11 Mar
2022
11 Mar
'22
4:40 p.m.
I'm a recent joiner to HECnet, and wondered if anyone else has played
around with Phase V software at all ?
Obviously HECnet itself is a Phase IV network, but you can connect Phase V
VAX endnodes using the Phase IV compatibility mode.
All my systems are simulated with simh, and getting an Ethernet connection
to work was very straightforward, there's just a bit more local
configuration required than with Phase IV.
However, I spent quite a few years at DEC working on VAX synchronous device
drivers, so I was interested to see whether I could get a WAN connection
going. The biggest issue was device support. DECnet-VAX Phase V doesn't
support any of the older microcoded interfaces (DMC11, DMV11, KMX11 etc),
and the later devices (DSV11, DSB32, DST32, etc) aren't supported by simh.
The only intersection is actually the DUP/DPV11, where there was a new VMS
driver (SEDRIVER) for Phase V, that replaces the
individual protocol-specific drivers that were previously used with these
devices.
I've spent quite a long time this week dredging stuff out of memory from 30
years ago, and still failing to get a DPV11 link to come up, until I
eventually realised that a) there are some significant differences between
the DUP and DPV, and b) simh does not support the DPV properly (how is one
supposed to know this?) ! So after changing my simulated microVAX into an
8600, and the DPV into a DUP, I've managed to get a DDCMP link to come up
between this and a PyDecnet router.
There still seem to be some rough edges, as it won't always come up until
PyDecnet is bounced, and it's also a bit slow for interactive use, as if
there's something that's only working on a retransmission, or timer expiry.
Anyway, it's been a fun nostalgia exercise to get this far, I'll probably
mess about with it some more. Let me know if you have any questions...
TOGATH$ mcr ncl sho routing circuit ddcmp-0 adj * all
Node 0 Routing Circuit DDCMP-0 Adjacency RTG$0001
at 2022-03-11-15:28:50.820+00:00Iinf
Identifiers
Name = RTG$0001
Status
Type = Autoconfigured
State = Up
Neighbor Node Type = Phase IV Router
Router NETs =
{
49::00-1D:AA-00-04-00-D5-75:00 (LOCAL:.TWPD)
}
TOGATH$ mcr ncl sho modem connect line ddcmp-0 all
Node 0 Modem Connect Line DDCMP-0
at 2022-03-11-15:29:33.660+00:00Iinf
Identifiers
Name = DDCMP-0
Status
UID = 5AFC9FC0-A14A-11EC-8002-AA000400C475
State = On
Interface State = Full Enabled
Actual Speed = 0
Loopback Mode = Null
Interface Type = Unknown
Modem Type = ""
Device Availability = Device Present
Request to Send = Asserted
Clear to Send = Asserted
Data Set Ready = Asserted
Data Terminal Ready = Asserted
Carrier Detect = Asserted
Signalling Rate Selector = Unknown
Signalling Rate Indicator = Unknown
Ring Indicator = Not Asserted
Remote Loopback = Not Asserted
Local Loopback = Not Asserted
Test Mode = Unknown
Characteristics
Communications Port = DPV-0-0
