Subject: Re: raidframe parity always dirty after reboot
To: Matthias Buelow <mkb@mukappabeta.de>
From: Greg A. Woods <woods@weird.com>
List: netbsd-users
Date: 04/28/2003 12:44:38
[ On Monday, April 28, 2003 at 18:20:30 (+0200), Matthias Buelow wrote: ]
> Subject: Re: raidframe parity always dirty after reboot
>
> But swap is a kernel entity;
No, "swap" is not really a kernel entity, it's a form of intermediate
external storage used by the kernel (just like any filesystem).
> surely your argument could be extended to
> the buffer cache aswell;
Nope. There's no way to turn on, turn off, or otherwise tune or
configure the buffer cache from a userlevel process (potential future
sysctl tweaks aside). Not only that but the buffer cache isn't a
stateful external thing either -- it's just part of volatile memory.
Strictly speaking my argument does extend to mount points though, and
indeed in traditional Unix systems the "umount" had to be done
explicitly at shutdown. I would not be unhappy if NetBSD required this
as well. I still tend to not trust the automatic kernel unmounts, even
though I've no technical reason for not trusting it, and so I often do
"umount -a" myself when shutting down from single user mode. I've even
been known to "umount -u -o ro /" too.
--
Greg A. Woods
+1 416 218-0098; <g.a.woods@ieee.org>; <woods@robohack.ca>
Planix, Inc. <woods@planix.com>; VE3TCP; Secrets of the Weird <woods@weird.com>