Sent from mobile device that advertises itself for no good reason
On 29 May 2014, at 05:55, Johnny Billquist <bqt at softjar.se> wrote:
On 2014-05-28 20:19, Cory Smelosky wrote:
On Wed, 28 May 2014, Johnny Billquist wrote:
I would guess on a bad XXDP image, since XXDP requires almost nothing
working. If RSRS/E and RSX works then I can't imagine XXDP not working.
As soon as I figure out how the hell to write a disk image from RSTS/E
I'll try a new image. ;)
Can't MOUNT FOREIGN so I need to read more manuals it appears.
If you have RSX-11M-PLUS V4.6 up and running, it already have a tool for doing this. :-)
I can't define remote nodes under RSX though. Which means copying doesn't work.
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
--
DECtec mailing list
http://dectec.info
To unsubscribe from this list see page at: http://dectec.info/mailman/listinfo/dectec_dectec.info
On 2014-05-29 18:15, H Vlems wrote:
Johnny,
I just checked the HECnet bridge link and it is down.
Is something wrong at your end, did I miss something?
No. Nothing would appear broken on my side. Other links are working, and you are not disabled. I sent you a mail earlier today asking to check that the address I have is correct. Did you not get that?
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
"David R. Lennon" <dave at thelennons.net> writes:
Hi,
That macro version that Brian posted doesn't seem to be working
(no output) under OpenVMS 8.4 IA64 but this other version he posted
does:
https://mail.encompasserve.org/anon/htnotes/note?f1=DEC_NETWORKING&f2=1247.…
Hmm... I'm busy at the moment addressing pool (as in swimming, not memory)
issues. Opened the damned thing on Memorial day only to have the filter's
pump motor seize up. Just purchased a new one and I'm installing it, but I
need to run off to the hardware store. I checked email before heading out.
I'll take a look at the two examples later. It's been more than 2 decades
since I authored those programs and I haven't run Phase IV here in a decade
or more.
--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG
I speak to machines with the voice of humanity.
"Bob Armstrong" <bob at jfcl.com> writes:
Source code?
You mean the VMS fiche ? Sorry - at one time I had access to a set, but
no more...
Well, these days, I read it from CD/DVD. ;)
1 259 %SBTTL 'NML$PROCESS_NICE Main command processing routine'
1 260
1 261 GLOBAL ROUTINE NML$PROCESS_NICE (msg_desc, resp_rtn): NOVALUE =
1 262
1 263 !++
1 264 ! This routine is the main command processing routine. NICE messages
1 265 ! are parsed to determine the requested function and then the proper
1 266 ! routine is called to perform the function.
1 267 !
1 268 ! Inputs:
1 269 !
1 270 ! msg_desc = Address of descriptor of NICE message
1 271 ! resp_rtn = Address of action routine to call with NICE response
1 272 ! The action routine is called with the following arguments:
1 273 ! 1) Address of descriptor of NICE response
1 274 !
1 275 ! Outputs:
1 276 !
1 277 ! None - control is returned after the last response has been passed
1 278 ! to the action routine.
1 279 !--
If you Google NML$PROCESS_NICE
Did that last night :-) There's not a lot, but I did find Nick deSmith's
netname.c example, and one other little code fragment, which was helpful.
The "Callback routine" is handed a descriptor.
I saw that. I was really hoping that the callback routine received an
extra generic parameter (like a Windows lParam or a VMS AST parameter) so
that the caller of NML$PROCESS_NICE could pass arbitrary additional
arguments to the callback. It doesn't look like that happens, though, so if
the callback wants to store or save any of the data that it gets, then it
has to do it globally.
It's far from perfect, I agree.
library module "$NMADEF" in the system called LIB.
Got that. Actually on OVMS 7.3 I don't have SYS$LIB_C.TLB or nmadef.h -
from some of the web comments I found, it sounds like those might exist only
on Alpha. No matter, though - $NMADEF is in MACRO32 is in LIB and it's
simple to extract $NMADEF to a MAR file. A little bit of emacs magic turned
it into a .h.
