to
look every night:
at 1:00 AM we will poll 1000/1 for any mail
at 3:30 AM we will route outgoing mail to our host
at 4:00 AM we run National Mail hour
at 5:00 AM we will hold mail for 1000/2 to pick up.
This takes 4 seperate routes. If you run SEAdog, they can all go
into your one ROUTE.DOG file. Here's how they should look:
Schedule E ; (route.e) polls 1000/1
HOLD ALL except 1000/1
SEND-TO ALL
GIVE-TO ALL
POLL 1000/1
- the HOLD statement places all mail in a hold status in case
another node calls in looking for his mail. DO NOT HOLD any
mail packets to nodes you are sending to or polling.
- the SEND-TO statement allows HOLD mail to be released if an
incoming call comes in.
- the GIVE-TO will give the mail you have for 1000/1 to 1000/1
when you call him.
- the POLL statement makes sure that you will call 1000/1 whether
or not you have mail for him. If you only want to call 1000/1
IF you have mail for him, don't use the POLL statement.
Next, we want to route our mail to our host. But, in case we did
not connect with 1000/1, we still want to keep it here and not
route it. Run a route that looks like this:
FidoNews 4-26 Page 30 13 Jul 1987
Schedule G ; (route.g) local routing
ROUTE-TO [your host net/node] ALL
NO-ROUTE 1000/1 1000/2
SEND-TO [your host net/node]
PICKUP [your host net/node]
GIVE-TO ALL
- the ROUTE-TO will route ALL to [your host net/node].
- the NO-ROUTE keeps mail you hold at a later time from being
sent over to your host for routing.
- the SEND-TO and PICKUP are here for you to be able to make the
call to your host if you have mail to send out.
- the GIVE-TO statement gives mail to incoming callers that have
mail waiting.
Next, we want to run the National Mail hour. Remember, we don't
want to send or route any of 1000/1 or 1000/2 mail, as they are
dealt with later. Here's the sample route file:
Schedule A ; (route.a) natl mail hour
NO-ROUTE 1000/1 1000/2
HOLD 1000/1 1000/2
SEND-TO ALL
GIVE-TO ALL
- the NO-ROUTE makes sure that the mail for 1000/1 and 1000/2
stays at your node.
- the HOLD keeps your node from calling these nodes directly.
- the SEND-TO calls the nodes you have not routed mail for, and
that you have mail at your location for.
- the GIVE-TO gives the mail to any incoming calls that have any
mail waiting.
You could also add a PICKUP ALL in the route so when you call a
node to drop off mail, you can pick up any mail for you that he
may have waiting.
Confused YET ? If not, I'll try harder.
Now, we're going to hold some mail for 1000/2. Here's where most
nodes get into trouble- the ability to actually give mail to an
incoming caller that has messages there.
Here's out holding route for 1000/2:
Schedule F ; (route.f) hold for 1000/2
HOLD ALL
SEND-TO ALL
GIVE-TO ALL
FidoNews 4-26 Page 31 13 Jul 1987
- the HOLD keeps your system from calling out.
- the SEND-TO releases packets to an incoming caller that has
mail waiting.
- the GIVE-TO actually gives the packet to the caller.
Ahhhhh, finished. Just remember this important rule:
* you MUST BE IN A MAIL EVENT TO GIVE MAIL TO A POLLING NODE, no
matter what type of system you run. *
Next week. we'll take a look at the schedule configurations, and
if you run SEAdog, how to put up a true 24 hour mail system with
your BBS.
Don't forget to call in and request our latest utility - MSGDB.
It can be requested by filename MSGDB.ARC . The arc file also
contains FIXDATE - a simple utility that corrects those off the
wall dates we see in the echo areas.
That's all for now. If you have any broadcasters calling into
your BBS, you may want to join one of our Broadcast Echos. Just
Net_Mail me at 100/517...
And that's the way it was...
from the Broadcast Booth,
Glen Jackson
-----------------------------------------------------------------
FidoNews 4-26 Page 32 13 Jul 1987
-- The Regular Irregular Column --
Dale Lovell
157/504
Well, things have gone somewhat better this week. The new
system has been performing flawlessly, and I actually enjoy the
new keyboard. Things are looking up here, if things keep working
out I may even become a public access board again (right now I'm
a private node, so mail to me has to be routed). This has made it
necessary for me to look into a few new things, so I'm going to
start off by asking for your help.
