Date: Sat, 5 Nov 2011 20:03:44 -0700 From: Xin LI <delphij@gmail.com> To: Rick Macklem <rmacklem@uoguelph.ca> Cc: freebsd-fs@freebsd.org, Josh Paetzel <jpaetzel@freebsd.org>, zkirsch@freebsd.org Subject: Re: [RFC] Should vfs.nfsrv.async be implemented for new NFS server? Message-ID: <CAGMYy3tT3vmqGq6ju7GVpA1WKZ=VdFSbvZH1FAezaxtNUMDnvQ@mail.gmail.com> In-Reply-To: <1558351773.1229453.1320542285788.JavaMail.root@erie.cs.uoguelph.ca> References: <1558351773.1229453.1320542285788.JavaMail.root@erie.cs.uoguelph.ca>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi, Rick, On Sat, Nov 5, 2011 at 6:18 PM, Rick Macklem <rmacklem@uoguelph.ca> wrote: > Hi, > > Josh Paetzel pointed out that vfs.nfsrv.async doesn't exist > for the new NFS server. > > I don't think I had spotted this before, but when I looked I > saw that, when vfs.nfsrv.async is set non-zero in the old server, > it returns FILESYNC (which means the write has been committed to > non-volatile storage) even when it hasn't actually done that. > > This can improve performance, but has some negative implications: > - If the server crashes before the write is committed to > =C2=A0non-volatile storage, the file modification will be lost. > =C2=A0(When a server replies UNSTABLE to a write, the client holds > =C2=A0 onto the data in its cache and does the write again if the > =C2=A0 server crashes/reboots before the client does a Commit RPC > =C2=A0 for the file. However, a reply of FILESYNC tells the client > =C2=A0 it can forget about the write, because it is done.) > - Because of the above, replying FILESYNC when the data is not > =C2=A0yet committed to non-volatile (also referred to as stable) > =C2=A0storage, this is a violation of RFC1813. > > I wouldn't want this to be the default, but am willing to > patch head based on the "backwards compatibility" argument. > My concern with these types of patches is that some people > will enable them without realizing the risk of data loss > that they introduce. > > So, how do others feel with respect to whether or not this > patch should be committed to head? I think the default of old NFS server was async=3D0? In general I'd prefer seeing this as an option but disabled by default, so administrators can override the option. Having async=3D1 by default doesn't seem to be a good idea in my opinion. Another thought is the async flag should be a per-mountpoint flag rather than a global flag, but that might be over-complicating things so just my $0.02. Cheers, --=20 Xin LI <delphij@delphij.net> https://www.delphij.net/ FreeBSD - The Power to Serve! Live free or die
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAGMYy3tT3vmqGq6ju7GVpA1WKZ=VdFSbvZH1FAezaxtNUMDnvQ>