Connection Type = NonSwitched
Communications Mode = Synchronous
Duplex = Full
Profile = ""
Modem Control = Full
Modem Options =
{
}
Encoding = Normal
Clock = External
Speed = 0
Alternate Speed = 0
Rate Select = High
Maximum Enable Transmit Timer = 2000
Suppress Test Indicator = True
Transmit Holdoff Timer = 0
Carrier Loss Timer = 15000
Minimum DTR Deassertion Timer = 2000
Maximum Disable Transmit Timer = 500
Call Accept Timer = 0
Successful Call Indication Timer = 30
Maximum DSR Deassertion Timer = 5000
Counters
Creation Time = 2022-03-11-14:48:46.780+00:00Iinf
Losses of Clock = 0
Cable Faults = 0
Times DCE Not Ready = 0
Transmit Enable Timeouts = 4
Losses Of Carrier = 0
Rate Fallbacks = 0
Test Indications = 0
Framing Errors = 0
Times Cable Detected = 0
Device Errors = 0
Times Reset = 4
c
CTF V1.0-00 Page 1
Trace started on 11-MAR-2022 15:45:28.40 Analyzed on 11-MAR-2022
15:45:56.95
Trace File DUA1:[TREVOR]CTF$TRACE.DAT;3 Output File DUA1:[TREVOR]XX.TXT;4
-----------+----+-----+<--------DDCMP
Frame-------->+--------------------------`
Time |Evnt|Data | Typ QS Res Num Ctl Adr Count|Data
`
hh mm ss cc| |Size | |
`
-----------+----+-----+<--------------------------->+--------------------------`
15:45:28.40| Tx| 85| DAT S 77 86 1 77| 0A C5 75 C4 75 01 21
35 E`
| 01 04 C3 02 EE 81 C6
01 2`
| 2E 30 85 02 03 E8 C4
01 0`
15:45:28.41|Rx | 22| DAT 85 78 1 14| 05 D5 75 0A AA AA AA
AA A`
15:45:28.56|Rx | 6| ACK 86 1 |
15:45:28.57| Tx| 6| ACK S 78 1 |
15:45:32.80| Tx| 85| DAT S 78 87 1 77| 0A C5 75 C4 75 01 21
35 E`
| 01 04 C3 02 EE 81 C6
01 2`
| 2E 30 85 02 03 E8 C4
01 0`
15:45:32.81|Rx | 62| DAT 86 79 1 54| 02 C4 75 C5 75 02 21
2E D`
| 01 04 C6 01 23 01 11
30 3`
| 03 E8
15:45:32.81| Tx| 25| DAT S 79 88 1 17| 02 C5 75 C4 75 01 21
09 6`
15:45:32.82|Rx | 6| ACK 87 1 |
15:45:32.94|Rx | 6| ACK 88 1 |
15:45:32.95|Rx | 62| DAT 88 80 1 54| 02 C4 75 C5 75 02 21
2E D`
| 01 04 C6 01 23 01 11
30 3`
| 03 E8
15:45:33.07| Tx| 6| ACK S 80 1 |
15:45:38.48|Rx | 31| DAT 88 81 1 23| 02 C4 75 C5 75 02 21
07 F`
15:45:38.48| Tx| 40| DAT S 81 89 1 32| 02 C5 75 C4 75 01 21
07 F`
| 00 00 00 00 00 00
15:45:38.49|Rx | 31| DAT 88 82 1 23| 02 C4 75 C5 75 02 21
07 F`
15:45:38.49| Tx| 25| DAT S 82 90 1 17| 02 C5 75 C4 75 01 21
09 6`
15:45:38.50|Rx | 6| ACK 89 1 |
15:45:38.60|Rx | 6| ACK 90 1 |
15:45:39.72|Rx | 58| DAT 90 83 1 50| 02 C4 75 C5 75 02 21
07 F`
| 00 00 00 00 00 00 01
02 1`
15:45:39.72| Tx| 25| DAT S 83 91 1 17| 02 C5 75 C4 75 01 21
09 6`
15:45:39.73| Tx| 25| DAT S 83 92 1 17| 02 C5 75 C4 75 01 21
09 6`
15:45:39.73| Tx| 80| DAT S 83 93 1 72| 02 C5 75 C4 75 01 21
07 F`
| 00 00 00 00 00 00 01
02 F`
| A0 13 00 30 02 30 60
F9 0`
15:45:39.73| Tx| 38| DAT S 83 94 1 30| 02 C5 75 C4 75 01 21
07 F`
| 00 00 00 00
15:45:39.73| Tx| 49| DAT S 83 95 1 41| 02 C5 75 C4 75 01 21
07 F`
| 00 18 00 42 70 50 00
A0 1`
Buffer: XX.TXT | Write | Insert |
Forward
11 Mar
11 Mar
4:47 p.m.
