Date: Sun, 27 Sep 1998 14:43:02 +0800 From: Peter Wemm <peter@netplex.com.au> To: Ollivier Robert <roberto@keltia.freenix.fr> Cc: freebsd-current@FreeBSD.ORG Subject: Re: Softupdates panics Message-ID: <199809270643.OAA11696@spinner.netplex.com.au> In-Reply-To: Your message of "Sat, 26 Sep 1998 22:03:33 %2B0200." <19980926220333.A20259@keltia.freenix.fr>
next in thread | previous in thread | raw e-mail | index | archive | help
Ollivier Robert wrote: > According to Don Lewis: > > I hadn't heard this. I don't know why that would be true. > > I remember people getting panics when running noatime with SU. I've got a > few myself (although they were more unstable than now). I thought it was async that was causing the problems. One of the main problems was that softupdates was doing a 'write everything' FSYNC vnode op and expecting all modified data to be written out. The async mount option "broke" the semantics of fsync() so that it didn't really wait until everything was on disk, so softupdate's sanity checks caught it. I think I saw somebody (Bruce?) change the semantics of VOP_FSYNC() back again. Cheers, -Peter To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199809270643.OAA11696>