Whatever your opinion of Linux, it is significant as a server OS. The RFC
for its removal says:

> Decnet is an obsolete network protocol that receives more attention
> from kernel janitors than users. It belongs in computer protocol
> history museum not in Linux kernel.
>
> It has been Orphaned in kernel since 2010.
> And the documentation link on Sourceforge says it is abandoned there.
>
> Leave the UAPI alone to keep userspace programs compiling.

<https://lore.kernel.org/lkml/20220731190646.97039-1-stephen@networkplumbe
r.org/>

I don't know how well VMS works as an NFS client or server. When I was
using VMS in the late 1990s, it wasn't great, but a lot of time has
passed since then. If anyone wants Linux to carry on supporting DECnet,
there will need to be credible volunteers to maintain it.

John

On 08/02/22 10:41, John Dallman wrote:
> Whatever your opinion of Linux, it is significant as a server OS. The RFC
> for its removal says:
>
>> Decnet is an obsolete network protocol that receives more attention
>> from kernel janitors than users. It belongs in computer protocol
>> history museum not in Linux kernel.
>>
>> It has been Orphaned in kernel since 2010.
>> And the documentation link on Sourceforge says it is abandoned there.
>>
>> Leave the UAPI alone to keep userspace programs compiling.
>
> <https://lore.kernel.org/lkml/20220731190646.97039-1-stephen@networkplumbe
> r.org/>
>
> I don't know how well VMS works as an NFS client or server. When I was
> using VMS in the late 1990s, it wasn't great, but a lot of time has
> passed since then. If anyone wants Linux to carry on supporting DECnet,
> there will need to be credible volunteers to maintain it.
>
> John

First question must be: Is it built into the kernel or is it, as it
should be, a loadable module ?. If it's a loadable module,
then it should be maintainable as a separate entity, providing all
the interface requirements are met...

Chris

On 2022-08-02, John Dallman <jgd@cix.co.uk> wrote:
> Whatever your opinion of Linux, it is significant as a server OS. The RFC
> for its removal says:
>
>> Decnet is an obsolete network protocol that receives more attention
>> from kernel janitors than users. It belongs in computer protocol
>> history museum not in Linux kernel.
>>

Best description of DECnet Phase IV I have ever seen. :-)

>> It has been Orphaned in kernel since 2010.
>> And the documentation link on Sourceforge says it is abandoned there.
>>

To be honest, I thought it had already been removed from Linux long ago.

>> Leave the UAPI alone to keep userspace programs compiling.

This bit I don't agree with. When things break, it's better to know as
early as possible. That means it should break during compilation, not
when you try to run the program.

>
><https://lore.kernel.org/lkml/20220731190646.97039-1-stephen@networkplumbe
> r.org/>
>
> I don't know how well VMS works as an NFS client or server. When I was
> using VMS in the late 1990s, it wasn't great, but a lot of time has
> passed since then. If anyone wants Linux to carry on supporting DECnet,
> there will need to be credible volunteers to maintain it.
>

You aren't even allowed (quite rightly) to run unencrypted protocols
such as telnet on many networks any more.

The time of DECnet Phase IV has well and truly passed.

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Walking destinations on a map are further away than they appear.

On 8/2/22 09:04, Simon Clubley wrote:
> On 2022-08-02, John Dallman <jgd@cix.co.uk> wrote:
>> Whatever your opinion of Linux, it is significant as a server OS. The RFC
>> for its removal says:
>>
>>> Decnet is an obsolete network protocol that receives more attention
>>> from kernel janitors than users. It belongs in computer protocol
>>> history museum not in Linux kernel.
>>>
>
> Best description of DECnet Phase IV I have ever seen. :-)
>
>>> It has been Orphaned in kernel since 2010.
>>> And the documentation link on Sourceforge says it is abandoned there.
>>>
>
> To be honest, I thought it had already been removed from Linux long ago.
>
>>> Leave the UAPI alone to keep userspace programs compiling.
>
> This bit I don't agree with. When things break, it's better to know as
> early as possible. That means it should break during compilation, not
> when you try to run the program.
>
>>
>> <https://lore.kernel.org/lkml/20220731190646.97039-1-stephen@networkplumbe
>> r.org/>
>>
>> I don't know how well VMS works as an NFS client or server. When I was
>> using VMS in the late 1990s, it wasn't great, but a lot of time has
>> passed since then. If anyone wants Linux to carry on supporting DECnet,
>> there will need to be credible volunteers to maintain it.
>>
>
> You aren't even allowed (quite rightly) to run unencrypted protocols
> such as telnet on many networks any more.
>
> The time of DECnet Phase IV has well and truly passed.