On Mar 11, 2022, at 10:40 AM, Trevor Warwick
<twarwick(a)gmail.com> wrote:
I'm a recent joiner to HECnet, and wondered if anyone else has played around with
Phase V software at all ?
Obviously HECnet itself is a Phase IV network, but you can connect Phase V VAX endnodes
using the Phase IV compatibility mode.
All my systems are simulated with simh, and getting an Ethernet connection to work was
very straightforward, there's just a bit more local configuration required than with
Phase IV.
However, I spent quite a few years at DEC working on VAX synchronous device drivers, so I
was interested to see whether I could get a WAN connection going. The biggest issue was
device support. DECnet-VAX Phase V doesn't support any of the older microcoded
interfaces (DMC11, DMV11, KMX11 etc), and the later devices (DSV11, DSB32, DST32, etc)
aren't supported by simh. The only intersection is actually the DUP/DPV11, where there
was a new VMS driver (SEDRIVER) for Phase V, that replaces the individual
protocol-specific drivers that were previously used with these devices.
I've spent quite a long time this week dredging stuff out of memory from 30 years
ago, and still failing to get a DPV11 link to come up, until I eventually realised that a)
there are some significant differences between the DUP and DPV, and b) simh does not
support the DPV properly (how is one supposed to know this?) ! So after changing my
simulated microVAX into an 8600, and the DPV into a DUP, I've managed to get a DDCMP
link to come up between this and a PyDecnet router.
There still seem to be some rough edges, as it won't always come up until PyDecnet is
bounced, and it's also a bit slow for interactive use, as if there's something
that's only working on a retransmission, or timer expiry. Anyway, it's been a fun
nostalgia exercise to get this far, I'll probably mess about with it some more. Let me
know if you have any questions...
Could you send me traces of the PyDECnet issues you're seeing? I have never seen a
VMS DDCMP, but I've run the PyDECnet DDCMP against several others and it all works
without the sort of troubles you mentioned. I'd like to understand what the problem
is and if PyDECnet either needs a bugfix, or a workaround for a VMS bug.
Some day I may do at least partial Phase V in PyDECnet. Not soon, it's a big job.
paul
4:53 p.m.
Could you send me traces of the PyDECnet issues you're seeing? I have
never seen a VMS DDCMP, but I've run the PyDECnet DDCMP against several
others and it all works without the sort of troubles you mentioned. I'd
like to understand what the problem is and if PyDECnet either needs a
bugfix, or a workaround for a VMS bug.
Some day I may do at least partial Phase V in PyDECnet. Not soon, it's a
big job.
paul
Hi Paul,
Yes, I will let you know after I've looked into it further. My gut feel is
it's something in the TCP session setup (maybe just a config issue) between
simh and PyDecnet.
Trevor
4:57 p.m.
On Mar 11, 2022, at 10:53 AM, Trevor Warwick
<twarwick(a)gmail.com> wrote:
Could you send me traces of the PyDECnet issues you're seeing? I have never seen a
VMS DDCMP, but I've run the PyDECnet DDCMP against several others and it all works
without the sort of troubles you mentioned. I'd like to understand what the problem
is and if PyDECnet either needs a bugfix, or a workaround for a VMS bug.
Some day I may do at least partial Phase V in PyDECnet. Not soon, it's a big job.
paul
Hi Paul,
Yes, I will let you know after I've looked into it further. My gut feel is it's
something in the TCP session setup (maybe just a config issue) between simh and PyDecnet.
Trevor
You could try UDP to see if that makes a difference, but both should work without
trouble.
paul
5:09 p.m.
Speaking of VMS and DECnet, I did a fix yesterday to the routing layer
in RSX in order to work with older versions of VMS that have a bug
preventing connectivity to be established.
Basically, the problem is that when the link is being established, both
sides exchange some information, which contains things like the link
maximum buffer size. Commonly this is 576 bytes. But if you announce a
larger buffer size, the remote end can use that if it wants to. This is
separate from the segment buffer size, which is used by DECnet itself
when communicating.
