Site Status
A corrected ARPANET host status report from September 1971.
McKenzie, A.M., "Site Status", RFC 240, DOI 10.17487/RFC0240, September 1971, <https://www.rfc-editor.org/info/rfc240>.
@misc{rfc240,
series = {Request for Comments},
number = 240,
howpublished = {RFC 240},
publisher = {RFC Editor},
doi = {10.17487/RFC0240},
url = {https://www.rfc-editor.org/info/rfc240},
author = {A.M. McKenzie},
title = {{Site Status}},
pagetotal = 4,
year = 1971,
month = sep,
} RFC text
Networking Group A. McKenzie Request for Comments: 240 BBN NIC #7665 30 September 1971 Categories: F, G.3 Rand 50ctu Updates: none Obsoletes: RFC 235 SITE STATUS Due to extensive typographic errors in RFC #235, I have decided to reissue the first of our experimental site status reports. In compiling RFC #235, I followed the principle that timeliness was somewhat more important than extremely careful editing; this seemed to me to be in keeping with the original RFC philosophy. I am sorry for any embarrassment this may have caused to anyone associated with the Network, and accept full responsibility for the content of RFC #235. I will try to do a better job in the future. Alex McKenzie SITE STATUS Beginning with this RFC, BBN will report on the status of most Network Hosts approximately once every two weeks. The information for these reports will be gained from talking to people at each site, and from experimental "data". These data will be the results of daily attempts to log into each of the Hosts which might be accessible to a Network user; the attempts will have been made from the BBN prototype Terminal IMP at a random time each weekday. Several Hosts are currently excluded from the daily testing. These Hosts fall into two categories: 1) Hosts which are note expected to be functioning on the Network as servers (available for use from other sites) for at least a month. Included here are: Network Site Computer Address 71 Rand PDP-10 74 Lincoln TX2 11 Stanford PDP-10 RFC #240 13 Case PDP-10 14 Carnegie PDP-10 15 Paoli B6500 2) Hosts which are currently intended to be users only. Included here are the Terminal IMPs presently in the Network (AMES, MITRE, and BBN[1]). This category also includes the Network Control Center computer (Network Address 5) which is use solely for gathering statistics from the Network. Finally, included among these Hosts are the following: Network Site Computer Address 7 Rand IBM-360/65 73 Harvard PDP-1 12 Illinois PDP-11 The tables on the next two pages condense the information on Host status for September 13 through September 24. PREDICTIONS NETWORK SITE COMPUTER STATUS or PREDICTION OBTAINED ADDRESS FROM 1 UCLA SIGMA - 7 Server Jon Postel 65 UCLA IBM-360/91 Remote Job Service now, Time-sharing in January Steve Wolf 2 SRI(NIC) PDP-10 October 11 John Melvin 66 SRI(AI) PDP-10 November Len Chaiten 3 UCSB IBM-360-75 Server Jim White 4 UTAH PDP-10 "Soon" Barry Wessler 5 BBN DDP-516 NCC Alex McKenzie 69 BBN PDP-10 Server Dan Murphy 6 MIT(Multics) H-645 "Soon" Mike Padlipsky 70 MIT(DM) PDP-10 Server Bob Bressler 7 RAND IBM 360/65 User only Eric Harslem 71 RAND PDP-10 January Eric Harslem 8 SDC IBM-360/67 October 11 Bob Long 9 HARVARD PDP-10 "Soon" Bob Sundberg 73 HARVARD PDP-1 User only Bob Sundberg 10 LINCOLN IBM-360/67 "Soon" Joel Winnet 74 LINCOLN TX2 Uncertain Tom Barkalow 11 STANFORD PDP-10 November Andy Moorer 12 ILLINOIS PDP-11 User only John Cravits 13 CASE PDP-10 December 15 Charles Rose 14 CARNEGIE PDP-10 January Hal VanZoeren 15 PAOLI B6500 Uncertain John Cravits RFC #240 16 AMES TIP Installed 17 MITRE TIP Installed 30 BBN TIP Prototype NETWORK SITE COMPUTER DATE AND TIME (P.M.) ADDRESS 9/13 9/14 9/15 9/16 9/17 9/20 4:30 3:30 6:00 10:30 1:30 12:30 1 UCLA SIGMA-7 O O O D D D 65[2] UCLA