Maybe, but for those of us who still enjoy true computer history
it is essential. I still use it. On the VAX, on the PDP-11 and,
yes, on Linux.

bill

chris wrote:

> Is it built into the kernel or is it, as it should be, a loadable module ?.
From a quick glance, it's the latter.
<https://github.com/torvalds/linux/blob/master/net/decnet/Kconfig>

> If it's a loadable module, then it should be maintainable as a separate
> entity
There's a big difference between maintainable and maintained. Out-of-tree
modules tend to bitrot quickly as the kernel makes no promise not to break
internal APIs

On 08/02/22 19:01, Andy Burns wrote:
> chris wrote:
>
>> Is it built into the kernel or is it, as it should be, a loadable
>> module ?.
> From a quick glance, it's the latter.
> <https://github.com/torvalds/linux/blob/master/net/decnet/Kconfig>
>
>> If it's a loadable module, then it should be maintainable as a separate
>> entity
> There's a big difference between maintainable and maintained.
> Out-of-tree modules tend to bitrot quickly as the kernel makes no
> promise not to break internal APIs

Very true, but it should still be possible to bring up to date with
enough effort. Apparently, development stopped for Linux in 2010, so
maybe past it. Decnet did run on FreeBSD, PCs, Solaris
and you could still run Pathworks if still available. Older protocols
can still be useful to recover data from old systems, or even to support
critical apps...

Chris

On 08/02/22 23:33, chris wrote:
> On 08/02/22 19:01, Andy Burns wrote:
>> chris wrote:
>>
>>> Is it built into the kernel or is it, as it should be, a loadable
>>> module ?.
>> From a quick glance, it's the latter.
>> <https://github.com/torvalds/linux/blob/master/net/decnet/Kconfig>
>>
>>> If it's a loadable module, then it should be maintainable as a separate
>>> entity
>> There's a big difference between maintainable and maintained.
>> Out-of-tree modules tend to bitrot quickly as the kernel makes no
>> promise not to break internal APIs
>
> Very true, but it should still be possible to bring up to date with
> enough effort. Apparently, development stopped for Linux in 2010, so
> maybe past it. Decnet did run on FreeBSD, PCs, Solaris
> and you could still run Pathworks if still available. Older protocols
> can still be useful to recover data from old systems, or even to support
> critical apps...
>
> Chris
>
>

Tru64 on Alpha had decnet support, but don't know if it required a
separate license...

Chris

On 2/08/2022 9:04 pm, Simon Clubley wrote:

> You aren't even allowed (quite rightly) to run unencrypted protocols
> such as telnet on many networks any more.
>
> The time of DECnet Phase IV has well and truly passed.
>
> Simon.
>

Can you not run it over TCP/IP and IPSec?

On 8/2/2022 6:33 PM, chris wrote:
> On 08/02/22 19:01, Andy Burns wrote:
>> chris wrote:
>>> Is it built into the kernel or is it, as it should be, a loadable
>>> module ?.
>>  From a quick glance, it's the latter.
>> <https://github.com/torvalds/linux/blob/master/net/decnet/Kconfig>
>>
>>> If it's a loadable module, then it should be maintainable as a separate
>>> entity
>> There's a big difference between maintainable and maintained.
>> Out-of-tree modules tend to bitrot quickly as the kernel makes no
>> promise not to break internal APIs
>
> Very true, but it should still be possible to bring up to date with
> enough effort. Apparently, development stopped for Linux in 2010, so
> maybe past it.

If enough effort is available then everything is possible.

Loadable module source code available that builds on all
recent Linux versions.

Binary modules available on all recent Linux versions on
common architectures.

Getting support for it back in the official kernel.

But there was an if a few lines above. And experience shows
that when it comes to actually doing some work, then it
becomes very difficult to find people.

Arne

=?UTF-8?Q?Arne_Vajh=c3=b8j?= <arne@vajhoej.dk> wrote:
>But there was an if a few lines above. And experience shows
>that when it comes to actually doing some work, then it
>becomes very difficult to find people.

