From: Jeffrey Layton Subject: Re: replicated mounts patch Date: Wed, 20 Apr 2005 11:54:04 -0400 Message-ID: <1114012444.9683.28.camel@barsoom.rdu.redhat.com> References: <1113875877.4581.15.camel@tleilax.poochiereds.net> <1113912312.1274.26.camel@barsoom.rdu.redhat.com> <1113933871.1274.56.camel@barsoom.rdu.redhat.com> <1113959940.5879.5.camel@tleilax.poochiereds.net> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: autofs-bounces@linux.kernel.org Errors-To: autofs-bounces@linux.kernel.org Content-Type: text/plain; charset="us-ascii" To: raven@themaw.net Cc: autofs@linux.kernel.org On Wed, 2005-04-20 at 23:24 +0800, raven@themaw.net wrote: > I've tested this and it certainly seems to work. > > But another thing came to mind. > > We've had complaints about ports and protocols being blocked for the > rpc_ping function. > > Any thoughts on how to work around that? My patch should address this somewhat -- it looks for the 'tcp' and 'nfsvers=' mount options and does the RPC ping according to those. So if the ping doesn't work due to blocked ports, then the mount won't work either. > The other thing that plagues us is that when we open TCP sockets they hang > around which causes port exaustion. > Thoughts on that as well? On this part, there's not much we can do. The kernel leaves sockets in FIN_WAIT for a little while in case any stray packets come in after the connection has been closed. I don't think much (if anything) can be done from userspace to affect this, other than to close the socket as soon as possible. -- Jeff