-- What's a LAN? --
As you may recall, last week's problems were caused by a
keyboard failure and a lack of my old XT clone. In an effort to
prevent that from ever happening again, I bought a stripped down
clone at a hamfest this weekend (someone else had just upgraded
to an AT and wanted to get rid of their old machine). After I got
it home, I dug up some old cards and brought it up to 640K with a
20 meg hard disk. While this is primarily intended as a backup
machine, I've been thinking about using it as a public bulletin
board. After I got everything hooked up and tested, an idea
occurred to me. Instead of having two separate machines; one for
me to work on, and another dedicated to a bulletin board; why not
attach the two with a local area network. This way I could go
over my mail, maintain the board, and not have to duplicate
message bases on two different machines. There's only one problem
with this, I don't know the first thing about LANs.
This isn't to say I don't approve of (or rather haven't
approved of) local area networks. It has always sounded like a
great way to do some things (like sharing an expensive printer).
I've just never been in an environment where it was practical. At
my current job, there's exactly one PC in the office and it sits
on my desk. It's a small office and no one else really knows how
to use a computer. I got the AT at the office because my boss
noticed how long it took the computer to recalculate some
spreadsheets and re-index a database. He asked me if there was a
way to speed it up, and the company ended up buying an AT for me
to work on. Most of what I've learned has come through actual
experience, and I've just never has a chance to install or work
on a LAN. Since I now have a situation where one might be useful,
I'm going to look into buying one. In this endeavor, I ask for
your help.
I currently have an AT clone running at 10 mhz with a 42 meg
MiniScribe hard disk (using SpeedStor to make it look like three
drives; a 21 meg drive for my work, a 10 meg drive for files, and
a 10 meg drive for messages with 1K clusters). The "new" XT clone
is running at 4.77 mhz and has a 20 meg Seagate ST225 with a
Western Digital controller. What I'd like to do is set up the XT
as a bulletin board, with all BBS programs, utilities, file
areas, and message bases residing on it's own hard drive. The AT
should ideally be able to read the message bases (through
SEAdog's user interface), and be able to do the bulletin board's
FidoNews 4-26 Page 33 13 Jul 1987
maintenance. If possible I would also like the AT to be able to
use the modem in the XT for dialing out (mainly to the Unix
machine I use occasionally) although it wouldn't be necessary.
Since I don't expect any LAN to fill these needs perfectly, I'd
appreciate hearing from your own experiences with ANY LAN. At the
end of the column there are several different adresses for me,
just make certain you specify which LAN you're using and what
software is being used to run it. I'll sum up the opinions and
advantages between the different networks sometime in the future.
-- The AT clone --
My first comment on the new AT clone is WOW! Unless you have
ever worked on an AT before, you would never believe the
difference in speed. As an example of this, my echomail
processing has dropped to an unbelievably short time. In spite of
my private status, I pass along echomail to 5 nodes in net 157. I
start processing the echomail at 6:30 every morning (NMH + 30
minutes). As soon as it's done, I send it out to the four nodes
that are running SEAdog. In the past, my system usually finished
up the echomail processing between 7:30 and 8:30 depending on the
amount of mail being sent out. Since I switched everything over
to the AT, the mail has gone out by 7:00 every day (at least it
starts calling out at 7:00). I'm using this as a comparison only
because it will hopefully mean something to most of the sysops in
FidoNet, at least I think it means more than a straight out
benchmark (Norton's SI of 9.8, just in case you like that
"benchmark").
-- Speedstor (Storage Dimensions, $99.95) --
The Miniscribe 6053 hard disk came with SpeedStor, which is
a partitioning "program." With any drive over 32 meg you run into
a problem, DOS won't recognize a larger hard drive (PC-DOS 3.3 is
supposed to "fix" this, but I've heard that it has problems with
non-IBM equipment). While it is possible to do strange things
with FDISK and your controller card to make these larger drives
act like several different hard drives (C,D,E,etc) it was never
that easy. Speedstor makes this task easy, while also allowing
you to use a "non-standard drive." In an AT, all the information
on the type of hard drive is stored in non-volatile memory. There
are at least 14 different types of hard drives that the AT knows
about (IBM started off with 14 and everyone supports at least
those same drive types) that can be used with no trouble.
Unfortunately the Miniscribe 6053 isn't usually one of them. In
the past this meant part of the hard disk was wasted, as the AT
didn't know how to address the extra cylinders or heads.
Speedstor gets around this by keeping it's own information on the
type of hard drive being used (I think in the non-volatile RAM,
but I'm not certain). When the system boots up, it loads the
Speedstor software (one line in the CONFIG.SYS) and your system
now knows about this non-standard drive.
