[COFF] Re: [TUHS] Re: Be there a "remote diff" utility?

[COFF] Re: [TUHS] Re: Be there a "remote diff" utility?

[John Cowan]

[-- Attachment #1: Type: text/plain, Size: 2411 bytes --]

The Newcastle Connection, aka Unix United, was an early experiment in
transparent networking: see <
https://web.archive.org/web/20160816184205/http://www.cs.ncl.ac.uk/research/pubs/articles/papers/399.pdf>
for a high-level description.  A name of the form "/../host/path"
represented a file or device on a remote host in a fully transparent way.
This was layered on V7 at the libc level, so that the kernel did not need
to be modified (though the shell did, since it was not libc-based at the
time).  MUNIX was an implementation of the same idea using System V as the
underlying system.

This appears to be a VHS vs. Betamax battle: NFS was not transparent, but
Sun had far more marketing clout.  However, the Manchester Connection
required a single uid space (as far as I can tell), which may also have
been a (perceived) institutional barrier.

On Thu, May 16, 2024 at 3:34 AM Ralph Corderoy <ralph@inputplus.co.uk>
wrote:

> Hi,
>
> I've set ‘mail-followup-to: coff@tuhs.org’.
>
> > > Every so often I want to compare files on remote machines, but all
> > > I can do is to fetch them first (usually into /tmp); I'd like to do
> > > something like:
> > >
> > >     rdiff host1:file1 host2:file2
> > >
> > > Breathes there such a beast?
>
> No, nor should there.  It would be slain less it beget rcmp, rcomm,
> rpaste, ...
>
> > > Think of it as an extension to the Unix philosophy of "Everything
> > > looks like a file"...
>
> Then make remote files look local as far as their access is concerned.
> Ideally at the system-call level.  Less ideal, at libc.a.
>
> > Maybe
> >
> >     diff -u <(ssh host1 cat file1) <(ssh host2 cat file2)
>
> This is annoyingly noisy if the remote SSH server has sshd_config(5)'s
> ‘Banner’ set which spews the contents of a file before authentication,
> e.g. the pointless
>
>     This computer system is the property of ...
>
>     Disconnect NOW if you have not been expressly authorised to use this
>     system.  Unauthorised use is a criminal offence under the Computer
>     Misuse Act 1990.
>
>     Communications on or through ...uk's computer systems may be
>     monitored or recorded to secure effective system operation and for
>     other lawful purposes.
>
> It appears on stderr so doesn't upset the diff but does clutter.
> And discarding stderr is too sloppy.
>
> --
> Cheers, Ralph.
>

[-- Attachment #2: Type: text/html, Size: 3361 bytes --]

[COFF] Re: [TUHS] Re: Be there a "remote diff" utility?

[Dan Cross]

On Thu, May 16, 2024 at 7:51 AM John Cowan <cowan@ccil.org> wrote:
> [snip]
> This appears to be a VHS vs. Betamax battle: NFS was not transparent, but Sun had far more marketing clout.  However, the Manchester Connection required a single uid space (as far as I can tell), which may also have been a (perceived) institutional barrier.

So did NFS, for that matter.

This is one of those areas where Unix appears creaky in comparison to
Plan 9. `ssh` is all about remote access to resources, whereas plan 9
was all about resource sharing: you'd set up a namespace with all of
the resources (exposed as files from wherever they ultimately came
from) you cared about, and then operate on those "locally"; the
resources were shared with you and access was transparent, via a
consistent, file-based interface. You want to `diff` two remote files?
Import the filesystems they're both on, mount those somewhere, and
`diff /n/host1/file /n/host2/file`.

I think the `sshfs`+FUSE model that Doug mentioned is about the
closest you're going to get these days.

        - Dan C.


        - Dan C.

[COFF] Re: [TUHS] Re: Be there a "remote diff" utility?

[Larry McVoy]

On Thu, May 16, 2024 at 09:45:38AM -0400, Dan Cross wrote:
> On Thu, May 16, 2024 at 7:51???AM John Cowan <cowan@ccil.org> wrote:
> > [snip]
> > This appears to be a VHS vs. Betamax battle: NFS was not transparent, but Sun had far more marketing clout.  However, the Manchester Connection required a single uid space (as far as I can tell), which may also have been a (perceived) institutional barrier.
> 
> So did NFS, for that matter.
> 
> This is one of those areas where Unix appears creaky in comparison to
> Plan 9. `ssh` is all about remote access to resources, whereas plan 9
> was all about resource sharing: you'd set up a namespace with all of
> the resources (exposed as files from wherever they ultimately came
> from) you cared about, and then operate on those "locally"; the
> resources were shared with you and access was transparent, via a
> consistent, file-based interface. You want to `diff` two remote files?
> Import the filesystems they're both on, mount those somewhere, and
> `diff /n/host1/file /n/host2/file`.

