A Beginner's Guide to the Internet/INTERNET.ISO/text/telediag/tc13-089.txt

Text File | 1996-05-06 | 17.9 KB | 446 lines

TELECOM Digest     Sat, 13 Feb 93 23:48:30 CST    Volume 13 : Issue 89

Index To This Issue:                     Moderator: Patrick A. Townson

    The Definitive Word on TTL Fields (Mark Boolootian)
    Re: Meaning of TTL in TCP/IP (Eric M. Carroll)
    Re: AT&T Are You Listening? (Jack Decker)
    Re: AT&T Are You Listening? (Ed Greenberg)
    Re: ANI on 800 Line w/o T1? (John Llorens)
    Re: ANI on 800 Line w/o T1? (John Higdon)
    Re: FCC Proposed Ruling on Scanners That Receive Cellphones (Ken Stox)
    Re: FCC Proposed Ruling on Scanners That Receive Cellphones (John Higdon)
----------------------------------------------------------------------

From: [e-mail] (Mark Boolootian)
Subject: The Definitive Word on TTL Fields
Date: Sat, 13 Feb 1993 13:01:45 -0800 (PST)


I don't want to drag this on ad infinitum, but it is important to get
the correct information out and that doesn't seem to be happening
here.

Matti Aarnio <[e-mail]> writes:

> Lately I have seen this "TTL is time in seconds" dream on many places.
> Who originated that idea?  Current IP version 4 does not contain an
> absolute time stamp in each datagram, rather it has a hop-count.

and, referring to TTL, again writes:

> Nope, it is not SECONDS, it is HOPS.  (Gomer does explain it well)
> (And it is explained well in the original RFCs too, see RFC 791).

While it is a minor point, the TTL field is in fact specified as being
in seconds in RFC 791 (and also in RFC 1122 - Host Requirements).
Here is the salient passage from 791:

    Time to Live:

    The time to live is set by the sender to the maximum time the
    datagram is allowed to be in the internet system.  If the datagram
    is in the internet system longer than the time to live, then the
    datagram must be destroyed.

    This field must be decreased at each point that the internet header
    is processed to reflect the time spent processing the datagram.
    Even if no local information is available on the time actually
    spent, the field must be decremented by 1.  The time is measured in
    units of seconds (i.e. the value 1 means one second).  Thus, the
    maximum time to live is 255 seconds or 4.25 minutes.  Since every
    module that processes a datagram must decrease the TTL by at least
    one even if it process the datagram in less than a second, the TTL
    must be thought of only as an upper bound on the time a datagram may
    exist.  The intention is to cause undeliverable datagrams to be
    discarded, and to bound the maximum datagram lifetime.


Since routers typically decrement the TTL by one, the TTL is typically
referred to as a hop count.  But according to the spec, it is
acceptable for a router to decrement the TTL by some larger value.  I
am unaware of any cases where this occurs, but it is within the realm
of possibilty.

>> By looking at Jack Decker's traceroute output, it seems that his TCP
>> packets are going through too many gateways.

> Nope, 30 is quite normal amount of router hops from one edge to
> another of the present Internet network. Even 50 is not that unusual.

I would like to challenge you to produce a traceroute for me with 30
hops in it (and no routing loops!).  My understanding of why a TTL of
60 is commonly used in TCP is that it virtually guarantees a packet
can move from one edge of the Internet to the other and back again
without being discarded (i.e. the maximum width of the Internet is <
30 hops).  50 hops is unbelievable.

I want to correct something *I* wrote in my last post.  I said:

> By the way, most implemenations of traceroute use a TTL of 30.

I should have said, most implementations of traceroute use a maximum
default TTL of 30 (the reason being is that is usually sufficient to
get where you need to go).


Mark Boolootian        [e-mail]        [phone]
Disclaimer:  booloo speaks for booloo and no other.

------------------------------

From: [e-mail] (Eric M. Carroll)
Subject: Re: Meaning of TTL in TCP/IP
Organization: UTCS Campus Access
Date: Sat, 13 Feb 1993 16:21:14 -0500


> Nope, it is not SECONDS, it is HOPS.   ...
> (And it is explained well in the original RFCs too, see RFC 791).

