Subject: Re: FFS reliability problems
To: None <tech-kern@netbsd.org>
From: Greg A. Woods <woods@weird.com>
List: tech-kern
Date: 06/12/2002 16:38:38
[ On Wednesday, June 12, 2002 at 20:51:33 (+0700), Robert Elz wrote: ]
> Subject: Re: FFS reliability problems
>
> Date: Mon, 10 Jun 2002 12:14:06 -0400 (EDT)
> From: der Mouse <mouse@Rodents.Montreal.QC.CA>
> Message-ID: <200206101614.MAA13292@Sparkle.Rodents.Montreal.QC.CA>
>
> | fsck's raison d'etre is to bring the filesystem into an internally
> | self-consistent state, a state that it could have been in upon a clean
> | shutdown. It is impossible to do this in the circumstances described
> | without either losing or recovering the data in question.
>
> The data was lost when the application unlinked it. Either the kernel
> (when the app exits/dies) or fsck (if the system crashed) are just
> completing the job.
Depending on the application the data often wasn't even written when the
unlink() was done....
--
Greg A. Woods
+1 416 218-0098; <gwoods@acm.org>; <g.a.woods@ieee.org>; <woods@robohack.ca>
Planix, Inc. <woods@planix.com>; VE3TCP; Secrets of the Weird <woods@weird.com>