Time for a new release announcement of TCP/IP for RSX-11M-PLUS.
Highlights:
. IP multicasting have been implemented
. TCP stability improvements
Detailed information on things that have been done since the last release:
IP:
. Added IP multicast support, and functions to enable this on UDP sockets.
UDP:
. Added functions for joining and leaving multicast groups on sockets.
TCP:
. Bugfix in TCP. Under some circumstances, TCP will stop receiving data
because of a calculation error on TCP sequence numbers.
. Improvement in TCP. Code accidentally sent unnecessary probes when a
socket is in Close Wait.
. Bugfix in TCP. Any ICMP error received for a socket caused the TCP
connection to close down. This should not happen for ICMP source quench
or ICMP timeout messages.
. Correct MSS computation and setup based on interface MTU.
IFCONFIG:
. Added ability to change MTU of interface in IFCONFIG.
FTPD:
. Bugfix in FTPD. Long home directory names caused FTPD to fail.
Some additional notes:
Some people might wonder why the multicast changes have been introduced,
but no other changes related to it. I wanted to get this change out now
in order to allow people the possibility to play with it, if anyone is
interested. For my own part, I next plan to look at mDNS, to allow RSX
to live in home networks without a proper DNS server, but still be
visible to other systems. mDNS depends on multicast groups.
In addition, the TCP corrections finally fixed some long standing
problems that I have been observing that have been very rare, but very
annoying to me. BQTCP is now behaving very well on the network, and I do
not actually have any known issues (at this time) that that are nagging
me. I hope this release will see a further reduction on work on
protocols like IP, TCP and UDP, and future work will be even more
focused on higher level protocols.
As usual, the distribution is available from:
ftp://mim.update.uu.se/bqtcp.dsk
ftp://mim.update.uu.se/bqtcp.tap
ftp://ftp.update.uu.se/pub/pdp11/rsx/tcpip/tcpip.dsk
The documentation is also available through ftp on Mim, or also at
http://mim.update.uu.se/tcpipdoc
I hope people find this update useful.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
NIA20 will be included along with an IP capable IMP interface.
On Feb 26, 2020 11:33 AM, Robert Armstrong <bob at jfcl.com> wrote:
>Rich Cornwell's simh KL10 is in the final stages of prep. You should see something
>in the next few weeks.
Really? I had no idea... Sounds like I am totally wrong about simh never emulating a KL.
Will it also emulate a DTE so you can marry it to one or more PDP-11 simhs? I know there was some talk of doing that for the PDP-15/76.
Bob
What are the alternatives for bridging DECnet systems across the Internet
that don't require any special hardware and can be self hosted on
OpenVMS/VAX?
I'm aware of Paul's Python router, but I didn't think it ran on a VAX. Am
I wrong about that? Is there a Python for OpenVMS/VAX? I believe there's a
port for AXP but I didn't know about one for VAX.
And I saw Rob mentioned his user mode DECnet router. Same questions -
what environment does that require? And I saw mention it being both
Multinet and simh DDCMp compatible - will it talk to a simh KS10/TOPS10/DMR
?
Bob
Hi all
I've moved my SIMH VAXen to a new host with a new version of SIMH (the
current 3.11 that I found - is there a newer version?)
Main host runs Devuan Linux (yes, I'm a systemd hater). VirtualBox VM
hosts the SIMH instance, also running Devuan Linux.
My area router, FRUGAL, boots fine and gets adjacency with my
Multinet-over-IP connections, that is, LEGATO, MIM, and 31.3:: as well as
other DECNet hosts (MISER, CHEAP)
However, the connection from FRUGAL "flaps" (going up and down) quite a
bit with local DECNet hosts as well as the ip-connection to 31.3::
I've tried eliminating the connection to 31.3:: as the culprit by turning
off the circuit, and the connection between FRUGAL:: and MISER:: (both
local nodes) continues to go up and down every few minutes.
I'd welcome some pointers, things to check, etc. Is there such a thing as
an adjacency timer perhaps I could crank up a bit?
Thanks
Fred
I just updated the current nodenames on a CISCO router after quite a
while using this. Posting this here in case this is useful for others
with CISCOs, possibly saving some awking and sedding.
This includes today's addition of A2RTR.
Best,
Supratim
--
Supratim Sanyal, W1XMT
39.19151 N, 77.23432 W
QCOCAL::SANYAL via HECnet
I have switched from using LEGATO as the area 2 router to using Paul's
pyDECnet router instead. I'm very impressed - it doesn't run under VMS, but
other than that pyDECnet does pretty much everything you could want to do
with DECnet or HECnet all in a single package. It only took me a couple of
hours to get it all set up, and it can manage all the Multinet links and the
one "Bilquist Bridge" link that were formerly attached to LEGATO.
The new router for area 2 is (what else?) A2RTR, 2.1023. Johnny, please
add this node name to your database.
LEGATO is still there, but it's been demoted to an L1 router and it
doesn't do anything now except route between HECnet and the real, physical,
machines in my house.
For the people who were/are connected to LEGATO there should be no real
difference except that you'll now see A2RTR on the other end of the link
instead. Everything else should work as before. Let me know if you find
some problem.
And a special thanks to Paul and Supratim for answering my silly questions
and helping me set it up.
Bob
I would not recomend DDCMP over TCP, as you have no idea what the
TCP implementaion does for timers etc and TCP is streams, you need
to identify the packet boundaries.
DDCMP over UDP gives DDCMP full control over retransmit timers etc
for making the connection reliable and what the actual netwrk is doing
is visible to the DDCMP layer.
-P
From: Peter Lothberg <roll at Stupi.SE>
>May I suggest sometng ugly, but not violating any layers,
>DDCMP over UDP.
For what it's worth, Ersatz-11 (my PDP-11 emulator) also supports DDCMP
over UDP, and I don't consider it ugly at all -- it's a very neat fit.
(E11 can also use TCP, or actual real async DDCMP over serial lines, and
there are the beginnings of support for DDCMP on sync lines too but
usable hardware for that is rare.)
John Wilson
D Bit