The TTL is in seconds. But a router is required to decrement it by at
least one. Thus it has hop-count properties. RFC791 defines it, and
RFC1122 clarifies it.

 From RFC791:

Time to Live:  8 bits

[Moderator's Note: This message quoted the very same text and has been
omitted here, since he repeats what the first message in this issue
said quoted from the same source by Mark B.  PAT]

 From Host Requirements RFC1122:

3.2.1.7  Time-to-Live: RFC-791 Section 3.2

DISCUSSION:
                 The TTL field has two functions: limit the lifetime of
                 TCP segments (see RFC-793 [TCP:1], p. 28), and
                 terminate Internet routing loops.  Although TTL is a
                 time in seconds, it also has some attributes of a hop-
                 count, since each gateway is required to reduce the TTL
                 field by at least one.


Eric Carroll     University of Toronto Computing & Communications
                 Network & Operations Services, Network Development

------------------------------

Date: Sat, 13 Feb 93 15:02:50 EST
From: [e-mail] (Jack Decker)
Subject: Re: AT&T Are You Listening?


In message <[e-mail]>, [e-mail] (Laurence
Chiu) wrote:

> I am thinking about MCI instead. They have a similar plan to ROW
> (don't remember the name) with similar rates but also they have
> Friends and Family. I can denote two international numbers F&F (China
> is allowed) and thus get savings to all the international destinations
> I call.

> I don't want to switch carriers. I like AT&T. Their lines are good
> (I've made consistent 14.4 connects all over the country using my
> cheap Supra modem), their customer service very friendly etc. But they
> just don't offer the calling plans I want. Anyway MCI advertise all
> the time on our local channel which carries Chinese broadcasting, in
> both Cantonese and Mandarin with toll-free numbers you can call which
> are answered by Chinese speaking operators so my wife is always asking
> me why I don't use MCI! I cannot explain modems to her that easily!

Why not use both carriers?

I would suggest you do this: Switch your line to MCI and use them for
all your voice calls.  You might want to try your modem calls at least
once; you may well be surprised to find that the line quality to the
places you call is perfectly acceptable.

However, should you find a place that you cannot call via MCI, simply
force the call to go via AT&T, by dialing 10288, then 1, the area code
and number.

If there is a location that you have consistent problems calling via
MCI, by all means call up MCI's customer service and complain!  But
honestly, you should not have that many problems.  All three of the
major carriers use digital fiber optic circuits now so you really
should not notice a difference in the quality of your modem connects,
unless you are calling some place out in the hinterlands.

But remember, as long as your phone line has been converted for "equal
access" (MOST lines in the U.S. have by now), you can ALWAYS force a
call to go by the carrier of your choice, simply by prefixing whatever
you'd normally dial with the carrier's "10xxx" code.  Thus, if you'd
normally dial 1-(xxx) xxx-xxxx, you'd instead dial 10xxx-1 (xxx)
xxx-xxxx.  The "10xxx" codes for the "big three" carriers are:

10222 - MCI
10288 - AT&T
10333 - Sprint

(I know this is all "common knowledge" for regular readers of the
TELECOM Digest, but I have to imagine that new folks stumble through
here at least once in a while!)  :-)

AT&T spends a LOT of money on advertising to convince you that their
quality is better.  In my mind, this is just about as valid as the
advertising that oil companies used to run to convince you that one
brand of gasoline was better than another.  In many cases, all the gas

stations in a town got their gas from the same source!  It was the
same gas, yet they all tried to convince the public that theirs was
better!

And we see a similar situation here, in that in many cases if you are
going to have trouble on a call, the trouble is most likely to be in
the local telco's lines from their point-of-presence (of which there
is often only one per LATA) to their local exchange, and ALL the
carriers use those same lines to complete calls.  If you try a call
over MCI and it doesn't work, and you then try to complete it over
AT&T and it does, that doesn't necessarily mean that AT&T is better,
it just means you got a different circuit from the local telco.  Had
you tried your second attempt on MCI again, you would probably have
been just as satisfied with the result.