If you are all trusting, behind a firewall, like the Sun campus was:

diff /net/host1/file /net/host2/file

Seems pretty darn similar and you don't set up a namespace other than
saying what you want to share in /etc/exports.

Seems far from creaky to me, no root access required (other than setting
up /etc/exports), any user can access any exported file.  I was at Sun
and it worked great.  Granted, you needed a working NFS implementation
and when I left Sun I found that Sun was really the only place that got
that right in the 1990's.  Linux has caught up, don't know about the
rest.

--lm

[COFF] Re: [TUHS] Re: Be there a "remote diff" utility?

[John Cowan]

[-- Attachment #1: Type: text/plain, Size: 1345 bytes --]

NFS v4 provides idmapping between client uids/gids and their server
equivalents using a config file.


On Thu, May 16, 2024 at 9:46 AM Dan Cross <crossd@gmail.com> wrote:

> On Thu, May 16, 2024 at 7:51 AM John Cowan <cowan@ccil.org> wrote:
> > [snip]
> > This appears to be a VHS vs. Betamax battle: NFS was not transparent,
> but Sun had far more marketing clout.  However, the Manchester Connection
> required a single uid space (as far as I can tell), which may also have
> been a (perceived) institutional barrier.
>
> So did NFS, for that matter.
>
> This is one of those areas where Unix appears creaky in comparison to
> Plan 9. `ssh` is all about remote access to resources, whereas plan 9
> was all about resource sharing: you'd set up a namespace with all of
> the resources (exposed as files from wherever they ultimately came
> from) you cared about, and then operate on those "locally"; the
> resources were shared with you and access was transparent, via a
> consistent, file-based interface. You want to `diff` two remote files?
> Import the filesystems they're both on, mount those somewhere, and
> `diff /n/host1/file /n/host2/file`.
>
> I think the `sshfs`+FUSE model that Doug mentioned is about the
> closest you're going to get these days.
>
>         - Dan C.
>
>
>         - Dan C.
>

[-- Attachment #2: Type: text/html, Size: 1947 bytes --]

[COFF] Re: [TUHS] Re: Be there a "remote diff" utility?

[Dan Cross]

On Thu, May 16, 2024 at 10:44 AM John Cowan <cowan@ccil.org> wrote:
> NFS v4 provides idmapping between client uids/gids and their server equivalents using a config file.

NFS v4 came much later, of course. But even earlier NFS
implementations provided something like this, and UID 0 has been
mapped to "nobody" for many decades. But that's not terribly relevant;
the point is that, by default, the NFS protocol relies on UIDs having
meaning, as opposed to properly authenticated principles independent
of the underlying implementation of "identity", as in e.g. AFS. To be
fair, it's my understanding that NFSv4 _does_ improve on the situation
here.

        - Dan C.


> On Thu, May 16, 2024 at 9:46 AM Dan Cross <crossd@gmail.com> wrote:
>>
>> On Thu, May 16, 2024 at 7:51 AM John Cowan <cowan@ccil.org> wrote:
>> > [snip]
>> > This appears to be a VHS vs. Betamax battle: NFS was not transparent, but Sun had far more marketing clout.  However, the Manchester Connection required a single uid space (as far as I can tell), which may also have been a (perceived) institutional barrier.
>>
>> So did NFS, for that matter.
>>
>> This is one of those areas where Unix appears creaky in comparison to
>> Plan 9. `ssh` is all about remote access to resources, whereas plan 9
>> was all about resource sharing: you'd set up a namespace with all of
>> the resources (exposed as files from wherever they ultimately came
>> from) you cared about, and then operate on those "locally"; the
>> resources were shared with you and access was transparent, via a
>> consistent, file-based interface. You want to `diff` two remote files?
>> Import the filesystems they're both on, mount those somewhere, and
>> `diff /n/host1/file /n/host2/file`.
>>
>> I think the `sshfs`+FUSE model that Doug mentioned is about the
>> closest you're going to get these days.
>>
>>         - Dan C.
>>
>>
>>         - Dan C.