[Info-vax] Reject Cause=Invalid Message Format, Erroneous Transport PDU='386200BD00000005'H

[Johnny Billquist]

On 2016-11-28 00:18, Supratim Sanyal wrote:
> Hi Johnny,
>
>> Could you please redesign that thing. I'm getting lots of logs about
> you
>> trying to access machines I control, on which you do not have access.
>
> Sorry, did not mean to flood logs. I am done, and will only run this
> once a week. There must a better design than brute-force like the
> following that I am doing now, will keep looking. Let me know if once-a-
> week flood is not acceptable either. Thanks.

Any regular operations causing log entries is bad. But it's not your NCP 
commands that are causing it. You seem to also be trying to read some 
file from each node. And that is what you are getting denied on. Atleast 
on my machines.

	Johnny

>
> $ DECNET_PHASE=0
> $! TRY PHASE IV NCP
> $ NCP TELL 'DECNETADDR' SHOW NODE 'DECNETADDR' STATUS
> $ NCPSTAT=$STATUS
> $ IF NCPSTAT .EQ. %X10000001
> $ THEN
> $    DECNET_PHASE=4
> $ ELSE
> $!   TRY PHASE V OSI NCL
> $    NCL SHOW NODE 'DECNETADDR'
> $    NCPSTAT=$STATUS
> $    IF NCPSTAT .EQ. %X10000001
> $    THEN
> $        DECNET_PHASE=5
> $    ENDIF
> $ ENDIF
>
> In article <o1ev0a$dd7$1 at Iltempo.Update.UU.SE>, bqt at softjar.se says...
>>
>> On 2016-11-27 14:51, Supratim Sanyal wrote:
>>>> You can not use NCP to get similar 'bogus' Phase IV information from
>>>> another Phase V node.
>>>
>>> Thanks, that is indeed the case. I can run NCP commands from RSX-11M-PLUS (NSP version = 4.0.0, Routing version = 2.0.0, Type = Endnode IV) to both VAXen (running DECnet Plus/OSI) but cannot get NCP to work between the two VAXen.
>>
>> To be correct:
>> There is a subset of NCP available on Phase V nodes, which can be used
>> to talk, in some limited form to Phase IV nodes.
>> If you want to talk between Phase V nodes, you need to use whatever
>> protocol they designed in Phase V, as it appears Phase V nodes do not
>> have a NICE listener.
>> Also, there is something slightly broken in Phase V with regards to
>> Phase IV compatibility in other aspects as well, as can be seen by my
>> reply of how a VMS and RSX box (both on phase IV) gives different errors
>> when trying to talk NICE to a Phase V node.
>>
>>> NCL does work between the two VAXen.
>>
>> That would be expected, yes.
>>
>>> I am trying to write a script to walk a mixed list of Phases IV and V nodes and get status, running from one of the DECnet Plus/OSI/Phase V nodes. This NCP for Phase IV and NCL for Phase V introduces a new twist.
>>
>> Could you please redesign that thing. I'm getting lots of logs about you
>> trying to access machines I control, on which you do not have access.
>>
>>> Is there a way to find out from a Phase V node what protocol another node is running (Phase IV or V) so that I can conditionally run NCP or NCL from the script?
>>
>> I can't answer that one, as I avoid Phase V. :-)
>>
>> 	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