Date: Wed, 19 Apr 2006 08:23:29 -0700 From: Alfred Perlstein <alfred@freebsd.org> To: Mohan Srinivasan <mohan_srinivasan@yahoo.com> Cc: src-committers@FreeBSD.org, cvs-src@FreeBSD.org, cvs-all@FreeBSD.org, ps@freebsd.org, Xin LI <delphij@FreeBSD.org>, mohans@freebsd.orgc Subject: Re: cvs commit: src/sys/nfsclient nfs_bio.c nfs_vnops.c Message-ID: <20060419152329.GO35896@elvis.mu.org> In-Reply-To: <20060418165348.55185.qmail@web30804.mail.mud.yahoo.com> References: <20060418133832.GF35896@elvis.mu.org> <20060418165348.55185.qmail@web30804.mail.mud.yahoo.com>
next in thread | previous in thread | raw e-mail | index | archive | help
* Mohan Srinivasan <mohan_srinivasan@yahoo.com> [060418 09:56] wrote: > Alfred, > > If the operation times out because the server rebooted (or because > of a network glitch), rather than losing data, the client can > potentially have the data written out successfully later. > > Other NFS client implementations (at least the Solaris client) > re-dirty pages on soft mount timeouts. Ok, I just wanted to be sure that we weren't turning soft mounts into hard mounts. The only problem though is that you have a weird failure mode now where the OS reports failure however it may later succeed, I think this is OK, esp if Solaris does it, but it is interesting. Seems OK, Thanks for clarifying! -Alfred
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060419152329.GO35896>