In addition to letting use your drive to it's full capacity,
Speedstor let's you do some more unusual things. While I'm fairly
certain that it will let you get past the 32 meg limit of DOS, I
FidoNews 4-26 Page 34 13 Jul 1987
didn't try it. What it does do very easily is allow each hard
drive partition act like a separate drive. Right now I have one
physical hard disk in the AT. But DOS "knows" about drives C,D
and E; and I don't have any RAM disks installed. Using Speedstor
I made three partitions on the hard disk. The first partition is
using one half of the hard drive and is a bootable DOS partition.
The other two partitions each have one quarter of the hard drive
and are usable by DOS (although I may change this in the future
to support XENIX). The nice thing is that the third partition has
a 1K cluster size. Under DOS 3.x cluster sizes are normally 2048
bytes (2K), this means your two line batch file is going to use
2K of your hard disk. If you're running a DOS 2.x it get's even
worse as those versions of DOS use an 8K (8192 bytes) cluster
size. Under Speedstor I can move the cluster size down to 512
bytes, although when I tried that with a 10 meg partition it gave
me some warning about not enough space in the FAT table and
CHKDSK would have problems (I chickened out and moved it up to
the next choice). Fido and Opus sysops take note, you don't have
to lose insufferable amounts of disk space (due to slack) because
of your message bases. I "gained" over 1 meg of disk space by
putting my message bases on this partition with it's 1K clusters.
If I had felt like spending a little more time I may have played
around with the partition size and cluster size and gotten
something with no warnings and a 512 byte cluster size, although
I don't think the space savings would have been as great.
If you aren't doing anything unusual, like I did, it
installs very quickly. I initially ran Speedstor's semi-automated
installation and ended up with two 21 meg partitions. All I had
to do was tell it what type of drive I was using. On the menu it
had a choice called manufacturer. After I choose it, I was able
to pick the drive out of a list of the three current Miniscribe
drives. Since my copy is the "Miniscribe version" I'd assume that
the full version would include a fuller list of drives from
several manufacturers. Other than that, about all I had to enter
was the bad track table for the drive. Almost everything else was
completely automatic. Miniscribe is now including Speedstor with
all of their larger drives, and from what my dealer tells me
Seagate is doing something similar with a program called Disk
Manager by OnTrack (Although he says it is nowhere near as
powerful as Speedstor). It's nice to see the hard drive
manufacturers are interested in the end user, not just getting
their money. While including Speedstor or Disk Manager with the
drive may kick up the price a few dollars, I feel better getting
everything I need at once rather than getting the drive and
finding out a need something else to really use it.
-- SYSEDIT and SuperKey --
After I got all the data transferred over to the AT, it came
time to change all my SYSTEM*.BBS files. Because of the different
way I set up the new system, I needed to change a lot of message
and file paths. In the past everything was on the same logical
drive (as opposed to a physical drive, of which I have only one),
while now it was on two logical drives. This wasn't too bad as I
dug up a copy of Eric Ewanco's SYSEDIT. SYSEDIT let's you go in
FidoNews 4-26 Page 35 13 Jul 1987
and edit the system files from DOS. It has a nice full screen
display and is very easy to use. I started doing every change
manually but quickly decided there had to be an easier way to do
things (I have over 20 echomail conferences going through my
machine). Enter SuperKey from Borland (list price $69.95) a
keyboard macro program. I quickly installed SuperKey and taught
it what needed to be done on all the message bases. I had to put
in what Borland refers to as a keyboard delay because SYSEDIT
didn't use the keyboard buffers, but that was the only
difficulty. In under five minutes I had all my system files
changed and was ready to go, or so I thought. I tried running
some of the echomail programs, but it couldn't find some needed
directories. After puzzling over this and checking the system
files I finally realized what I had forgotten. I had changed all
the system files, but had forgotten about SEAdog's CONFIG.DOG,
AREAS.DOG, and the Fido/Opus MAIL.SYS file. After correcting my
oversight it was finally changed over and everything has been
running fine since. SYSEDIT and SuperKey are both excellent
programs and helped make this switch over much easier than I had
anticipated. I've seen I program similar to SYSEDIT for OPUS
systems, but I haven't managed to find it in the Opus files
directory. Programs like SYSEDIT do a lot to make a sysop's life
easier, I'd encourage you to dig up a program like this (SYSEDIT
in particular for Fido sysops) if you are currently running a
bulletin board.
