Re: Problem with Tramp 2.5.2 (with emacs 28.1) when talking to Solaris

[Michael Albinus]

Simon Matthews <spm@flirble.org> writes:

Hi Simon,

> Just to confirm that the latest git version seems to work without problems.

Thanks for confirmation. This change will appear with the next Emacs
28.2 release. It will also be contained in the next GNU ELPA Tramp
version, 2.5.2.4, scheduled for the next days.

> However checking uname for SunOS does seem quite fragile (eg for Solaris
> derivatives like Illumos). An option to disable adding -w or even changing
> the test to check that it gets some expected results would be preferred.

That would be an alternative, yes. But experience shows, that many users
don't find the proper option; that's why I prefer that Tramp detects
such problems itself. Using uname is good and serves most
constellations.

If you could send me the output of "uname -sr" of such a Solaris
derivate, I'm open to add this to `tramp-sunos-unames'. Reading
<https://gist.github.com/natefoo/7af6f3d47bb008669467> it looks like
"SunOS 5.11" is used. OTOH,
<https://github.com/gdamore/illumos-core/blob/core/README.uname> tells
something different.

Since there are several cases SunOS differs in its implementation, it
might be the better case for users to modify this constant anyway. Look
for places `tramp-sunos-unames' is used in tramp-sh.el. Maybe we shall
document this approach?

> Simon

Best regards, Michael.