IBM 360/91 O O O O O D 2 SRI(NIC) PDP-10 D D D D D D 66 SRI(AI) PDP-10 D D D D D D 3 UCSB IBM-360/75 O O O O O O 4 UTAH PDP-10 D D D D D D 69 BBN PDP-10 O 1/2 0 O T 1/2 O O 6 MIT(Multics) H-645 R R R D 1/2 O D 70 MIT(DM) PDP-10 T T T O O T 8 SDC IBM-360/67 D D D D T D 9 HARVARD PDP-10 T D T T T T 10 LINCOLN IBM-360/67 D 1/2 0 1/2 0 D T D NETWORK SITE COMPUTER DATE AND TIME (P.M.) ADDRESS 9/21 9/22 9/23 9/24 4:30 3:30 2:00 5:00 1 UCLA SIGMA-7 O D T O 65[2] UCLA IBM 360/91 O O O O 2 SRI(NIC) PDP-10 D O D D 66 SRI(AI) PDP-10 D D D D 3 UCSB IBM-360/75 O O O O 4 UTAH PDP-10 D D D D 69 BBN PDP-10 O 1/2 O O O 6 MIT(Multics) H-645 1/2 O T T R 70 MIT(DM) PDP-10 1/2 O O D D 8 SDC IBM-360/67 D D D D 9 HARVARD PDP-10 D T D D 10 LINCOLN IBM-360/67 D D 1/2 0 D RFC #240 where D = Dead. (Destination Host either dead orinaccessible (due to network partitioning or local IMP failure) from the BBN Terminal IMP). R = Refused. (Destination Host returned a CLS to the initial RFC.) T = Timed Out. (Destination Host did not respond in any way to the initial RFC, although not dead.) 1/2 0 = 1/2 Open. (Destination Host opened a connection but then either immediately close it, or did not respond any further.) O = Open. (Destination Host opened a connection and was accessible to users.) Endnotes [1] The BBN Terminal IMP (Network Address 158) is a prototype, and as such is frequently not connected to the Network, but being used to refine and debug the Terminal IMP programs. [2] The UCLA IBM-360 is at the moment only able to handle Remote Job Service. BBN is not equipped to test this, but is assuming that receipt of their canned message indicates that RJS is also functioning. [ This RFC was put into machine readable form for entry ] [ into the online RFC archives by Roy Zimmer 9/97 ]
AI ELI5
On this page
§1Introduction
RFC 240 reissues the first ARPANET site status report from BBN. The original RFC 235 contained so many typographical errors that the author, Alex McKenzie, decided to republish it with corrections. BBN committed to issuing these status reports approximately every two weeks thereafter, gathering information from daily automated tests and contact with site operators.
A researcher had to say sorry for a report with too many typos and printed a clean copy instead.
§2Daily Testing and Host Status
BBN performed daily login tests from the BBN prototype Terminal IMP at a random time each weekday to see which hosts were reachable. The results used five status codes: O for Open (host responded and was accessible), D for Dead (unreachable), T for Timed Out (no response within the timeout), R for Refused (returned a close), and 1/2 O for 1/2 Open (opened a connection but then closed or stopped responding).
Some hosts were excluded from daily testing: those not expected to be network servers for at least a month, such as new PDP-10s at Rand, Stanford, Case, and Carnegie; and hosts intended as users only, including Terminal IMPs and the Network Control Center.
Testers called each computer every day and marked down if it answered, timed out, or refused the call. Some computers were too new or too special to test every day.
§3Host Predictions and Status Summary
For each host, the report included predictions about when it would become a network server, obtained from site operators. Most hosts reported status as either Server or User only. Availability varied widely: UCLA's SIGMA-7 and UCSB's IBM-360/75 were consistently open, while others like UTAH's PDP-10 and SRI's PDP-10 were frequently dead or timed out. MIT's Multics H-645 and Lincoln's IBM-360/67 showed intermittent connectivity.
Each computer's home site said when it might be ready to talk to others, and the tests showed which ones actually answered the calls.