From owner-freebsd-questions Mon Jul 10 5: 3:24 2000 Delivered-To: freebsd-questions@freebsd.org Received: from relay1.inwind.it (relay1.inwind.it [212.141.53.67]) by hub.freebsd.org (Postfix) with ESMTP id DE5B737B749 for ; Mon, 10 Jul 2000 05:03:20 -0700 (PDT) (envelope-from bartequi@inwind.it) Received: from bartequi.ottodomain.org (212.141.78.153) by relay1.inwind.it; 10 Jul 2000 14:03:16 +0200 From: Salvo Bartolotta Date: Mon, 10 Jul 2000 13:04:05 GMT Message-ID: <20000710.13040500@bartequi.ottodomain.org> Subject: Re: Softupdates question To: Doug Barton Cc: freebsd-questions@FreeBSD.ORG, djkanter@northwestern.edu References: <20000709005612.A89313@localhost.localdomain> <20000709.23515500@bartequi.ottodomain.org> <3969172D.D3A30104@gorean.org> X-Mailer: SuperCalifragilis X-Priority: 3 (Normal) MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG >>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<< On 7/10/00, 1:22:05 AM, Doug Barton wrote regarding Re: Softupdates question: > Salvo Bartolotta wrote: > > If performance is your main concern, you may wish to enable > > softupdates and the "noatime" option for all filesystems except for > > "/". This last filesystem ("/") should (normally) be used only with > > the "noatime" option. > I've read many posts from some really smart people that said not= only does > noatime not help you in combination with softupdates, but it can actually > hurt you. Also, you shouldn't be doing a lot of i/o on your root filesystem > anyway, so I wouldn't advise using noatime on that one either. > Good luck, > Doug Dear Doug Barton, thank you very much for writing. My understanding was that the noatime option reduced writes, as per TFM (mount(8)):
noatime Do not update the file access time when reading from a file. This option is useful on filesystems where there are large numbers of files and performance is more criti- cal than updating the file access time (which is rarely ever important). This option is currently only supported on local filesystems.
Essentially, I thought that avoiding these writes in conjunction with softupdates (smart metadata management) would not do harm. Also, a number of posts had showed that a few people were actually using softupdates *and* noatime. However, I searched -questions, -fs, -hackers, -stable, and -arch (!), once again for the string "softupdate* and noatime". Alas, I have not found any clues: rather, it seems that most authoritative posts (even explicitly) suggest using softupdates **and** noatime in order to improve performance. Ok, I am a little confused now :-) Would you be so kind as to shed more light on the whole matter (also providing appropriate pointers/material/evidence) ? I have been using softupdates (+ noatime) for a few months, so I am particularly intested in correctly understanding the potential issues and/or dangers connected with this filesystem policy and noatime. Many thanks in advance and best regards, Salvo To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message