[HECnet] Disallow login on DZ line in VMS

[Johnny Billquist]

On 2021-12-21 03:00, Brian Schenkenberger, VAXman- wrote:
> Johnny Billquist <bqt at softjar.se> writes:
> 
>> That /NOTYPE_AHEAD would disable logins in VMS is a bit surprising.
>> In my eyes, that's very unintuitive. /NOINTERACTIVE seems much more
>> sensible (yay for RSTS/E).
> 
> When a terminal has no associated process and it receives unsolicited
> input, it forks into routine UNSOL in module TTYSUB.  UNSOL notifies
> the job controller of the occurrance by sending a message to the job
> cntroller's permanet mailbox.  This message contains the device name
> and precipitates the creation of a login process.  There are various
> other actions based upon the type of terminal and other attributes of
> the device that may occur prior to the actions of the job controller
> but without TYPE_AHEAD, none of this occurs.

Like I said - I do find it a bit unintuitive that all of this processing 
is so tied to the type ahead attribute.

>> Sounds like VMS also have /NOINTERACTIVE...?
> 
> VMS has no /NOINTERACTIVE.  Interactive is a process right assigned
> to a process that is instantiated via an unsolicited interrupt on a
> terminal device.

I did see someone else who claimed it existed, but wasn't documented. I 
haven't checked at all myself.

>> In RSX, /NOTYPE_AHEAD just means you don't have any typeahead. If you
>> try typing something when nothing is reading, the characters are just
>> thrown away. But if a read is in progress, things works just as normal.
> 
> I thought this was a question concerning VMS.

Well, I commented on Paul, who wrote what RSTS/E did, so it was already 
reflecting on more than VMS. But yes, it's a topic drift.

   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