Extract it and use SDL!
The real problem is that there are no comments in $NMADEF, and my NICE
spec (AA-X437A-TK) only gives numeric values for everything. It's kind of
"take your best guess" as to which named constant actually corresponds to
the item you want.
Avoid the nameless constant values; include NMADEF from the library and use
its constant definitions.
Agreed!
Call NML$INITIALIZE; no parameters.
Call NML$PROCESS_NICE; NML request by descriptor, callback routine
Call NML$TERMINATE; no parameters.
The only serious piece I'm missing at this point is, what do you do if you
want to talk to NML on another node besides your own (e.g. "TELL yoyo SHOW
EXEC CHAR") ? There doesn't seem to be any NICE function that can be
embedded in the data stream to do this, and that wouldn't really make sense
anyway. That'd be like asking the local NML to talk to the remote one on
your behalf, and I don't think that's the way it works.
It seems like there should be some parameter to either NML$INITIALIZE or
NML$PROCESS_NICE that gives the node and access control information, but I
don't find any reference to that.
Thanks again for all the help, Brian.
Read the NML spec. You can do all you want via NICE. You simply need to
get the information about the remote node as known by your node and then,
compose the NICE message with the DECnet address in the NICE message. The
rest of the code all looks the same.
--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG
I speak to machines with the voice of humanity.
Hi,
That macro version that Brian posted doesn't seem to be working (no output) under OpenVMS 8.4 IA64 but this other version he posted does:
https://mail.encompasserve.org/anon/htnotes/note?f1=DEC_NETWORKING&f2=1247.…
The code he posted does have this comment:
; Read function options (RTFM. Sec 6.8 of DNA NML Network Mgt Spec #AA-X437A-TK)
which is a manual officially titled "DECnet Digital Network Architecture Phase IV Network Management Functional Specification", AA-X437A-TK and a scan is available online at:
http://bitsavers.trailing-edge.com/www.computer.museum.uq.edu.au/pdf/AA-X43…
Regards,
Dave
Quoting "Brian Schenkenberger, VAXman-" <system at tmesis.com>:
"Bob Armstrong" <bob at jfcl.com> writes:
Hey, thanks, Brian, for the example! Where are these NML$ library
routines documented? That's what I was looking for...
Source code? It's been so long since I wrote that example that I really do
not recall where *that* info was found. I typically document my code with
a RTFM reference. Of course, the only thin in that didty I posted was the
DNA NML Network Management Specification. There are only 3 entry points in
NMLSHR.EXE which you can see with an ANALYZE/IMAGE command.
Call NML$INITIALIZE; no parameters.
Call NML$PROCESS_NICE; NML request by descriptor, callback routine
Call NML$TERMINATE; no parameters.
The "Callback routine" is handed a descriptor. The first byte of the data
in the descriptor is a status. The status values, along with all of the
rest of the NMA constants, can be found in the library module "$NMADEF" in
the system called LIB. SYS$LIBRARY:LIB.MLB for macro; LIB.REQ for Bliss;
SYS$LIB_C.TLB for C.
If you Google NML$PROCESS_NICE, you will find some other tidbits of info.
One that was found when I did this is by friend Nick De Smith. It can be
found in the DECUS Library Compendium (DECUSlib.COM) which I too maintain.
NDS's example is written in C. I'd suggest NOT following NDS's example
too closely. Avoid the nameless constant values; include NMADEF from the
library and use its constant definitions.
Also, these calls will only work if you have DECnet phase IV installed on
your system. DECnet Phase V interface is a whole 'nother ball o' wax.
--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG
I speak to machines with the voice of humanity.
Source code?
You mean the VMS fiche ? Sorry - at one time I had access to a set, but
no more...
If you Google NML$PROCESS_NICE
Did that last night :-) There's not a lot, but I did find Nick deSmith's
netname.c example, and one other little code fragment, which was helpful.
