Date: Mon, 9 Sep 1996 13:58:03 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: karl@Mcs.Net (Karl Denninger) Cc: jkh@time.cdrom.com, henrich@crh.cl.msu.edu, freebsd-current@freebsd.org Subject: Re: Grrr. NFS to a Sun (Slowaris 5.5.1) Message-ID: <199609092058.NAA01665@phaeton.artisoft.com> In-Reply-To: <199609091917.OAA16267@Jupiter.mcs.net> from "Karl Denninger" at Sep 9, 96 02:17:44 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> > You'd be right in that impression - NFS clientry is now one of the > > easiest way to crash yourself in -current. > > Uh, why is it that my news server, which NFS serves the spool to some 30 > clients, doesn't blow up? :-) In simplest terms, "because you aren't using the broken code". The client code is FS code. The Server code is a kernel process for an rpc server using a system call that never returns to user space (the oldest and crudest way to get a kernel thread). In general, there are a lot of NFS client *and* server problems, most of which derive from the framework. For instance, the NFS server cookie stuff for VOP_READDIR() if an exported FS is pure garbage. A theoretical soloution hashed out between Doug Rabson and myself about 6 months ago is waiting for an implementor in the -current list archives. There are also issues of principle, like who should be handling the export options (right now, it's per FS list mangement, which explains why you can't NFS serve some FS types). Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199609092058.NAA01665>