-- Winding down... --
Since I became interested in local area networks this past
week, the book I'm going to recommend is going to reflect my
current concerns. "Networking IBM PCs, A Practical Guide" by
Michael Durr (QUE books, $18.95) has provided me with a lot of
information on how a LAN works. While the copy I got ($10 at the
hamfest I attended) is somewhat out of date, many of the specific
LANs mentioned are still with us. It provides a description on
several different LANs (probably most of the one's available in
1984, the year it was written) and goes over the features
available under the network software. It covers many topics that
I have found to be of interest such as administration,
performance, maintenance and such. It has helped explain a lot of
what goes on in a LAN, and has helped me decide what I'll want
included in the LAN I buy (Such as a non-dedicated server). While
the book has raised many questions for me, it has gotten me to
start thinking about more than the brand name. I think I could
now shop intelligently for a LAN, and have a much better chance
of coming out with something that will do the job versus
something the salesman wanted to sell me. After all, isn't that
the whole idea behind learning a little about something before
you go out and buy it?
I once again welcome your comments, even ask for them if you
can help me out with information on LANs. Down below you'll find
several addresses for me. Some of you have asked about Compuserve
or The Source mailboxes or account numbers. Unfortunately I don't
maintain an account on any of these systems, only on the more
public access networks. Since I'm a private node in the nodelist,
FidoNews 4-26 Page 36 13 Jul 1987
all FidoNet mail to me should be routed through 157/1 or 157/0
(same board really). If you send me anything through US mail,
please make sure I have some sort of return address (are you
hearing this "Ender Wiggin?") in order to send you a reply. Next
week I should be looking at a game, so all you fellow gaming
people who have sent me mail be patient. I haven't forgotten you.
-- Late Breaking News --
Couple of quickies here that "came in" just as I was
finishing this column. First, the latest version of the echomail
conference list is now available from Thomas Kenny at 107/316.
You can file request under the name "ECHOLIST.ARC." Also, Hewlett
Packard Laser Jet printer owners will be pleased to hear that
Hewlett Packard hopes to have a PostScript update for all of
their laser printers available by the end of the year. I'm a
strong supporter of PostScript, even if IBM has endorsed it
(Whadda mean Thom? I thought everybody who wrote on the computer
scene was supposed ta hate IBM. You mean I can actually support
an IBM decision? S'Allright....)
Dale Lovell
3266 Vezber Drive
Seven Hills, OH 44131
FidoNet: 157/504 (or 1:157/504.1 for an extended address)
uucp:
decvax\
>!cwruecmp!hal\
cbosgd/ >!ncoast!lovell
/--!necntc/
ames---\ /
talcott \/
harvard /
sri-nic/
-----------------------------------------------------------------
FidoNews 4-26 Page 37 13 Jul 1987
=================================================================
WANTED
=================================================================
INFORMATION FEEDS WANTED
I am currently involved in a project in which I have a need for
"information providers" who are willing to do the following.
1) Write to their State Lottery Commission and request a complete
history of all numbers drawn for each of the game(s) played in
that state.
2) Send a daily "crash mail" message containing a listing of the
game(s) played that day, and the number(s) drawn for those
games AS SOON AFTER THE DRAWING AS POSSIBLE!
At this time I am willing to pay $10 to the first person in each
of the following states who sends me the "history" listed in (1)
above. (Please send a "crashmail" message before you send the
data to insure that you are indeed the first.)
ARIZONA, CALIFORNIA, CONNECTICUT, DELAWARE, D.C., ILLINOIS,
IOWA, MARYLAND, MASSACHUSETTS, MICHIGAN, MISSOURI, NEW
JERSEY, NEW YORK, "N.NEW ENGLAND," OHIO, OREGON, RHODE
ISLAND, WASHINGTON & WEST VIRGINIA
As for (2) I am unable to promise any payment at this time,
however, if the concept is sold it may mean payment for this
information in the near future!
Please direct all inquiries to me, John Penberthy at 129/200,
129/28, or voice 412-364-3951.
-----------------------------------------------------------------
FidoNews 4-26 Page 38 13 Jul 1987
=================================================================
NOTICES
=================================================================
The Interrupt Stack
1 Aug 1987
Third Annual BBS Picnic in Edison, NJ. Please register before
July 10th. Admission is $7 for adults, $4 for 12 and under,
free for 5 and under. Contact John Kelley at 107/331 for
details.
20 Aug 1987
Start of the Fourth International FidoNet Conference, to be
held at the Radisson Mark Plaza Hotel in Alexandria, VA.
Contact Brian Hughes at 109/634 for more information. This is
FidoNet's big annual get-together, and is your chance to meet
all the people you've been talking with all this time. We're
hoping to see you there!
24 Aug 1989
Voyager 2 passes Neptune.
If you have something which you would like to see on this
calendar, please send a message to FidoNet node 1:1/1.
