From: Ian Kent <raven@themaw.net>
Subject: Re: [NFS] Attempting to mount a large number of directories
Date: Tue, 08 Jan 2008 12:07:43 +0900
Message-ID: <1199761663.3155.17.camel@raven.themaw.net>
References: <200712241200.08410.lists@fabis-site.net>
	 <1199695504.3156.69.camel@raven.themaw.net>
	 <200801072228.58507.lists@fabis-site.net>
Mime-Version: 1.0
Content-Type: text/plain
Cc: linux-nfs@vger.kernel.org
To: Fabian Steiner <lists-A1xC5SIKT5unwkN7LPkuEg@public.gmane.org>
In-Reply-To: <200801072228.58507.lists-A1xC5SIKT5unwkN7LPkuEg@public.gmane.org>
Sender: linux-nfs-owner@vger.kernel.org


On Mon, 2008-01-07 at 22:28 +0100, Fabian Steiner wrote:
> Ian Kent wrote:
> >> [...]
> > I see no-one else wanted to buy into this.
> > I'm not really surprised.
> 
> Thanks a lot for your explainations! Now it is much easier to understand how 
> things work on the low-level area.
> 
> > Judging by your attempt to use the insecure option I assume you actually
> > know that you're running out of port space on the client. The server, of
> > course, consumes only a few ports, like ports for mountd, portmap (or
> > rpcbind) and NFS and because it's a server and it reuses the same port
> > since the different port used on the client makes the tupple
> > clientaddr,clientport,serveraddr,serverport unique. Point is that port
> > space is consumed on the client not the server.
> >
> > So running out of port space is a client problem and the reason it
> > happens is that autofs and mount probe to see if the server is up and
> > what version of NFS is available etc. before doing the mount. This can
> > result in as many a 9 ports consumed for each mount and the ports that
> > aren't continuing to be used can't be reused before a timeout has
> > passed. I think it's about 60 seconds, and is required for the TCP
> > implementation to function properly. So you run out of available ports
> > fairly quickly if you issue many mount requests at once.
> >
> > Quite a bit of work has been done to reduce the port usage over time and
> > I think it has been included in recent nfs-utils releases so perhaps a
> > later nfs-utils will help.
> 
> On the affected machines nfs-utils-1.0.7 has been installed so far. Picking up 
> your hint we forced an upgrade from Ubuntu Dapper to Ubuntu Gutsy with 
> version 1.1.1~git-20070709 being installed and obviously this solved the 
> majority number of problems. Now approx. 400 directories can be mounted. Of 
> course, the entire issue requires further investigation; we are quite content 
> about the current situation, though.

Great, you may also find that specifying the protocol as a mount option,
proto=tcp for example, will reduce the overhead further.

Ian