[ih] A laugh and a question

[David L. Mills]

Louis,

One of my enduring war stories with the Fuzzball was during the checkout 
of the GGP program written by Ginny Strazisar at BBN. I started the 
Fuzzball program with GGP running and noticed that first the gateway I 
was talking to crashed, then each of the other GGP gateways crashed in 
turn. I later discovered that my GGP update was flawed and that each GGP 
gateway that received it copied its neighbors and then expired.

Funny thing is the same kind of problem brought down the entire AT&T 
network in January 1990. Took them ten hours to fix the problem and 
bring the network back up. Problem was a misplaced "}" in one of the 3B2 
computer programs. An overloaded ESS switch in Manhattan rebooted and 
sent out a recovery message that rebooted its neighbor and the dance was 
on for all 114 switches in the network.

Dave

Louis Mamakos wrote:

> I'm pretty certain that the fuzzball PING existed before the PING that
> Mike wrote at BRL. I distinctly remember crashing one of their VAXen on
> the eve of the NCP/TCP transition with PING from one of my fuzzballs.
> There was a double-free() for the mbuf containing the ICMP echo. I
> don't recall if they were running the BBN stack, or an early 4.1BSD 
> version.
>
> In fact, I crashed it two or three times before I noticed the connection
> between my testing activities and er, the failure of my tests. This I
> confirmed in a conversation with Ron Natalie, then of BRL. "Ron, did
> your VAX just crash? ..uh, yeah, why do you ask?" This might have been
> BRL-VGR, but I might be imagining that.
>
> I also recall testing the ICMP code in the UNIVAC 1100 IP stack Mike
> Petry and I wrote during the same time, perhaps during 1981 - 1983, with
> the fuzzball PING command as well.
>
> louie
>
>
> Noel Chiappa wrote:
>
>> Interesting: RFC889 (December 1983) says:
>>
>> Among the various measurement packages is the original PING (Packet
>> InterNet Groper) program used over the last six years for numerous tests
>> and measurements of the Internet system and its client nets. This program
>> contains facilities to send various kinds of probe packets, including 
>> ICMP
>> Echo messages, process the reply and record elapsed times and other
>> information in a data file, as well as produce real-time snapshot
>> histograms and traces.
>>
>> The date on that RFC is very interesting, because Mike's Ping history 
>> page
>> (http://ftp.arl.mil/~mike/ping.html) says:
>>
>> My original impetus for writing PING for 4.2a BSD UNIX came from an 
>> offhand
>> remark in July 1983 by Dr. Dave Mills .. in which he described some work
>> that he had done on his "Fuzzball" LSI-11 systems to measure path latency
>> using timed ICMP Echo packets.
>> In December of 1983 I encountered some odd behavior of the IP network at
>> BRL. Recalling Dr. Mills' comments, I quickly coded up the PING program
>>
>> Presumably the work described in RFC889 predates December 1983 (the 
>> date on
>> the RFC); and that material was also what Mike referred to in his 
>> mention of
>> "described some work that he had done".
>>
>> So your "PING (Packet InterNet Groper)" program predates Mike's, and 
>> your use
>> of the name is at least contemporaneous, and probably pre-dates, 
>> Mike's. You
>> don't happen to have any older rough drafts of 889 lying around, do 
>> you? If
>> so, that could definitively show your use of the term pre-dating Mike's.
>>
>>
>> Just out of curiousity, I assume you had the sonar analogy in mind 
>> when you
>> came up with the backronym (?) "PING (Packet InterNet Groper)"?
>>
>> Do you recall if the term "ping" was already in use in the community 
>> at the
>> time, or was that something you introduced? (I just don't recall, alas!)
>>
>> Noel
>>