Subject: Re: PR 36963
To: None <tech-kern@NetBSD.org>
From: Jan Danielsson <jan.m.danielsson@gmail.com>
List: tech-kern
Date: 09/23/2007 00:06:12
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

Daniel Carosone wrote:
> On Thu, Sep 20, 2007 at 02:22:05PM +0200, Jan Danielsson wrote:
>> It actually boots a memory disk with a file system image from the USB
>> memory stick, so the real root is RAM at all times, once the kernel has
>> been loaded.
> 
> FWIW, I have the same setup (so I can remove the usb stick after
> boot).  I haven't used it extensively (laziness; I never quite got
> around to repartitioning and removing the small boot partition on the
> HD that this was supposed to eliminate) but I never had or noticed
> issues like this.

   Hmm.. I'm trying to decide if that is good or bad news. It's very
noticeable for me; postfix will always fail on boot due to the
permission problems.

> I was using X and not really logging in on console, other than in
> single user, but I can try again sometime soon now there are more
> specific ideas about what to look for and how to trigger it.

   I don't really have to work to trigger it -- in fact, booting is
enough. Once I have "deactivated" the problem, by logging out, I just
have to do something like update pkgsrc, and the problem pops up again.

> Oh, one other possible clue/difference: In order to work around issues
> with getty and Xconsole fighting some time ago, my /etc/ttys looks
> like this:
> 
> console "/usr/libexec/getty Pc"         vt100   off secure
> ttyE0   "/usr/libexec/getty Pc"         vt220   on secure
> 
> and so on for more ttyE?'s.  So maybe there are fewer things with
> console itself open.  Does doing something similar on your machine
> alter the behaviour at all?

   Mine looks like that too. :-(

- --
Kind regards,
Jan Danielsson

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (NetBSD)

iD8DBQFG9ZHUuPlHKFfKXTYRCguMAKCIilFBBNYaYXQRgTcKJ9q6Oe3oIACfWmYq
N/QjC6P1jfzHUNcOKBjKbno=
=3ZF/
-----END PGP SIGNATURE-----