FYI, LSSM/area 61 may experience a network outage tomorrow (Monday
July 6th) due to infrastructure work the upstream provider is conducting
in town. They've said that they expect the work will be complete by end
of the day (US Eastern time) tomorrow.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
Does anybody know hoe to.set up multinet for hecnet?
Begin forwarded message:
>
> From: Johnny Billquist <bqt(a)softjar.se>
> Subject: Re: [HECnet] Getting back online..
> Date: 27. May 2026 at 12.09.22 GMT+3
> To: Sampsa Laine <sampsa(a)mac.com>
>
> I'm no expert on Multnet under VMS. There is some tool or something, where you set the link up, telling what host and port it should connect to, and that you should be the one initiating the connection.
>
> You could ask on the mailing list about these details.
>
> Johnny
>
> On 2026-05-27 11:04, Sampsa Laine wrote:
>> I get the fiollowing from ncp:
>> NCP>sho known circuits
>> Known Circuit Volatile Summary as of 27-MAY-2026 09:02:24
>> Circuit State Loopback Adjacent
>> Name Routing Node
>> QNA-0 on
>> TCP-0-0 on -starting
>>> On 27. May 2026, at 11.49, Johnny Billquist <bqt(a)softjar.se> wrote:
>>>
>>> Hi.
>>> Ok, so set up your side to connect to hecnet.softjar.se, port 7717, using tcp.
>>> That should do it.
>>>
>>> Johnny
>>>
>>> On 2026-05-27 09:59, Sampsa Laine wrote:
>>>> Sorry to bug you but I could use the connection details,,,
>>>>> On 27. May 2026, at 8.41, Johnny Billquist <bqt(a)softjar.se> wrote:
>>>>>
>>>>> If you can wait a couple of hours. I'm about to leave home for work...
>>>>>
>>>>> Johnny
>>>>>
>>>>> On 2026-05-27 07:37, Sampsa Laine wrote:
>>>>>> oK address and port?
>>>>>> Sampsa
>>>>>>> On 27. May 2026, at 8.36, Johnny Billquist <bqt(a)softjar.se> wrote:
>>>>>>>
>>>>>>> Of course I can.
>>>>>>> But you really should be doing multinet with tcp and not udp...
>>>>>>>
>>>>>>> Johnny
>>>>>>>
>>>>>>> On 2026-05-27 03:14, Sampsa Laine wrote:
>>>>>>>> Can anyone peer with me over multinet/udp..?
>>>>>>>> _______________________________________________
>>>>>>>> HECnet mailing list -- hecnet(a)lists.dfupdate.se
>>>>>>>> To unsubscribe send an email to hecnet-leave(a)lists.dfupdate.se
>>>>>>>
>>>>>>> --
>>>>>>> Johnny Billquist || "I'm on a bus
>>>>>>> || on a psychedelic trip
>>>>>>> email: bqt(a)softjar.se || Reading murder books
>>>>>>> pdp is alive! || tryin' to stay hip" - B. Idol
>>>>>>>
>>>>>
>>>>> --
>>>>> Johnny Billquist || "I'm on a bus
>>>>> || on a psychedelic trip
>>>>> email: bqt(a)softjar.se || Reading murder books
>>>>> pdp is alive! || tryin' to stay hip" - B. Idol
>>>>>
>>>
>>> --
>>> Johnny Billquist || "I'm on a bus
>>> || on a psychedelic trip
>>> email: bqt(a)softjar.se || Reading murder books
>>> pdp is alive! || tryin' to stay hip" - B. Idol
>>>
>
> --
> Johnny Billquist || "I'm on a bus
> || on a psychedelic trip
> email: bqt(a)softjar.se || Reading murder books
> pdp is alive! || tryin' to stay hip" - B. Idol
>
Seems I might have a problem with my Qbus SCSI controller on my 11/93
(PONDUS::). It started acting up the other day, and things are not
running as they should now.
Anyone have one to spare? Happy to pay for it, but looking for something
less than Ebay if I can. Need one that does both disk and tape.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt(a)softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
On 2026-05-18 16:04, Johnny Billquist wrote:
> Thomas...