However, older versions of VMS go bonkers if the link maximum size is
larger than some value. My guess would be if the size is larger than the
local link maximum buffer size, but it might be if it's larger than the
segment buffer size as well, or something complete else. I have not
explored this deeply.
But I changed RSX to instead of announcing the actual link maximum
buffer size, it will announce the segment buffer size.
This will cause the routing layer to split packets up in smaller parts
when it might not have needed to, but apart from that, things will work.
Newer versions of VMS do not have this problem, but it's observable on
VMS V5.4 at least.
The effect if the problem is hit is that the other side will announce
the circuit to come up, followed pretty soon by it being down again, and
this will repeat for ever. And the link never actually becomes operational.
The reason I've observed this, done the patch, and have the problem is
in the end because of a "problem" in the Qbus ethernet controllers. It's
a known problem, which is that if the controller runs out of buffers in
the middle of a receive which requires multiple buffers, the controller
will never cause an interrupt, and the host will not know that reception
of packets have stopped. (And reception have stopped since there are no
free buffers...)
This is usually not a problem for DECnet with 576 byte buffers, since
the segment size is 576 bytes. However, with TCP/IP, you normally have a
MTU of 1500. Which will require 3 buffers to receive full sized packets.
So I've seen this happen lots of times on loaded systems with high
TCP/IP packet rates. Which is what started my investigation.
The obvious solution to all of this is to instead tell DECnet to use
1500 byte buffers. That way you will never have a packet that requires
more than one buffer, and you will not get the hung ethernet controller.
But when you tell DECnet to use 1500 byte buffers, the routing layer
will then tell that the link maximum size is 1500, which then leads to
the problem mentioned above.
And of course, I know of one VMS V5.4 host on HECnet, to which Multinet
links then started failing, while it was working fine pretty much
everywhere else.
So I had to not use large buffers, and instead artificially try and tell
TCP/IP to not use large packets, in order to try and avoid ethernet
controller stops. (Oh, the joy!)
Anyway, I'm happy to tell that I've now included a "fixed" version of
the DECnet routing module for RSX so that this now works, with a slight
potential penalty for DECnet.
Johnny
On 2022-03-11 16:47, Paul Koning wrote:
--
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 Mar 11, 2022, at 10:40 AM, Trevor Warwick
<twarwick(a)gmail.com> wrote:
I'm a recent joiner to HECnet, and wondered if anyone else has played around with
Phase V software at all ?
Obviously HECnet itself is a Phase IV network, but you can connect Phase V VAX endnodes
using the Phase IV compatibility mode.
All my systems are simulated with simh, and getting an Ethernet connection to work was
very straightforward, there's just a bit more local configuration required than with
Phase IV.
However, I spent quite a few years at DEC working on VAX synchronous device drivers, so I
was interested to see whether I could get a WAN connection going. The biggest issue was
device support. DECnet-VAX Phase V doesn't support any of the older microcoded
interfaces (DMC11, DMV11, KMX11 etc), and the later devices (DSV11, DSB32, DST32, etc)
aren't supported by simh. The only intersection is actually the DUP/DPV11, where there
was a new VMS driver (SEDRIVER) for Phase V, that replaces the individual
protocol-specific drivers that were previously used with these devices.
I've spent quite a long time this week dredging stuff out of memory from 30 years
ago, and still failing to get a DPV11 link to come up, until I eventually realised that a)
there are some significant differences between the DUP and DPV, and b) simh does not
support the DPV properly (how is one supposed to know this?) ! So after changing my
simulated microVAX into an 8600, and the DPV into a DUP, I've managed to get a DDCMP
link to come up between this and a PyDecnet router.
There still seem to be some rough edges, as it won't always come up until PyDecnet is
bounced, and it's also a bit slow for interactive use, as if there's something
that's only working on a retransmission, or timer expiry. Anyway, it's been a fun
nostalgia exercise to get this far, I'll probably mess about with it some more. Let me
know if you have any questions...