The "Callback routine" is handed a descriptor.
I saw that. I was really hoping that the callback routine received an
extra generic parameter (like a Windows lParam or a VMS AST parameter) so
that the caller of NML$PROCESS_NICE could pass arbitrary additional
arguments to the callback. It doesn't look like that happens, though, so if
the callback wants to store or save any of the data that it gets, then it
has to do it globally.
library module "$NMADEF" in the system called LIB.
Got that. Actually on OVMS 7.3 I don't have SYS$LIB_C.TLB or nmadef.h -
from some of the web comments I found, it sounds like those might exist only
on Alpha. No matter, though - $NMADEF is in MACRO32 is in LIB and it's
simple to extract $NMADEF to a MAR file. A little bit of emacs magic turned
it into a .h.
The real problem is that there are no comments in $NMADEF, and my NICE
spec (AA-X437A-TK) only gives numeric values for everything. It's kind of
"take your best guess" as to which named constant actually corresponds to
the item you want.
Avoid the nameless constant values; include NMADEF from the library and use
its constant definitions.
Agreed!
Call NML$INITIALIZE; no parameters.
Call NML$PROCESS_NICE; NML request by descriptor, callback routine
Call NML$TERMINATE; no parameters.
The only serious piece I'm missing at this point is, what do you do if you
want to talk to NML on another node besides your own (e.g. "TELL yoyo SHOW
EXEC CHAR") ? There doesn't seem to be any NICE function that can be
embedded in the data stream to do this, and that wouldn't really make sense
anyway. That'd be like asking the local NML to talk to the remote one on
your behalf, and I don't think that's the way it works.
It seems like there should be some parameter to either NML$INITIALIZE or
NML$PROCESS_NICE that gives the node and access control information, but I
don't find any reference to that.
Thanks again for all the help, Brian.
Bob
"Bob Armstrong" <bob at jfcl.com> writes:
Hey, thanks, Brian, for the example! Where are these NML$ library
routines documented? That's what I was looking for...
Source code? It's been so long since I wrote that example that I really do
not recall where *that* info was found. I typically document my code with
a RTFM reference. Of course, the only thin in that didty I posted was the
DNA NML Network Management Specification. There are only 3 entry points in
NMLSHR.EXE which you can see with an ANALYZE/IMAGE command.
Call NML$INITIALIZE; no parameters.
Call NML$PROCESS_NICE; NML request by descriptor, callback routine
Call NML$TERMINATE; no parameters.
The "Callback routine" is handed a descriptor. The first byte of the data
in the descriptor is a status. The status values, along with all of the
rest of the NMA constants, can be found in the library module "$NMADEF" in
the system called LIB. SYS$LIBRARY:LIB.MLB for macro; LIB.REQ for Bliss;
SYS$LIB_C.TLB for C.
If you Google NML$PROCESS_NICE, you will find some other tidbits of info.
One that was found when I did this is by friend Nick De Smith. It can be
found in the DECUS Library Compendium (DECUSlib.COM) which I too maintain.
NDS's example is written in C. I'd suggest NOT following NDS's example
too closely. Avoid the nameless constant values; include NMADEF from the
library and use its constant definitions.
Also, these calls will only work if you have DECnet phase IV installed on
your system. DECnet Phase V interface is a whole 'nother ball o' wax.
--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG
I speak to machines with the voice of humanity.
On 2014-05-28 20:19, Cory Smelosky wrote:
On Wed, 28 May 2014, Johnny Billquist wrote:
I would guess on a bad XXDP image, since XXDP requires almost nothing
working. If RSRS/E and RSX works then I can't imagine XXDP not working.
As soon as I figure out how the hell to write a disk image from RSTS/E
I'll try a new image. ;)
Can't MOUNT FOREIGN so I need to read more manuals it appears.
If you have RSX-11M-PLUS V4.6 up and running, it already have a tool for doing this. :-)
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
