[HECnet] Oldest VAX/VMS for VAX-11/730 on HECNET

[Mark J. Blair]

> On Dec 9, 2021, at 4:09 AM, Johnny Billquist <bqt at softjar.se> wrote:
> 
> 
> 
> On 2021-12-09 08:29, Mark J. Blair wrote:
>>> On Dec 8, 2021, at 11:18 PM, Supratim Sanyal <supratim at riseup.net> wrote:
>>> 
>>> Looks like it. The lowest I could go so far is 31.31 (XLIV) @ 4.4. We should try 3.4+ with the key sometime.
>> I tried 3.5 with the license installed, and then upgraded it to 3.7 and tried again. It asks for an address in the range 1..1023, but I didn't see any way to specify an area. When I tried starting the network in any of them, I got errors like:
>>> $ @startnet
>>> %RUN-S-PROC_ID, identification of created process is 0001000E
>>> %OPCOM,  8-DEC-2021 18:33:36.97, message from user DECNET
>>> DECnet starting
>>> %NCP-I-NMLRSP, listener response - Invalid parameter value, Maximum address
>>> Executor node = 2.617 (PUGGLE)
>>> 
>>> %RUN-S-PROC_ID, identification of created process is 0001000C
>>> %NCP-I-NMLRSP, listener response - Invalid parameter value, Maximum address
>>> Executor node = 2.617 (PUGGLE)
> 
> Hmm. Interesting. I know that if you only give one number on a phase IV node, it will assume it's in the current area, or if area is sortof not defined, then it's in area 1.
> You seem to have it understand that it was in area 2. How did that happen?
> It would suggest that you managed to correctly enter a full phase IV address.

It prompted me for an address in the range 1-1023, but I wasn't paying close attention and I typed in "2.617". I thought that maybe it blindly accepted the string and got confused, so I ran NETCONFIG again and typed in "617". That gave me the same errors, but I didn't happen to copy the error messages at that stage.

> I'd guess it is the related to the error you got on 3.5 and 3.7. And that seems to have come from a machine with the address 2.699.

2.699 is my PyDECnet router that connects to Robert.

> I suspect you might get even 3.5 working. But you need to go into NCP and change a thing.
> 
> At least with RSX, you have a parameter that tells which is the highest node number it should handle, and if you see, or configure, a number higher than that, you'll get some kind of error similar to what you have described above.

Aha! I will look for the maximum address parameter and give it another try when I have a chance. Maybe that parameter was just set low by default in the VMS versions which gave me those errors.

If I can get a 3.x release running on HECNET, that would be neat.


-- 
Mark J. Blair, NF6X <nf6x at nf6x.net>
https://www.nf6x.net/