>
> On 5/18/26 20:37, Thomas DeBellis wrote:>
>> For better or worse, Tops-20 really does not support a ordinary user
>> program making any sort of a connection to an unnamed node. This is as
>> opposed to IP, where you can specify the address in dotted form
>> ([a.b.c.d], perhaps). I have mixed feelings about this as the NRT,
>> CTERM and (I believe) LAT host don't care and a SYSTAT will show
>> something like (2.512). FAL didn't used to care, either. Onthe other
>> hand, you can't make an outgoing connection to such a node, which
>> seems anti-social on the surface.
>
> This is no different than most other DECnet implementations. I believe
> VMS allows doing numeric hosts, but that's the only one, as far as I
> know.
> But there is a big difference in being able to do outgoing
> connections, and dealing with incoming connections.
> Now you're talking about outgoing connections, but your issue was with
> incoming ones.
RSTS/E is happy with numeric DECnet hosts:
$ dir 41.501::
Node: 41.501
User:
Password:
System Password:
Directory 41.501::SYS$SPECIFIC:[FAL$SERVER]
Name .Typ Size Prot Access Date Time UIC
NETSERVER.LOG;7 0/9 < 56> 18-May-26 18-May-26 16:19
[000376,000373]
NETSERVER.LOG;6 2/9 < 56> 23-Feb-26 23-Feb-26 14:51
[000376,000373]
NETSERVER.LOG;5 4/9 < 56> 23-Feb-26 23-Feb-26 07:25
[000376,000373]
NETSERVER.LOG;4 3/9 < 56> 22-Feb-26 22-Feb-26 21:21
[000376,000373]
Event type 33.0, Remote file access
Occurred 18-May-26 16:18:15.4 on node 41.509 (SPCPI2)
Access: Remote
Function: DIRECTORY
Remote node = 41.501 (41.501)
Remote process = 17 0 0
Local process = 0 20 254 NFT007
User =
File accessed = 0
NETSERVER.LOG;3 3/9 < 56> 22-Feb-26 22-Feb-26 19:04
[000376,000373]
NETSERVER.LOG;2 8/9 < 56> 22-Feb-26 22-Feb-26 18:09
[000376,000373]
NETSERVER.LOG;1 2/9 < 56> 22-Feb-26 22-Feb-26 17:55
[000376,000373]
Total of 22/63 blocks in 7 files.
$ set host 41.501::
Connection Established to VAX/VMS Node 41.501
Welcome to OpenVMS (TM) VAX Operating System, Version V7.3
Username: ^Z
Error reading command input
End of file detected
Control returned to node SPCPI2
This is a two part question.
I recently had FAL crash twice because of an ANONYMOUS request from a
node that I did not have yet defined (BV20:: [1.40]).  In general, I do
not feel comfortable providing ANONYMOUS services to unknown nodes, so a
lot of my software checks the remote node name. I wasn't doing that
check on the connection interrupt and the downstream code gracefully
handled that by doing the check itself and then crashing...
The first part of remediation is to not crash on an unknown node, and
therefore I need to reject such a connection while I figure out what to
do about all the rest of the code that assumes that there is a regular
batch job connecting to MIM:: to get the latest node list. FAL /really/
wants the node name...
The disconnect text is (obviously) "Unknown Node" which fortunately fits
in the 'generous' 16 byte disconnection message. The disconnect code
itself is something else. Which is better?
* DCX10 (decimal 10), INVALID NODE NAME
* DCX34 (decimal 34), ACCESS NOT PERMITTED
On one hand, no node name at all can be seen as a form of an invalid
node name. On the other hand, this is a policy decision to not allow
the access, so perhaps DCX34 is better?
The second part is what is the right thing to do. Again, I don't feel
comfortable providing service to an unknown node, but that's me: I'm an
Information Security Officer these days. On the other hand, maybe it's
a little fascist to be that strict, considering that this is a hobbyist
network.
What I'm considering is another RESTRICT keyword, UNKNOWN-NODES, which
would take either ALLOW or DISALLOW as selections. Or something like
that. I would have to make changes to all the downstream code that
would choke, maybe by phonying up a node name. Tops-20 does this in the
NRT server.
Email could not be made to work as Tops-20 does not allow connections to
node numbers.
