Subject: Re: install/31297
To: Peter <firstname.lastname@example.org>
From: Rui Paulo <rpaulo@NetBSD.org>
Date: 09/12/2005 02:14:55
Content-Type: text/plain; charset=us-ascii
On 2005.09.12 04:09:48 +0300, Peter wrote:
| On Mon, 12 Sep 2005 email@example.com wrote:
| >Synopsis: mounting volumes with -o async speeds up file write=20
| >operations 500% (5 times)
| >State-Changed-From-To: open->feedback
| >State-Changed-By: firstname.lastname@example.org
| >State-Changed-When: Mon, 12 Sep 2005 00:52:06 +0000
| >Mouting file systems with the async is evil and very dangerous.
| >>From the mount(8) man page:
| > async All I/O to the file system should be done asyn-
| > chronously. In the event of a crash, it is
| > impossible for the system to verify the integrit=
| > of
| > data on a file system mounted with this option. =
| > You
| > should only use this option if you have an appli=
| > tion-specific data recovery mechanism, or are=20
| > willing
| > to recreate the file system from scratch.
| >I think this PR should be closed.
| Mounting filesystems without -o async while plugged into an UPS is=20
| not so dangerous.
Most of the times, systems that are to be taken to a datacenter after
instalation and configuration were not installed using an UPS, so this
probahly not a very common case.
| Imho not using -o async is dangerous to the installer's health. It=20
| increases the coffee consumption 5 times over the -o async option, and=20
| the blood pressure upon seeing that the data throughput on UDMA2 disks=20
| is only about 150kB/sec without -o async in the conditions shown (many=20
| small files unpacked from tar).
You can make sysinst more silent about that. :-)
| Also during install the worst thing that can happen is that one has to=20
| reinstall. Since installs wipe nearly everything (esp. if the disklabel=
| was edited).
You are not considering the updates made through sysinst.
-- Rui Paulo
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (NetBSD)
-----END PGP SIGNATURE-----