Bug#718416: closed by Dmitry Bogatov <KAction at debian.org> ()

[Thorsten Glaser]

On Fri, 16 Nov 2018, Adam Borowski wrote:

> Starting X will also kill any scrollback, unless you recompile your kernel

Sure, but when I’m on a server VM or use startx, this is moot.


On Fri, 16 Nov 2018, Ian Jackson wrote:

> I agree that messing with existing installations is probably wrong.
> (Or, at least, too complex to get right.)  Thorsten, do you agree with
> that ?

Somewhat. But yes, existing installations probably work already,
and my main use case for this is to debug a new installation.

On the other hand, this occasionally comes up on an existing
installation, checking for a regression after a system upgrade,
or even on one I did not initially set up. I’d be much in favour
if it could also be fixed for existing installations, thus, but
I’m not ignorant of the trouble this could invite and won’t insist
too much.

> But I don't see anyone having made a coherent argument about what the
> behaviour of new installations should be.

Thanks.

> I found Dmitry's argument
> about "hurting those, who are accustomized to `clear` behaviour"
> relevant to existing installations but much less so for new ones.

I don’t find it relevant to the “modern desktop” users — those all
boot from GRUB in graphical mode via plymouth straight to X11 or
worse anyway.

For this reason I believe that not quieting the kernel messages
and not clearing the screen, for a console login system, are the
right thing to do even *more* than some time ago.

> I think the ideal default behaviour would be for the *first* getty to
> spawn on tty1 not to clear the screen.

Yes.

> We probably can't manage that.

Why?

> But failing that my current opinion is that we should change the
> default inittab to pass --noclear for tty1.

Indeed. And I think that ought to have been done in 2013 already,
but doing it now is better than later or never.

This is orthogonal to other fixes for this problem anyway.

bye,
//mirabilos
PS: Sorry for being a bit late in catching up to some eMails,
    a lot of them has piled up, but I’m buried in $dayjob and
    other work with deadlines.
-- 
«MyISAM tables -will- get corrupted eventually. This is a fact of life. »
“mysql is about as much database as ms access” – “MSSQL at least descends
from a database” “it's a rebranded SyBase” “MySQL however was born from a
flatfile and went downhill from there” – “at least jetDB doesn’t claim to
be a database”	‣‣‣ Please, http://deb.li/mysql and MariaDB, finally die!