Date: Mon, 10 Jul 2000 10:06:08 -0700 From: Doug Barton <DougB@gorean.org> To: Salvo Bartolotta <bartequi@inwind.it> Cc: freebsd-questions@FreeBSD.ORG, djkanter@northwestern.edu Subject: Re: Softupdates question Message-ID: <396A0280.98065D7E@gorean.org> References: <20000709005612.A89313@localhost.localdomain> <20000709.23515500@bartequi.ottodomain.org> <3969172D.D3A30104@gorean.org> <20000710.13040500@bartequi.ottodomain.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Salvo Bartolotta wrote: > 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) ? Here is about as clear a post as you're going to get: http://www.FreeBSD.org/cgi/getmsg.cgi?fetch=106939+109577+/usr/local/www/db/text/1998/freebsd-current/19980927.freebsd-current Try looking at the situation logically. The noatimes option causes access time not to be written out when a file is accessed. This is part of the metadata that softupdates optimizes the writing of. Therefore at best, eliminating that one small piece of metadata is not likely to have any beneficial effect. On the other side of the coin, using softupdates & noatime together used to cause kernel panics. It doesn't cause them anymore, but the code to work around that bug might be affecting your performance negatively. I'm not sure on that, but my whole point is that it is so unlikely that the noatime option is gaining you anything that it's not worth the risk. Doug -- "Live free or die" - State motto of my ancestral homeland, New Hampshire Do YOU Yahoo!? To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?396A0280.98065D7E>