Could you send me traces of the PyDECnet issues you're seeing? I have never seen a
VMS DDCMP, but I've run the PyDECnet DDCMP against several others and it all works
without the sort of troubles you mentioned. I'd like to understand what the problem
is and if PyDECnet either needs a bugfix, or a workaround for a VMS bug.
Some day I may do at least partial Phase V in PyDECnet. Not soon, it's a big job.
paul
_______________________________________________
HECnet mailing list -- hecnet(a)lists.dfupdate.se
To unsubscribe send an email to hecnet-leave(a)lists.dfupdate.se
4:51 p.m.
I know there are phase V nodes on HECnet, and I don't think you need to
pretend to be an endnode either.
But with that said, I know that phase V in general just makes life more
miserable, and I know some people "upgraded" back to phase IV...
As for DPV, it might just be that nobody have tested it. It's not a
device that gets much attention. I think Paul might have played with the
DUP11 recently, which might have helped get that into shape...
Johnny
On 2022-03-11 16:40, Trevor Warwick wrote:
I'm a recent joiner to HECnet, and wondered if anyone else has played
around with Phase V software at all ?
Obviously HECnet itself is a Phase IV network, but you can connect Phase
V VAX endnodes using the Phase IV compatibility mode.
All my systems are simulated with simh, and getting an Ethernet
connection to work was very straightforward, there's just a bit more
local configuration required than with Phase IV.
However, I spent quite a few years at DEC working on VAX synchronous
device drivers, so I was interested to see whether I could get a WAN
connection going. The biggest issue was device support. DECnet-VAX
Phase V doesn't support any of the older microcoded interfaces (DMC11,
DMV11, KMX11 etc), and the later devices (DSV11, DSB32, DST32, etc)
aren't supported by simh. The only intersection is actually the
DUP/DPV11, where there was a new VMS driver (SEDRIVER) for Phase V, that
replaces the individual protocol-specific drivers that were previously
used with these devices.
I've spent quite a long time this week dredging stuff out of memory from
30 years ago, and still failing to get a DPV11 link to come up, until I
eventually realised that a) there are some significant differences
between the DUP and DPV, and b) simh does not support the DPV properly
(how is one supposed to know this?) ! So after changing my simulated
microVAX into an 8600, and the DPV into a DUP, I've managed to get a
DDCMP link to come up between this and a PyDecnet router.
There still seem to be some rough edges, as it won't always come up
until PyDecnet is bounced, and it's also a bit slow for interactive use,
as if there's something that's only working on a retransmission, or
timer expiry. Anyway, it's been a fun nostalgia exercise to get this
far, I'll probably mess about with it some more. Let me know if you have
any questions...
TOGATH$ mcr ncl sho routing circuit ddcmp-0 adj * all
Node 0 Routing Circuit DDCMP-0 Adjacency RTG$0001
at 2022-03-11-15:28:50.820+00:00Iinf
Identifiers
Name = RTG$0001
Status
Type = Autoconfigured
State = Up
Neighbor Node Type = Phase IV Router
Router NETs =
{
49::00-1D:AA-00-04-00-D5-75:00 (LOCAL:.TWPD)
}
TOGATH$ mcr ncl sho modem connect line ddcmp-0 all
Node 0 Modem Connect Line DDCMP-0
at 2022-03-11-15:29:33.660+00:00Iinf
Identifiers
Name = DDCMP-0
Status
UID =
5AFC9FC0-A14A-11EC-8002-AA000400C475
State = On
Interface State = Full Enabled