-----------------------------------------------------------------
Latest Software Versions
BBS Systems Node List Other
& Mailers Version Utilities Version Utilities Version
Dutchie 2.51 EDITNL 3.3 ARC 5.21
Fido 11w LISTGEN 05.25.86 ARCmail 0.60
Opus 0.00 Prune 1.40 EchoMail 1.31
SEAdog 4.00 TestList 8.3 FastEcho 2.00
TBBS 2.0M XlatList 2.81 Renum 3.30
* Recently changed
Utility authors: Please help keep this list as current as
possible by reporting new versions to 1:1/1. It is not our
intent to list all utilities here, only those which verge on
necessity.
-----------------------------------------------------------------
FidoNews 4-26 Page 39 13 Jul 1987
OFFICIAL REGISTRATION FORM
FOURTH INTERNATIONAL FIDONET CONFERENCE
RADISSON MARK PLAZA HOTEL
ALEXANDRIA, VA.
AUGUST 20 - 23, 1987
Name _________________________________ Date _____________
Address ______________________________
City & State _________________________
Phone (Voice) ________________________
Net/Node Number ______________________
Phone (Data) _________________________
Number in Your Party _________________
Staying at the Radisson? _____________
Number of Rooms? _____________________
Arrival Date? ________________________
Departure Date? ______________________
Registration Fees: How Many Total
Full Conference $60.00 each ________ $________
Late registration $10.00 each ________ $________
(after Aug. 1)
Friday Night Banquet $30.00 each ________ $________
Saturday Luncheon $16.50 each ________ $________
Total Amount Included (Registration and Meals) $________
IFNA MEMBERS ONLY:
How many in your party will
be attending the Sunday morning
Board of Directors meeting? ________
Send your registration form and a check or money order to:
Fourth International FidoNet Conference
212 E. Capitol St., Washington, D.C. 20003
Attn: Brian H. Hughes -- voice: (202) 543-4200
This registration form does not include hotel accomodations. If
you wish to stay at the Radisson Mark Plaza Hotel, please contact
them directly and mention you are with the FidoNet Conference.
Conference room rates are $80/night for single or double
occupancy, and $20/night for an extra cot.
Radisson Mark Plaza Hotel
5000 Seminary Rd., Alexandria, Va. 22311
1-800-228-9822
-----------------------------------------------------------------
FidoNews 4-26 Page 40 13 Jul 1987
Bob Morris 141/333
Chairman, Elections and Nominations Committee
The next two pages are your Official ballot for the Election of
the IFNA Board of Directors. The following are the few rules
which must prevail in this election:
1. You must send a legible copy of this ballot to the address
listed on the ballot. It must be signed and bear your
net/node number.
2. You may vote for any one person in your region for the
position of Regional Director. This vote is to be cast in the
LEFT column of the ballot.
3. You may vote for any eleven people in any regions for the
position of Director at Large. These votes are to be cast in
the RIGHT column of the ballot.
4. Voting will continue until the end of registration at the
Conference in August. The results will be read during the
opening of the business meeting on the first day of the
conference.
5. Write-in Votes will be accepted and are requested during this
election.
FidoNews 4-26 Page 41 13 Jul 1987
IFNA Board Of Directors
Ballot
Regional At Large
Region 10:
Steve Jordan _________ ________
Region 11:
Ryugen Fisher _________ ________
Theodore Polczynski _________ ________
Region 12:
Region 13:
Don Daniels _________ ________
John Penberthy _________ ________
Thom Henderson _________ ________
Gee Wong _________ ________
Brian Hughes _________ ________
Region 14:
Ben Baker _________ ________
Ken Kaplan _________ ________
Brad Hicks _________ ________
Region 15:
David Dodell _________ ________
Larry Wall _________ ________
Region 16:
Bob Hartman _________ ________
Hal Duprie _________ ________
Region 17:
Rob Barker _________ ________
Bob Swift _________ ________
Region 18:
Wes Cowley _________ ________
FidoNews 4-26 Page 42 13 Jul 1987
Region 19:
Mark Grennan _________ ________
Wynn Wagner _________ ________
Region 2:
Henk Wevers _________ ________
Write-in candidates:
___________________ _________ ________
___________________ _________ ________
Name ______________________________ Net/Node ___________
Signature______________________________ Date ___________
Please complete this and mail it to:
Robert Morris
IFNA Elections Committee
210 Church Street
West Haven, Ct. 06516
or bring it with you when you come to the conference in August.
These ballots will be counted by myself since with 200 members
the charges for a CPA would be very high. Hard copies will be
made available to anyone wishing to insure that their vote was
included.
Thank You
Bob Morris
Elections and Nominations Committee
-----------------------------------------------------------------
--
Tim Pozar
UUCP pozar@hoptoad.UUCP
Fido 125/406
USNail KLOK-FM
77 Maiden Lane
San Francisco CA 9'SOR