Date: Thu, 19 Feb 2004 11:56:47 +0100 From: "Poul-Henning Kamp" <phk@phk.freebsd.dk> To: Andre Guibert de Bruet <andy@siliconlandmark.com> Cc: current@freebsd.org Subject: Re: Review/Test: Pseudo-device unit number management patch Message-ID: <26055.1077188207@critter.freebsd.dk> In-Reply-To: Your message of "Thu, 19 Feb 2004 03:09:31 EST." <20040219025604.D76134@alpha.siliconlandmark.com>
next in thread | previous in thread | raw e-mail | index | archive | help
In message <20040219025604.D76134@alpha.siliconlandmark.com>, Andre Guibert de Bruet writes: >The likes of ATA disks include serial numbers in words 10-19. These could >be used to positively identify a drive when its availability has come >into question. But how can you tell if the disk has been mounted and modified in another computer before it came back ? There are N ways you can add bandaids and heuristics, but you can not truly solve the problem, unless you constantly keep a very strong hash of the entire disk contents in RAM, and check it again if it comes back which is if not practically impossible, then at least prohibitively expensive for the added benefit. Our current operating model is both cheap performance wise, and simple to understand for users: Do not remove a storage device you are still using. I do not think you should expect this to change significantly in the future. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk@FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?26055.1077188207>