[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[ih] inter-network communication history

On 8 Nov 2019 at 11:16, Jack Haverty via Internet-history wrote:

> Some other factoids I recalled:
> ??? - The original ARPANET management software at BBN was called
> "U",
> for Utilities.? Later we developed a successor tool, cleverly called
> "NU", for New Utilities.?? I'm pretty sure that the Internet
> management
> was integrated into NU when the NOC assumed 24x7 operations
> responsibility.

'U' I think was the system that Steve Butterfield hacked up on TENEX.  Before 
that there was the PDP-1 and after that a standalone management system on a 
spare 316 in the computer room.  Those systems had a "light box" showing which 
lines and IMPs were up and down.   The 316 system also had a hack where it 
understood the network topology and so when the network separated,  it figured 
out the IMP(s) and LINE(s) potentially responsible for the outage and suppressed 
the tidal wave of IMP down and LINE down reports.  It allowed the NOC folk to 
know what to look at/whom to call.  [it also suppressed the tidal wave of IMP 
UP and LINE UP messages when the network was reconnected].  The most 
handy case was if something happened to IMP 5.  Instead of reporting *every* 
IMPand Line down, it just say "IMP 5 down".

And of course we downloaded patches [since at that time the system was being 
assembled on the PDP-1].

There was no MIB and nothing fancy: the IMPs reported their status every 56{?} 
seconds and it was from those packets the up/down/connectivity was figured 
out.  and patches and other investigations were just done via a reasonable 
interface to the on-IMP DDT.

When TENEX got its ARPAnet connection, Steve hacked up a package that kinda 
worked like the stuff we had cobbled up.  I have *no* idea when the IMP sources 
moved to TENEX nor who cobbled up an assembler for them there. [I'm 
assuming, withouth knowing, that they _did_ get moved to TENEX].   

            Bernie Cosell
       bernie at fantasyfarm.com
-- Too many people; too few sheep --

Internet-history mailing list
Internet-history at elists.isoc.org