Actual Speed = 0
Loopback Mode = Null
Interface Type = Unknown
Modem Type = ""
Device Availability = Device Present
Request to Send = Asserted
Clear to Send = Asserted
Data Set Ready = Asserted
Data Terminal Ready = Asserted
Carrier Detect = Asserted
Signalling Rate Selector = Unknown
Signalling Rate Indicator = Unknown
Ring Indicator = Not Asserted
Remote Loopback = Not Asserted
Local Loopback = Not Asserted
Test Mode = Unknown
Characteristics
Communications Port = DPV-0-0
Connection Type = NonSwitched
Communications Mode = Synchronous
Duplex = Full
Profile = ""
Modem Control = Full
Modem Options =
{
}
Encoding = Normal
Clock = External
Speed = 0
Alternate Speed = 0
Rate Select = High
Maximum Enable Transmit Timer = 2000
Suppress Test Indicator = True
Transmit Holdoff Timer = 0
Carrier Loss Timer = 15000
Minimum DTR Deassertion Timer = 2000
Maximum Disable Transmit Timer = 500
Call Accept Timer = 0
Successful Call Indication Timer = 30
Maximum DSR Deassertion Timer = 5000
Counters
Creation Time = 2022-03-11-14:48:46.780+00:00Iinf
Losses of Clock = 0
Cable Faults = 0
Times DCE Not Ready = 0
Transmit Enable Timeouts = 4
Losses Of Carrier = 0
Rate Fallbacks = 0
Test Indications = 0
Framing Errors = 0
Times Cable Detected = 0
Device Errors = 0
Times Reset = 4
c
CTF V1.0-00 Page 1
Trace started on 11-MAR-2022 15:45:28.40 Analyzed on 11-MAR-2022
15:45:56.95
Trace File DUA1:[TREVOR]CTF$TRACE.DAT;3 Output File DUA1:[TREVOR]XX.TXT;4
-----------+----+-----+<--------DDCMP
Frame-------->+--------------------------`
Time |Evnt|Data | Typ QS Res Num Ctl Adr Count|Data
`
hh mm ss cc| |Size | |
`
-----------+----+-----+<--------------------------->+--------------------------`
15:45:28.40| Tx| 85| DAT S 77 86 1 77| 0A C5 75 C4 75 01
21 35 E`
| 01 04 C3 02 EE 81
C6 01 2`
| 2E 30 85 02 03 E8
C4 01 0`
15:45:28.41|Rx | 22| DAT 85 78 1 14| 05 D5 75 0A AA AA
AA AA A`
15:45:28.56|Rx | 6| ACK 86 1 |
15:45:28.57| Tx| 6| ACK S 78 1 |
15:45:32.80| Tx| 85| DAT S 78 87 1 77| 0A C5 75 C4 75 01
21 35 E`
| 01 04 C3 02 EE 81
C6 01 2`
| 2E 30 85 02 03 E8
C4 01 0`
15:45:32.81|Rx | 62| DAT 86 79 1 54| 02 C4 75 C5 75 02
21 2E D`
| 01 04 C6 01 23 01
11 30 3`
| 03 E8
15:45:32.81| Tx| 25| DAT S 79 88 1 17| 02 C5 75 C4 75 01
21 09 6`
15:45:32.82|Rx | 6| ACK 87 1 |
15:45:32.94|Rx | 6| ACK 88 1 |
15:45:32.95|Rx | 62| DAT 88 80 1 54| 02 C4 75 C5 75 02
21 2E D`
| 01 04 C6 01 23 01
11 30 3`
| 03 E8
15:45:33.07| Tx| 6| ACK S 80 1 |
15:45:38.48|Rx | 31| DAT 88 81 1 23| 02 C4 75 C5 75 02
21 07 F`
15:45:38.48| Tx| 40| DAT S 81 89 1 32| 02 C5 75 C4 75 01
21 07 F`
| 00 00 00 00 00 00
15:45:38.49|Rx | 31| DAT 88 82 1 23| 02 C4 75 C5 75 02
21 07 F`
15:45:38.49| Tx| 25| DAT S 82 90 1 17| 02 C5 75 C4 75 01
21 09 6`
15:45:38.50|Rx | 6| ACK 89 1 |
15:45:38.60|Rx | 6| ACK 90 1 |
15:45:39.72|Rx | 58| DAT 90 83 1 50| 02 C4 75 C5 75 02
21 07 F`
| 00 00 00 00 00 00
01 02 1`
15:45:39.72| Tx| 25| DAT S 83 91 1 17| 02 C5 75 C4 75 01
21 09 6`
15:45:39.73| Tx| 25| DAT S 83 92 1 17| 02 C5 75 C4 75 01
21 09 6`
15:45:39.73| Tx| 80| DAT S 83 93 1 72| 02 C5 75 C4 75 01
21 07 F`
| 00 00 00 00 00 00
01 02 F`
| A0 13 00 30 02 30
60 F9 0`
15:45:39.73| Tx| 38| DAT S 83 94 1 30| 02 C5 75 C4 75 01
21 07 F`
| 00 00 00 00
15:45:39.73| Tx| 49| DAT S 83 95 1 41| 02 C5 75 C4 75 01
21 07 F`
| 00 18 00 42 70 50
00 A0 1`
Buffer: XX.TXT | Write | Insert |
Forward
_______________________________________________
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
5:18 p.m.