Mind you, I can find things to complain about regarding all the
carriers, but with AT&T my biggest complaint is that AT&T tries to
make it sound as though their lines are so vastly superior to that of
their competitors when the truth is that there really isn't any
perceptible difference in most cases (certainly nothing you could hear
on a voice call ... and in the past I have seen a few cases where you
couldn't make a modem connect over AT&T, while using MCI produced a
near-perfect connection.  Still, that could also have been a local
telco problem).

AT&T's new fax commercials really get me ... they talk about their fax
guarantee but insofar as I can determine from the short amount of time
that the fine print remains on the screen, they only guarantee to
credit your bill for the cost of any failed fax call.  Have the other
carriers refused to do that?  I rather doubt it, since essentially we
are talking a "poor connection" and the major carriers have all issued
credits for those for quite some time!


Jack Decker | Internet: [e-mail] | Fidonet: 1:154/8

------------------------------

From: [e-mail] (Ed Greenberg)
Subject: Re: AT&T Are You Listening?
Organization: Netcom Online Communications Services (408-241-9760 login: guest)
Date: Sat, 13 Feb 1993 13:22:39 GMT


In article <[e-mail]> John Higdon <[e-mail].
com> writes:

> For instance, keep AT&T as your PIC and then sign up with MCI for the
> international plan, keeping your MCI account secondary. Then when you
> call China (or whereever), prepend '10222' to the call. Frontline
> salesslime will tell you that you have to "switch" to take advantage
> of a particular plan, but that is 99.9% hogwash.

> I have used various plans offered by Sprint and MCI, but I have never
> "left" AT&T. There is no need.

I can second this.  I have an MCI personal 800 number. Even though my
PIC is AT&T, I have a "Friends and Family" account with one number on
it -- mine.  I discount the cost of my 800 calls home by 20% this way.
The whole thing was set up for me by a courteous, respectful, _telecom_
literate_ MCI rep who understood me perfectly when I said that I had
no intention of switching PICs but wanted an MCI 800 number.  She
explained that I could add F&F to a secondary account and set it up.
I checked and checked, waiting to be slammed, and imagine my surprise
when I wasn't!


Edward W. Greenberg  | Home: [phone] | [e-mail]
1600 Stokes St. #24  | Work: [phone] | DoD#: 0357
San Jose, CA  95126  | Fax:  [phone] | KM6CG (ex WB2GOH)

------------------------------

From: [e-mail] (John Llorens)
Subject: Re: ANI on 800 Line w/o T1?
Organization: Compaq Computer Corp.
Date: Sat, 13 Feb 1993 17:34:17 GMT


In article <[e-mail]> John Higdon <john@zygot.
ati.com> writes:

> Tas Dienes <[e-mail]> writes:

>> Does anybody know if it is possible to get ANI on an 800 line without
>> having to get T1 service?  I just have a couple of regular (actually,
>> Centranet) lines - local service is GTE, 800 is Sprint.  Sprint says
>> no, but I was wondering if anybody else can?

> In order to receive realtime ANI from a long distance carrier, you
> must have a "trunk-side connection". All connections from your telco's
> switch are "line-side connections". So the answer is no, you cannot
> get realtime ANI without having a direct trunk connection to a
> carrier's switch.

> There is a way around this however. Stay tuned.

You can recieve realtime ANI and DNIS on reqular voice lines via a
DTMF header that begins the call.  While this information is being
transfered to the called, the caller is receiving a ring back tone.

When you (called) have processed the ANI/DNIS (database lookup,
operator available, etc ...) you send a special DTMF to the switch to
connect the voice path.  This method requires a Dialogic or equivalent
tone detection/generation equipment at the receiver's end.

Who offers this service?

   Arch Telecom
   Houston TX
   [phone] or 1-800-882-2947

How much does it cost?

   ATT Readyline rates!

Who is the carrier?

   ATT (ISDN inbound and outbound for extremely fast call setup times)

Features in addition to ANI/DNIS delivery?

   ANI Blocking/Passing can be programmed via touchtone.
   Takes effect immediately.

Destination number redirect to any domestic number.

   Can be programmed via touchtone.
   Takes effect immediately.

Can be used as a calling card.