DECNET is kind of cool on VMS systems, in that it allows file transfer
with transparent transfer of file attributes, something scp does not.

However, once you're talking between a VMS system and a Linux system,
that's no longer a big issue. Unless you have a system that cannot
possibly support tcp/ip, there is little reason to use decnet for that
today.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

On 8/2/2022 8:11 PM, Scott Dorsey wrote:
> =?UTF-8?Q?Arne_Vajh=c3=b8j?= <arne@vajhoej.dk> wrote:
>> But there was an if a few lines above. And experience shows
>> that when it comes to actually doing some work, then it
>> becomes very difficult to find people.
>
> DECNET is kind of cool on VMS systems, in that it allows file transfer
> with transparent transfer of file attributes, something scp does not.
>
> However, once you're talking between a VMS system and a Linux system,
> that's no longer a big issue. Unless you have a system that cannot
> possibly support tcp/ip, there is little reason to use decnet for that
> today.

DECnet is legacy and hobbyist today.

But 100 active hobbyist developers could create a lot.

But they don't exist.

Arne

On Tuesday, August 2, 2022 at 5:16:33 PM UTC-7, Arne Vajhøj wrote:

(snip)

> DECnet is legacy and hobbyist today.
>
> But 100 active hobbyist developers could create a lot.
>
> But they don't exist.
As long as old versions of Linux are still around, those should be fine
for any hobbyists that need old DECnet.

On 2022-08-03 01:08, Richard Maher wrote:
> On 2/08/2022 9:04 pm, Simon Clubley wrote:
>
>> You aren't even allowed (quite rightly) to run unencrypted protocols
>> such as telnet on many networks any more.
>>
>> The time of DECnet Phase IV has well and truly passed.
>>
>> Simon.
>>
>
> Can you not run it over TCP/IP and IPSec?

Yes you can.

Johnny

On 08/03/22 10:26, Johnny Billquist wrote:
> On 2022-08-03 01:08, Richard Maher wrote:
>> On 2/08/2022 9:04 pm, Simon Clubley wrote:
>>
>>> You aren't even allowed (quite rightly) to run unencrypted protocols
>>> such as telnet on many networks any more.
>>>
>>> The time of DECnet Phase IV has well and truly passed.
>>>
>>> Simon.
>>>
>>
>> Can you not run it over TCP/IP and IPSec?
>
> Yes you can.
>
> Johnny

So not a problem in security terms then. There's a long thread on
the classic computer mailing list discussing it here:

cctalk@classiccmp.org,

may have to adjust the link...

Chris

On 8/2/2022 9:04 AM, Simon Clubley wrote:
> On 2022-08-02, John Dallman <jgd@cix.co.uk> wrote:
>> Whatever your opinion of Linux, it is significant as a server OS. The RFC
>> for its removal says:
>>
>>> Decnet is an obsolete network protocol that receives more attention
>>> from kernel janitors than users. It belongs in computer protocol
>>> history museum not in Linux kernel.
>>>
>
> Best description of DECnet Phase IV I have ever seen. :-)
>
>>> It has been Orphaned in kernel since 2010.
>>> And the documentation link on Sourceforge says it is abandoned there.
>>>
>
> To be honest, I thought it had already been removed from Linux long ago.
>
>>> Leave the UAPI alone to keep userspace programs compiling.
>
> This bit I don't agree with. When things break, it's better to know as
> early as possible. That means it should break during compilation, not
> when you try to run the program.
>
>>
>> <https://lore.kernel.org/lkml/20220731190646.97039-1-stephen@networkplumbe
>> r.org/>
>>
>> I don't know how well VMS works as an NFS client or server. When I was
>> using VMS in the late 1990s, it wasn't great, but a lot of time has
>> passed since then. If anyone wants Linux to carry on supporting DECnet,
>> there will need to be credible volunteers to maintain it.
>>
>
> You aren't even allowed (quite rightly) to run unencrypted protocols
> such as telnet on many networks any more.
>
> The time of DECnet Phase IV has well and truly passed.
>
> Simon.
>

Nobody is forcing Simon and the rest of the do-gooders/despots to use DECnet,
but why are they so intent on denying others the capability of doing so?

--
David Froble Tel: 724-529-0450
Dave Froble Enterprises, Inc. E-Mail: davef@tsoft-inc.com
DFE Ultralights, Inc.
170 Grimplin Road
Vanderbilt, PA 15486