The only intersection is actually the
DUP/DPV11, where there was a new VMS driver (SEDRIVER) for Phase V ...
Sorry to be off topic, but I have a real MicroVAX-III with a DPV11 and I didn't
think it was supported by VMS at all. Does it work with DECnet Phase IV as well, or just
Phase V?
Of course my next problem will be to find something to connect it to :)
Bob
5:52 p.m.
I'm fairly sure that before Phase V, the DPV11 was only supported by VAX
PSI and VMS/SNA, not Decnet. It used to have a different driver for
whatever protocol you wanted to use, and the DUP11 drivers were different
again. In Phase V we produced a single multiprotocol
(DDCMP/HDLC/SDLC/Bisync) driver that supported both DPV and DUP.
On Fri, 11 Mar 2022 at 16:18, Robert Armstrong <bob(a)jfcl.com> wrote:
The only intersection is actually the
DUP/DPV11, where there was a new VMS driver (SEDRIVER) for Phase V ...
Sorry to be off topic, but I have a real MicroVAX-III with a DPV11 and I
didn't think it was supported by VMS at all. Does it work with DECnet
Phase IV as well, or just Phase V?
Of course my next problem will be to find something to connect it to :)
Bob
_______________________________________________
HECnet mailing list -- hecnet(a)lists.dfupdate.se
To unsubscribe send an email to hecnet-leave(a)lists.dfupdate.se
6:53 p.m.
But Phase V includes HDLC as a supported data link protocol, right? For that matter,
while PSI support tends to be a separate product it was, architecturally speaking, part of
DECnet.
paul
On Mar 11, 2022, at 11:52 AM, Trevor Warwick
<twarwick(a)gmail.com> wrote:
I'm fairly sure that before Phase V, the DPV11 was only supported by VAX PSI and
VMS/SNA, not Decnet. It used to have a different driver for whatever protocol you wanted
to use, and the DUP11 drivers were different again. In Phase V we produced a single
multiprotocol (DDCMP/HDLC/SDLC/Bisync) driver that supported both DPV and DUP.
On Fri, 11 Mar 2022 at 16:18, Robert Armstrong <bob(a)jfcl.com
<mailto:bob@jfcl.com>> wrote:
The only intersection is actually the
DUP/DPV11, where there was a new VMS driver (SEDRIVER) for Phase V ...
Sorry to be off topic, but I have a real MicroVAX-III with a DPV11 and I didn't
think it was supported by VMS at all. Does it work with DECnet Phase IV as well, or just
Phase V?
Of course my next problem will be to find something to connect it to :)
Bob
_______________________________________________
HECnet mailing list -- hecnet(a)lists.dfupdate.se <mailto:hecnet@lists.dfupdate.se>
To unsubscribe send an email to hecnet-leave(a)lists.dfupdate.se
<mailto:hecnet-leave@lists.dfupdate.se>
_______________________________________________
HECnet mailing list -- hecnet(a)lists.dfupdate.se
To unsubscribe send an email to hecnet-leave(a)lists.dfupdate.se
782
days inactive
782
days old
8 comments
4 participants
participants (4)
-
Johnny Billquist -
Paul Koning -
Robert Armstrong -
Trevor Warwick