   Speed dial codes

I think they also offer this service in combination with ATT's new
Vari-a-bill 900 service.

etc ... I don't remember all the services/features...

I don't know of anyone else using this method for switched long
distance service. I hope the info is useful to those in need of this
service.


John Llorens    Compaq Computer Corporation   [e-mail]

------------------------------

Date: Sat, 13 Feb 93 13:32 PST
From: [e-mail] (John Higdon)
Reply-To: John Higdon <[e-mail]>
Organization: Green Hills and Cows
Subject: Re: ANI on 800 Line w/o T1?


tim gorman <[e-mail]> writes:

> First, there are ways to get ANI over 800 lines without T1 service.
> It depends entirely on what your long distance provider offers.

> If the carrier can offer standard trunking with the ANI sent as an MF
> (or possible DTMF) digit stream and your PBX can handle the specific
> format, then a T1 is not a requirement.

> If your PBX can handle Feature Group D signaling formats,

Were you responding to the same post I was? The original poster said
that he had ordinary POTS (well Centrex anyway) from GTE and no
special equipment. I assumed he was asking if he could casually get
ANI from a carrier without T1.

All POTS circuits are "line side". I never claimed that you could not
get trunk connections from telco (I have them in certain
applications), but the question was not about PBXes, complex
alternatives to IXC direct trunking or anything else. It was about ANI
on POTS. Of course, it is certainly possible for a carrier to provide
there is no such offering at this time.

There is even a rumored service from AT&T that provides ANI via the
CNID mechanism to customers. Only problem with that (and why I did not
mention it) is that the poster is from California, the land of
politically correct, anonymous thinking. (An aside: the pest who
called me fifteen times this morning, hanging up each time I answered
will have no worry about HIS identity being discovered. Only in
California!)


John Higdon  |   P. O. Box 7648   |   [phone]     |       FAX:
[e-mail] | San Jose, CA 95150 | 10288 0 700 FOR-A-MOO | [phone]


------------------------------

From: [e-mail] (Ken Stox)
Subject: Re: FCC Proposed Ruling on Scanners That Receive Cellphones
Reply-To: [e-mail] (Ken Stox)
Organization: Ministry of Extremely Silly Walks
Date: Sat, 13 Feb 93 17:55:41 GMT


> 3. As defined in 47 CFR part 15 scanning receivers, or "scanners,"
> are radio receivers that automatically switch between four or more
> frequencies anywhere within the 30-960 Mhz band....

I guess my TV set will become illegal too!! It has a mode which
"scans" all channels to see if they are transmitting, and then when
you hit the channel up or down buttons only "active" stations will be
shown.

You know, that may not be such a bad idea... :->


#include <std_disclaimer.h>
Ken Stox     Consultant to A.C. Nielsen    [e-mail]

------------------------------

Date: Sat, 13 Feb 93 12:57 PST
From: [e-mail] (John Higdon)
Reply-To: John Higdon <[e-mail]>
Organization: Green Hills and Cows
Subject: Re: FCC Proposed Ruling on Scanners That Receive Cellphones


[e-mail] (Robert Paul Loeber) writes:

> I know this may sound like a dumb question ... but what is the FCCs
> definition of a "radio receiver"?

> get a cellular phone (instead of a scanner) ...

Also, what is the FCC's definition of "easily modified"? I happen to
know someone who has rewritten the firmware for a very popular
handheld phone to allow it to act as a powerful monitor of cellular
communications. Unlike an ordinary scanner, this modified product has
the ability to follow a conversation after a handoff.

Believe me, if I wanted to listen in on cellular traffic I would not
waste any time or effort with my venerable Yaesu. The [name withheld]
"Special Edition" cellular phone would be my weapon of choice!

Scanner laws will be just about as effective as gun laws -- only much
sillier. The FCC is seriously deluded if it thinks it can win a
technological war with anyone. The below-average moron outguns the FCC
in the brain cell department.


John Higdon  |   P. O. Box 7648   |   [phone]     |       FAX:
[e-mail] | San Jose, CA 95150 | 10288 0 700 FOR-A-MOO | [phone]

------------------------------

End of TELECOM Digest V13 #89
*****************************